WO2012016531A1 - 一种接入网设备的入网方法及系统 - Google Patents

一种接入网设备的入网方法及系统 Download PDF

Info

Publication number
WO2012016531A1
WO2012016531A1 PCT/CN2011/077991 CN2011077991W WO2012016531A1 WO 2012016531 A1 WO2012016531 A1 WO 2012016531A1 CN 2011077991 W CN2011077991 W CN 2011077991W WO 2012016531 A1 WO2012016531 A1 WO 2012016531A1
Authority
WO
WIPO (PCT)
Prior art keywords
address
access
packet
protocol packet
network
Prior art date
Application number
PCT/CN2011/077991
Other languages
English (en)
French (fr)
Inventor
陆宏成
杨春晖
Original Assignee
北京乾唐视联网络科技有限公司
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 北京乾唐视联网络科技有限公司 filed Critical 北京乾唐视联网络科技有限公司
Priority to US13/814,728 priority Critical patent/US9154404B2/en
Publication of WO2012016531A1 publication Critical patent/WO2012016531A1/zh

Links

Classifications

    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04LTRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L45/00Routing or path finding of packets in data switching networks
    • H04L45/14Routing performance; Theoretical aspects
    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04LTRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L12/00Data switching networks
    • H04L12/28Data switching networks characterised by path configuration, e.g. LAN [Local Area Networks] or WAN [Wide Area Networks]
    • H04L12/2854Wide area networks, e.g. public data networks
    • H04L12/2856Access arrangements, e.g. Internet access
    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04LTRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L12/00Data switching networks
    • H04L12/28Data switching networks characterised by path configuration, e.g. LAN [Local Area Networks] or WAN [Wide Area Networks]
    • H04L12/2854Wide area networks, e.g. public data networks
    • H04L12/2856Access arrangements, e.g. Internet access
    • H04L12/2869Operational details of access network equipments

Definitions

  • the present invention relates to the technical field of a novel network, and in particular, to a network access method for an access network device, a network access system for an access network device, a node server, and an access switch.
  • Background technique
  • New networks including the Internet, enable the exchange of information and other information resources between different individuals and institutions.
  • Networks typically include technologies such as path, transport, signaling, and network management. These techniques have been widely found in various literatures. The general introduction to this is:
  • Path technologies connecting terminals to a wide-area transport network have evolved from 14.4, 28.8 and 56K modems to include ISDN, Tl, cable modem, DSL, Ethernet and wireless Connected technology.
  • Synchronous Optical Network SONET
  • DWDM Closed Wavelength Division Multiplexing
  • ATM Asynchronous Transfer Mode
  • RPR Resilient Packet Ring
  • IP Internet Protocol
  • IP-based network such as the Internet
  • voice such as telephony
  • video and data networks will be inevitable.
  • IP-based trains that integrate all types of networks have already left the station, and some passengers are extremely enthusiastic about the trip. Others are reluctantly dragged, crying, calling, kicking and enumerating IP. Defects. But no matter what its flaws, IP has been adopted as an industry standard, and no other technology except it has such great potential and room for development. , (taken from
  • TCP/IP Transmission Control Protocol/Internet Protocol
  • ARPANET ARPANET
  • Application layer - Application layer is a general term for all applications that users are targeting.
  • the TCP/IP protocol family has multiple protocols at this level to support different applications, and many of the familiar Internet-based applications are inseparable from these protocols.
  • WWW World Wide Web
  • FTP File Transfer Protocol
  • SMTP file transfer protocol
  • DNS domain name resolution
  • Telnet protocol remote login, etc.
  • the function of the transport layer one layer is mainly to provide communication between applications.
  • the TCP/IP protocol family has TCP and UDP protocols at this layer.
  • Network layer 1 is a very important layer in the TCP/IP protocol family. It mainly defines the IP address format, so that data of different application types can be smoothly transmitted on the Internet.
  • the IP protocol is a network layer protocol.
  • Network interface layer This is the lowest layer of TCP / IP software, responsible for receiving IP packets And send it through the network, or receive physical frames from the network, extract IP datagrams, and hand them over to the IP layer.
  • IP IP-to-Network Interface
  • the main reason for their incompatibility is because of the basic unit of data they transmit (technically called "frames").
  • the format is different.
  • the IP protocol is actually a set of protocol software consisting of software programs that convert all kinds of different "frames" into an "IP packet” format. This conversion is one of the most important features of the Internet, making all kinds of computers Interoperability can be achieved on the Internet, which is characterized by "openness.”
  • a packet is also a form of packet switching, which means that the transmitted data is segmented into “packets” and then transmitted. However, it belongs to the “connectionless type", and each "package” (grouping) that is marked is transmitted as an "independent message”, so it is called “data packet”. In this way, it is not necessary to connect a circuit before starting communication, and each data packet does not have to be transmitted through the same path, so it is called “connectionless type”. This feature is very important, in the case of text messaging, it greatly enhances the robustness and security of the network.
  • Each packet has two parts: a header and a message.
  • the header has a necessary content such as a destination address, so that each packet can accurately reach the destination without going through the same path. Recombination at the destination is restored to the originally sent data. This requires IP to have the function of packet packing and assembly.
  • the data packet can also change the length of the data packet according to the packet size specified by the network.
  • the maximum length of the IP data packet can reach 65535 bytes.
  • QoS Quality of Service
  • the network transmission quality is specifically represented by packet loss and bit error.
  • Computer files are not sensitive to errors in transmission. Even if most of the data packets are lost during transmission, as long as there is a TCP retransmission mechanism, the computer will still consider the network available. However, if the packet loss and bit error rate are higher than one thousandth, the video quality will be degraded for synchronous video.
  • the empirical data tells us that high-quality video communication even requires less than one in 100,000 lost packets and errors.
  • the test data of the current network environment shows that most of the packet loss occurs inside the router, and the error generated in the fiber transmission is almost negligible.
  • “Inte Serv” is built on the basis of independent stream resource reservation, using Resource Reservation Setup Protocol (RSVP).
  • RSVP Resource Reservation Setup Protocol
  • RSVP 2Mbps bandwidth can only be macroscopically. If the 1 second data set is sent in the first half of the second, it will cause problems and form periodic burst traffic. Because the core idea of the IP Internet is to do its best, at every network node, the switch always tries to forward data as fast as possible. When a video is circulated through a multi-level switch, the traffic distribution will inevitably become uneven. Multiple non-uniform non-synchronous flows together will result in greater non-uniformity over time, that is, network traffic must be periodically blocked. As the number of video users increases, there is no upper limit for periodic blocking. When the internal storage capacity of the switch exceeds the value of the switch, the result of packet loss is directly caused.
  • Diff Serv tries to provide a better than best-effort network service. This method does not require complex network-wide resource reservation, and the implementation is very simple. Just put “priority, mark, network switch in each packet first. Handling video data with "priority,” the basic principle is that banks can issue gold cards to VIP customers, which can effectively reduce the queue time of high-end customers. This method sounds good, but it still doesn't work.
  • HDTV quality video content can be transmitted.
  • most of today's web servers use Gigabit Ethernet ports, and their instantaneous traffic is dozens of times that of HDTV. If there are many similar websites that happen to collide together, the burst traffic generated in a short period of time will exceed the usage of the entire network. HDTV is required to be able to fill any lenient network. Statistical analysis shows that this type of collision is very frequent.
  • the IP Internet attempts to use storage to absorb transient traffic, with the result that transmission delays are increased. Due to limited storage capacity, there is no upper limit for bursty traffic. Therefore, the storage method can only improve the chance of packet loss of this device, and the burst traffic absorbed by this node will put more pressure on the next node. Video traffic continues to flow, and the storage mode of the switch intensifies the convergence of bursts to weak nodes, and network packet loss is inevitable.
  • the technical problem to be solved by the embodiments of the present invention is to provide a network access method for an access network device, which is used to ensure a stable transmission rate of data packet transmission and avoid delay, thereby ensuring network transmission quality.
  • the embodiment of the invention further provides a network access system of the access network device, a node server and an access switch, which are used to ensure the implementation and application of the above method in practice.
  • a network access method of an access network device which includes the following steps: The access switch is powered on, and all downlink protocol packets are directed to the CPU module in the internal downlink protocol packet address table;
  • the access switch Receiving, by the access switch, a downlink protocol packet sent by the node server, and according to the setting of the downlink protocol packet address table, directing the downlink protocol packet to a CPU module of the access switch, where the CPU module generates an uplink protocol packet, and Sending to the node server;
  • the downlink protocol packet includes an access network address to be allocated;
  • the node server sends an access network command to the access switch, where the access network command includes an access network address of the access switch, where the access network address is the to-be-allocated access in the downlink protocol packet received by the access switch.
  • the access switch updates its internal downlink protocol packet address table so that only the protocol packet whose destination address is its own access network address is directed to the CPU module.
  • the access switch enters the network.
  • the "net” entered refers to a novel network.
  • This new type of network is a centralized control network structure.
  • the network can be a tree network, a star network, a ring network, etc., but on this basis, a centralized control node is needed in the network to control the entire network.
  • the new network is divided into two parts: the access network and the metropolitan area network.
  • the devices in the access network can be mainly divided into three categories: node servers, access switches, and terminals (including various set-top boxes, code boards, memories, etc.).
  • the node server is a node in the access network that functions as a centralized control, and can control the access switch and the terminal.
  • the node server can be directly connected to the access switch or directly connected to the terminal.
  • the devices in the metropolitan area network can also be divided into three categories: metropolitan area servers, node switches, and node servers.
  • the node server is the node server of the access network part, that is, the node server belongs to both the access network part and the metropolitan area network part.
  • the metropolitan area server is a node that functions as a centralized control function in the metropolitan area network, and can control the node switch and the node server.
  • the metro server can be directly connected to the node switch or directly to the node server. It can be seen that the whole new network is a layered and centralized control network structure, and the network controlled by the node server and the metropolitan area server can be various structures such as a tree type, a star type, and a ring type.
  • the access switch is also connected to other subordinate access network devices.
  • the node server sends a port allocation packet to the access switch that has entered the network. That is, as a preferred embodiment of the present invention, when the access switch that has entered the network receives the port allocation packet sent by the node server, the method further includes the following steps:
  • the access switch that has entered the network directs the port allocation packet whose destination address is its own access network address to the CPU module.
  • the downlink port directed by each port downlink protocol packet is set.
  • the method further includes the following steps:
  • the access switch directs the port downlink protocol packet to the corresponding downlink port according to the setting of the internal downlink protocol packet address table; the port downlink protocol packet includes an access network address to be allocated;
  • the node server receives the port uplink protocol packet sent by the lower-level access network device connected to the downlink port of the access switch, sending a network access command to the lower-level access network device, where the network access command includes the lower-level connection
  • the access network address of the network access device where the access network address is the access network address to be allocated in the port downlink protocol packet received by the lower-level access network device.
  • the subordinate access network device may include an access switch or a terminal.
  • the access switch updates its internal downlink protocol packet address table according to the network access command, and sets a protocol packet whose destination address is its own access network address to the CPU module. It also sends an incoming network command response to the node server.
  • the terminal When the lower-level access network device is a terminal, the terminal sends a network access command response to the node server.
  • the node server When the node server receives the network access command response sent by the access switch, it knows that the access switch has entered the network, and then periodically sends a state downlink protocol packet to the access switch, that is, as a preferred embodiment of the present invention.
  • the access switch that has entered the network receives the state downlink protocol packet that is sent by the node server, the method further includes the following steps: The access switch that has been connected to the network directs the downlink protocol packet whose destination address is its own access network address to the CPU module according to the setting of its internal downlink protocol packet address table.
  • the CPU module generates a status uplink protocol packet and sends it to the node server.
  • the node server For a lower-level access network device connected to an access switch, the node server also periodically sends a state downlink protocol packet to the lower-level access network device, that is, as a preferred embodiment of the present invention, when When the access switch receives the status downlink protocol packet sent by the node server, the method further includes the following steps:
  • the access switch that has entered the network, according to the setting of the internal downlink protocol packet address table, directs the state downlink protocol packet whose destination address is the access network address of the lower-level access network device to the corresponding port; the state downlink protocol packet is The port is transmitted to the corresponding subordinate access network device;
  • the subordinate access network device generates a status uplink protocol packet for the received state downlink protocol packet, and sends the packet to the node server.
  • the access switch may also be configured with an uplink protocol packet address table.
  • the method further includes the following steps:
  • the access switch may further set a data packet address table.
  • the method further includes the following steps:
  • the embodiment of the present invention may further include a process in which the node server accesses the network. After the node server is powered on, the registration information of the lower-level access network device is imported to the CPU, the metropolitan area network address is obtained, and the own access network address is configured.
  • the process of connecting the node server to the network may include the following steps:
  • the node server enters a metropolitan area network, where the metropolitan area network is a network with centralized control functions, including a metropolitan area server, a node switch, and a node server, where the node exchanges
  • the machine is connected between the metropolitan area server and the node server; and the step of the node server entering the metropolitan area network includes:
  • the metropolitan area network device is connected to the metropolitan area network, and the metropolitan area server having the centralized control function in the metropolitan area network allocates a protocol label and a metropolitan area network address to the networked device;
  • the metropolitan area network device includes a node switch and a node server, and the protocol label is used to describe a connection between the metropolitan area network device and the metropolitan area server; when there is more between the same metropolitan area network device and the metropolitan area server When connecting, the metropolitan area server assigns different protocol labels to each connection;
  • the metropolitan area server For each service request across the metropolitan area network, the metropolitan area server assigns a data tag for the corresponding service, the data tag being used to describe the connection between the node servers involved in the service.
  • the label is divided into an inbound label and an outgoing label, and the inbound label refers to a label of the data packet entering the metropolitan area server or the node switch, and the outgoing label refers to the label of the data packet leaving the metropolitan area server or the node switch;
  • the inbound and outbound tags of the same packet are different tags or the same tag. More specifically, the metropolitan area network device accesses the metropolitan area network, and the steps of assigning the protocol label and the metropolitan area network address by the metropolitan area server include the following substeps:
  • the metropolitan area server sends a metropolitan area query tag packet to all its downlink ports, and each metropolitan area query tag packet contains a to-be-used protocol tag assigned by the metropolitan area server;
  • the node server After the node server is powered on, it receives the metropolitan area query label packet sent by the metropolitan area server, and then returns the metropolitan area response label packet to the metropolitan area server, where the metropolitan area response label packet includes the serial number of the metropolitan area network device and is received.
  • the port number of the metropolitan area query tag package
  • the metropolitan area server After receiving the metropolitan area response label packet, the metropolitan area server verifies whether the metropolitan area network device is registered according to the serial number in the packet. If it is already registered, it sends a network access command to the port of the metropolitan area network device that receives the metropolitan area query label packet.
  • the network entry command includes a metropolitan area network address allocated by the metropolitan area server for the metropolitan area network device and the inactive protocol label;
  • the node server accesses the network into the metropolitan area network in the manner of an IP node, that is, the metropolitan area network is an IP network structure.
  • IP node that is, the metropolitan area network is an IP network structure.
  • the embodiment of the present invention may further include the following steps: the node server sets, in its internal downlink protocol packet address table, a downlink port that is respectively directed by each downlink protocol packet;
  • the node server sends a downlink protocol packet through the corresponding downlink port according to the setting of the downlink protocol packet address table.
  • the embodiment of the present invention may further include the following steps:
  • the node server generates a port allocation package including port allocation information
  • the node server sends a port allocation packet through the corresponding downstream port according to the setting of its internal downlink protocol packet address table.
  • the node server is further provided with an address information table, and the address information table records address occupation information, device identification information, and device resource information.
  • the operation for the address information table may include the following:
  • the write device resource information is the port information of the current node server.
  • the address information table is updated according to the setting of the downlink port that is directed by each downlink protocol packet in the internal downlink protocol packet address table of the node server, and the update operation includes:
  • the address of the access network to be allocated in each downlink protocol packet is respectively written in the blank entry of the address information table, and the address occupation information is marked as the address to be used.
  • Scenario 3 Update the address information table according to the setting of the downlink port of the other access device that is directed by the port downlink protocol packet, where the update operation includes: The address of the to-be-assigned access network in each port downlink protocol packet is respectively written in the blank entry of the address information table, and the address occupation information is marked as the address to be used.
  • the device resource information is updated to be the port information of the current access switch, and the access network address information of the access network device to which the port is connected.
  • Scenario 5 When the node server receives the network access command response sent by the terminal, updating the address information table, where the update operation includes:
  • the device resource information is updated to be the port information of the current terminal, and the access network address information of the access network device to which the port is connected.
  • Scenario 6 If the node server does not receive the status uplink protocol packet sent by the access switch within a preset time, it stops sending the status downlink protocol packet to the access switch, and updates the address information table;
  • the update operations include:
  • Scenario VII If the node server does not receive the state uplink protocol packet sent by the lower-level access network device within a preset time, stop sending the state downlink protocol packet to the lower-level access network device, and update the address information.
  • Table; the update operation includes:
  • the address information table of the node server may further include uplink and downlink traffic information of the current access network device port, in order to facilitate the control and management of the communication path of the lower-level access network device and the data transmission.
  • the node server may further be provided with an uplink protocol packet address table, and the method further includes the following steps:
  • the node server When the node server is powered on, it is set in its internal uplink protocol packet address table, and all uplink protocol packets are directed to the CPU module.
  • the node server may further be configured with a data packet address table, and the method further includes the following steps:
  • the direction of all packets is set to be turned off in its internal packet address table.
  • the node server For a subordinate access network device that requests access, the node server usually first determines whether it is legal, and sends a network access command to it when it is legal.
  • the method further includes The following steps:
  • the CPU module of the node server determines whether the registration information of the access switch exists according to the network access command response packet, and if yes, determines that the access switch is legal; otherwise, it determines that the access switch is invalid;
  • the CPU module of the node server determines whether the registration information of the access network device exists according to the network access command response packet, and if yes, determines that the access network device is legal; otherwise, determines that the access network device is invalid;
  • the registration information may include a device type and a device identification.
  • the embodiment of the invention further includes a node server, which includes the following modules:
  • a downlink protocol packet sending module configured to send a downlink protocol packet to the access switch, where the downlink protocol packet includes an access network address to be allocated;
  • An uplink protocol packet receiving module configured to receive an uplink protocol packet that the access switch replies to the downlink protocol packet;
  • a first network access command sending module configured to send an access network command to the corresponding access switch according to the received uplink protocol packet, where the network access command includes an access network address allocated to the access switch, where the access network The address is the address of the access network to be allocated in the downlink protocol packet received by the access switch.
  • the node server may further include the following modules:
  • a port allocation packet sending module configured to send a port allocation packet to the access switch that has entered the network, where the port allocation packet includes port allocation information, and the port allocation information is that each port downlink protocol packet is directed to each downlink of the access switch. Port information.
  • the node server may further include the following modules: a port downlink protocol packet sending module, configured to send a port downlink protocol packet to the access switch that has entered the network; and the port downlink protocol packet includes a to-be-allocated The access network address of the port; the port uplink protocol packet receiving module, configured to receive a port uplink protocol packet sent by a lower-level access network device connected to a downlink port of the access switch;
  • a second network access command sending module configured to send a network access command to the lower-level access network device, where the network access command includes an access network address of the lower-level access network device, where the access network address is the lower-level access The access network address to be allocated in the port downlink protocol packet received by the network device.
  • the node server may further include:
  • the network access response receiving module is configured to receive an incoming network command response sent by the lower-level access network device.
  • the subordinate access network device may include an access switch or a terminal.
  • the node server further includes the following modules:
  • a status downlink protocol packet sending module configured to periodically send a status downlink protocol packet to an access switch that has entered the network;
  • the first receiving module of the status uplink protocol packet is configured to receive a status uplink protocol packet that is replied by the access switch for the status downlink protocol packet.
  • the node server further includes the following modules:
  • the second uplink receiving module of the state uplink protocol packet is configured to receive a state uplink protocol packet that is sent by the lower-level access network device for the downlink protocol packet in the state.
  • the node server further includes the following modules:
  • the initialization module is configured to import the registration information of the lower-level access network device to the CPU after power-on, obtain the metropolitan area network address, and configure its own access network address.
  • the node server may further include the following modules:
  • a protocol label and an address obtaining module configured to obtain an allocated protocol label and a metropolitan area network address from the metropolitan area server after accessing the metropolitan area network, where the protocol label is used to describe a connection between the node server and the metropolitan area server;
  • the upper-level connection device includes a node switch and a metropolitan area server;
  • the metropolitan area network is a network with centralized control functions.
  • the utility model includes a metropolitan area server, a node switch, and a node server, wherein the node switch is connected between the metropolitan area server and the node server;
  • a data label obtaining module configured to obtain, for each service request across the metropolitan area network, a data label of the allocated service to the secondary service from the metropolitan area server, where the data label is used to describe a connection between the node servers involved in the service ;
  • a label adding module configured to send a protocol packet or a data packet to a protocol packet or a data packet sent by the node server to the metropolitan area network;
  • a label deletion module is configured to remove a corresponding protocol label or data label from a protocol packet or a data packet received from the metropolitan area network.
  • the label is divided into an inbound label and an outgoing label, and the inbound label refers to a label of the data packet entering the metropolitan area server or the node switch, and the outgoing label refers to the label of the data packet leaving the metropolitan area server or the node switch;
  • the inbound label and the outgoing label of the same data packet are different labels, or are the same label; the label includes a protocol label and a data label. More preferably, the node server may further include:
  • An address-to-label mapping table configured to record, for each service across the metropolitan area network, a binding relationship between an access network address and an outgoing label of two terminals across the metropolitan area network; wherein, the access network address is each node The address assigned by the server to the networked device connected to it;
  • the label adding module searches the outbound label corresponding to the protocol packet or the data packet sent by the node server to the metropolitan area network according to the address-tag mapping table, and adds the found outbound label to be sent.
  • the node server may further include:
  • the protocol package label table is configured to set all the metropolitan area protocol packets to the CPU module when the node server is powered on, and the metropolitan area protocol package includes the metropolitan area query label package and the metropolitan area service application package sent by the metropolitan area server.
  • the node server may further include the following modules:
  • the response packet label table is used to direct the metropolitan area response label packet to the corresponding uplink port;
  • the response packet label table initialization module is configured to set the direction of all the urban response label packets to be closed when the node server is powered on;
  • the response packet label table configuration module is configured to: after receiving the metropolitan area query label packet sent by the metropolitan area server, modify the response packet label table of the metropolitan area server, and direct the metropolitan area response label packet corresponding to the protocol label to receive the metropolitan area query.
  • the upstream port of the tag packet is configured to: after receiving the metropolitan area query label packet sent by the metropolitan area server, modify the response packet label table of the metropolitan area server, and direct the metropolitan area response label packet corresponding to the protocol label to receive the metropolitan area query.
  • the node server may include the following modules:
  • the IP access network module is configured to enter the metropolitan area network by using an IP node, where the metropolitan area network is an IP network structure.
  • the node server may further include
  • the No. 0 table initializes the configuration module, which is used to initialize its internal downlink protocol packet address table for the downward shutdown of each downlink protocol packet upon power-on.
  • the node server may further include:
  • the first setting module of the No. 0 table is configured to set, in the internal downlink protocol packet address table, a downlink port respectively directed by each downlink protocol packet;
  • the downlink protocol packet sending module sends the downlink protocol packet through the corresponding downlink port according to the setting of the downlink protocol packet address table. More preferably, the node server may further include:
  • a port allocation packet generating module configured to generate a port allocation package including the port allocation information
  • the port allocation packet sending module sends the port allocation packet through a corresponding downlink port according to the setting of its internal downlink protocol packet address table.
  • the node server is further provided with an address information table, and the address information table records address occupation information, device identification information, and device resource information.
  • the node server may further include the following modules according to different situations of operation of the address information table:
  • the node server further includes:
  • the address information table initializing module is configured to initialize an address information table when the power is turned on, where the initializing includes:
  • the write device resource information is the port information of the current node server.
  • the node server also includes:
  • the first update module of the address information table is configured to update the address information table according to the setting of the downlink port that is directed by each downlink protocol packet in the downlink protocol packet address table of the node server, and the update includes:
  • the address of the access network to be allocated in each downlink protocol packet is respectively written in the blank entry of the address information table, and the address occupation information is marked as the address to be used.
  • the node server also includes:
  • the second update module of the address information table is configured to update the address information table according to the setting of the downlink port of the other access device that is directed to the port downlink protocol packet, where the update includes:
  • the node server also includes:
  • the third update module of the address information table is configured to update the address information table when receiving an incoming network command response sent by the access switch, where the update includes:
  • the device resource information is updated to be the port information of the current access switch, and the access network address information of the access network device to which the port is connected.
  • the node server also includes:
  • the address information table fourth update module is configured to update the address information table when receiving a network access command response sent by the lower-level access network device, where the update includes:
  • the address occupation information in the access network address corresponding entry of the lower-level access network device is marked as "the address is used;
  • the device resource information is updated to be the port information of the current lower-level access network device, and the access network address information of the access network device to which the port is connected.
  • the node server also includes:
  • the first downlink sending module of the state downlink protocol packet is configured to stop sending the state downlink protocol packet to the access switch when the state uplink protocol packet sent by the access switch is not received within a preset time;
  • a fourth update module of the address information table configured to update the address information table when the sending of the status downlink protocol packet to the access switch is stopped;
  • the node server also includes:
  • the second downlink sending module of the state downlink protocol packet is configured to stop sending the state downlink protocol packet to the lower-level access network device after receiving the state uplink protocol packet sent by the lower-level access network device within a preset time;
  • the fifth update module of the address information table is configured to update the address information table when the sending of the status downlink protocol packet to the lower-level access network device is stopped; the updating includes:
  • the address information table further includes uplink and downlink traffic information of the current access network device port, in order to facilitate control and management of the communication path of the lower-level access network device and the data transmission.
  • the node server is further provided with an uplink protocol packet address table, and the node server further includes:
  • Table 1 initializes the configuration module, which is used to set in the internal uplink protocol packet address table when powering up, and all uplink protocol packets are directed to the CPU module.
  • the node server is further provided with a data packet address table, and the node server further includes:
  • the packet address table configuration module when used for power-on, sets the direction of all packets to be closed in its internal packet address table.
  • the node server For a subordinate access network device that requests access, the node server usually first determines whether it is legal, and sends a network access command to it when it is legal.
  • the node server further includes :
  • a first verification module configured to determine whether the registration information of the access switch exists, and if yes, determine that the access switch is legal; otherwise, determine that the access switch is invalid; if legal, trigger the first network access command sending module .
  • a second verification module configured to determine whether the registration information of the access network device exists, and if yes, determine that the access network device is legal; otherwise, determine that the access network device is invalid; if legal, trigger the second access network Command sending module.
  • the embodiment of the invention also discloses an access switch, which comprises the following modules:
  • the No. 0 table initialization configuration module is configured to set all downlink protocol packets to the CPU module in the internal downlink protocol packet address table when the power is turned on; a downlink protocol packet receiving module, configured to receive a downlink protocol packet sent by the node server, and according to the setting of the downlink protocol packet address table, direct the downlink protocol packet to a CPU module of the access switch, where the downlink protocol packet is included Include an access network address to be allocated; an uplink protocol packet replying module, configured to generate an uplink protocol packet by the CPU module, and send the uplink protocol packet to the node server;
  • the first network access command receiving module is configured to receive a network access command sent by the node server, where the network access command includes an access network address of the access switch, where the access network address is a downlink protocol packet received by the access switch.
  • the first setting module of the No. 0 table is used to update the internal downlink protocol packet address table, and only the protocol packet whose destination address is its own access network address is directed to the CPU module.
  • the access switch After accessing the network, the access switch further includes the following modules:
  • a port allocation packet receiving module configured to receive a port allocation packet sent by the node server after entering the network; the port allocation packet includes port allocation information, where the port allocation information is that each port downlink protocol packet is directed to each downlink port of the access switch. Information;
  • a first guiding module configured to direct a port allocation packet whose destination address is its own access network address to the CPU module
  • the second setting module of the No. 0 table is configured to set, according to the port allocation information, a downlink port directed by each port downlink protocol packet in its internal downlink protocol packet address table.
  • the access switch further includes the following modules:
  • a port downlink protocol packet receiving module configured to receive a port downlink protocol packet sent by the node server after entering the network; the port downlink protocol packet includes an access network address to be allocated;
  • the second guiding module is configured to direct the port downlink protocol packet to the corresponding downlink port according to the setting of the internal downlink protocol packet address table.
  • the access switch further includes a network access command response module, configured to send a network access command response to the node server.
  • the receiving switch further includes: when receiving the status downlink protocol packet that is sent by the node server. a state downlink protocol packet receiving module, configured to receive a state downlink protocol packet periodically sent by the node server after entering the network;
  • the third guiding module is configured to direct, according to the setting of the internal downlink protocol packet address table, the downlink protocol packet whose destination address is the access network address of the user to the CPU module;
  • the status uplink protocol packet replying module is configured to generate a status uplink protocol packet by the CPU module, and send the status uplink protocol packet to the node server.
  • the access switch further includes:
  • the fourth guiding module is configured to, according to the setting of the internal downlink protocol packet address table, direct the state downlink protocol packet whose destination address is the access network address of the lower-level access network device to the corresponding port; The port is transmitted to the corresponding lower level access network device.
  • the access switch is further provided with an uplink protocol packet address table, and the access switch further includes:
  • Table 1 initializes the configuration module, which is used to set all upstream protocol packet-oriented uplink ports in its internal uplink protocol packet address table when powering up.
  • the access switch is further provided with a data packet address table, and the access switch further includes:
  • the packet address table configuration module is used to set the direction of all packets to be closed in the internal packet address table on power-up.
  • the embodiment of the present invention further discloses a network access system of an access network device, where the access network device includes a node server and an access switch connected to the lower level, and the node server includes sending a downlink protocol packet to the access switch. a downlink protocol packet sending module, and a first network access command sending module that sends an incoming network command according to an uplink protocol packet replied by the access switch;
  • the access switch includes:
  • the No. 0 table initialization configuration module is configured to set all downlink protocol packets to the CPU module in the internal downlink protocol packet address table when the power is turned on; a downlink protocol packet receiving module, configured to direct, according to the setting of the downlink protocol packet address table, the received downlink protocol packet to a CPU module of the access switch, where the downlink protocol packet includes an access network address to be allocated ;
  • An uplink protocol packet replying module configured to generate an uplink protocol packet by the CPU module, and send the uplink protocol packet to the node server;
  • the first network access command receiving module is configured to receive a network access command sent by the node server, where the network access command includes an access network address of the access switch, where the access network address is a downlink protocol packet received by the access switch.
  • the first setting module of the No. 0 table is used to update the internal downlink protocol packet address table, and only the protocol packet whose destination address is its own access network address is directed to the CPU module.
  • the access switch is also connected to other subordinate access network devices.
  • the node server sends a port allocation packet to the access switch that has entered the network, which is a preferred implementation of the present invention.
  • the node server further includes a port allocation packet sending module, configured to send a port allocation packet to the access switch that has entered the network, where the port allocation packet includes port allocation information, and the port allocation information is a downlink protocol packet of each port. Information directed to each downlink port of the access switch;
  • the access switch further includes:
  • a first guiding module configured to direct a port allocation packet whose destination address is its own access network address to the CPU module
  • the second setting module of the No. 0 table is configured to set, according to the port allocation information, a downlink port directed by each port downlink protocol packet in its internal downlink protocol packet address table.
  • the node server further includes a port downlink protocol packet sending module, configured to send a port downlink protocol packet to the access switch that has entered the network, where the port downlink protocol packet includes an access network address to be allocated;
  • the access switch further includes:
  • the second guiding module is configured to direct the port downlink protocol packet to the corresponding downlink port according to the setting of the internal downlink protocol packet address table.
  • the access network device further includes a lower-level access network device connected to a downlink port of the network access switch, where the node server further includes, for the lower-level access network The device sends a second network access command sending module to the network access command; the lower-level access network device includes:
  • a port uplink protocol packet replying module configured to generate a port uplink protocol packet for the received port downlink protocol packet, and send the packet to the node server;
  • the second network access command receiving module is configured to receive a network access command sent by the node server, where the network access command includes an access network address of the lower-level access switch, where the access network address is a receiving port of the lower-level access switch. The address of the access network to be allocated in the downlink protocol packet.
  • the lower-level access network device includes an access switch or a terminal.
  • the access switch further includes: a third setting module of the No. 0 table, configured to update the internal downlink protocol packet address table according to the network access command, and set the purpose.
  • the protocol packet whose address is its own access network address is directed to the CPU module.
  • the node server When the node server receives the network access command response sent by the access switch, it knows that the access switch has entered the network, and then periodically sends a state downlink protocol packet to the access switch, that is, as a preferred embodiment of the present invention.
  • the node server further includes a state downlink protocol packet sending module configured to periodically send a state downlink protocol packet to the access switch that has entered the network;
  • the access switch further includes:
  • the third guiding module is configured to direct, according to the setting of the internal downlink protocol packet address table, the downlink protocol packet whose destination address is the access network address of the user to the CPU module;
  • the status uplink protocol packet replying module is configured to generate a status uplink protocol packet by the CPU module, and send the status uplink protocol packet to the node server.
  • the node server For a lower-level access network device connected to an access switch, the node server also periodically sends a state downlink protocol packet to the lower-level access network device, that is, as a preferred embodiment of the present invention, the access The switch also includes: The fourth guiding module is configured to, according to the setting of the internal downlink protocol packet address table, direct the state downlink protocol packet whose destination address is the access network address of the lower-level access network device to the corresponding port; The port is transmitted to the corresponding lower level access network device.
  • the lower-level access switch further includes:
  • the status uplink protocol packet replying module is configured to generate a status uplink protocol packet for the received status downlink protocol packet, and send the status uplink protocol packet to the node server.
  • the terminal further includes:
  • the status uplink protocol packet replying module is configured to generate a status uplink protocol packet for the received status downlink protocol packet, and send the status uplink protocol packet to the node server.
  • the access switch is further provided with an uplink protocol packet address table, and the access switch further includes:
  • Table 1 initializes the configuration module, which is used to set all upstream protocol packet-oriented uplink ports in its internal uplink protocol packet address table when powering up.
  • the access switch further includes a data packet address table, and the access switch further includes:
  • the packet address table configuration module is used to set the direction of all packets to be closed in the internal packet address table on power-up.
  • the node server when the node server is connected to the network, the node server further includes: an initialization module, configured to import the registration information of the lower-level access network device after power-on to
  • CPU obtain the metropolitan area network address, and configure its own access network address.
  • the node server further includes the following modules for accessing the metropolitan area network:
  • a protocol label and an address obtaining module configured to obtain an allocated protocol label and a metropolitan area network address from the metropolitan area server after accessing the metropolitan area network, where the protocol label is used to describe a connection between the node server and the metropolitan area server;
  • the upper-level connection device includes a node switch and a metropolitan area server;
  • the metropolitan area network is a network with centralized control functions.
  • the utility model includes a metropolitan area server, a node switch, and a node server, wherein the node switch is connected between the metropolitan area server and the node server;
  • a data label obtaining module configured to obtain, for each service request across the metropolitan area network, a data label of the allocated service to the secondary service from the metropolitan area server, where the data label is used to describe a connection between the node servers involved in the service ;
  • a label adding module configured to send a protocol packet or a data packet to a protocol packet or a data packet sent by the node server to the metropolitan area network;
  • a label deletion module is configured to remove a corresponding protocol label or data label from a protocol packet or a data packet received from the metropolitan area network.
  • the label is divided into an inbound label and an outgoing label, and the inbound label refers to a label of the data packet entering the metropolitan area server or the node switch, and the outgoing label refers to the label of the data packet leaving the metropolitan area server or the node switch;
  • the inbound label and the outgoing label of the same data packet are different labels, or are the same label; the label includes a protocol label and a data label.
  • the node server further includes:
  • An address-to-label mapping table configured to record, for each service across the metropolitan area network, a binding relationship between an access network address and an outgoing label of two terminals across the metropolitan area network; wherein, the access network address is each node The address assigned by the server to the networked device connected to it;
  • the label adding module searches the outbound label corresponding to the protocol packet or the data packet sent by the node server to the metropolitan area network according to the address-tag mapping table, and adds the found outbound label to be sent.
  • the node server further includes:
  • the protocol package label table is configured to set all the metropolitan area protocol packets to the CPU module when the node server is powered on, and the metropolitan area protocol package includes the metropolitan area query label package and the metropolitan area service application package sent by the metropolitan area server.
  • the node server further includes:
  • the response packet label table is configured to direct the metropolitan area response label packet to the corresponding uplink port respectively;
  • the response packet label table initialization module is configured to set the direction of all the metropolitan area response label packets to be closed when the node server is powered on;
  • the response packet label table configuration module is configured to: after receiving the metropolitan area query label packet sent by the metropolitan area server, modify the response packet label table of the metropolitan area server, and direct the metropolitan area response label packet corresponding to the protocol label to receive the metropolitan area query.
  • the upstream port of the tag packet is configured to direct the metropolitan area response label packet to the corresponding uplink port respectively.
  • the node server further includes an IP access network module for accessing the metropolitan area network, and is configured to enter the metropolitan area network by using an IP node, where the metropolitan area network is an IP network structure.
  • the node server further includes:
  • the No. 0 table initializes the configuration module, which is used to initialize its internal downlink protocol packet address table for the downward shutdown of each downlink protocol packet upon power-on.
  • the node server further includes:
  • the first setting module of the No. 0 table is configured to set, in the internal downlink protocol packet address table, a downlink port respectively directed by each downlink protocol packet;
  • the downlink protocol packet sending module sends the downlink protocol packet through the corresponding downlink port according to the setting of the downlink protocol packet address table.
  • the node server further includes:
  • a port allocation packet generating module configured to generate a port allocation package including the port allocation information
  • the port allocation packet sending module sends the port allocation packet through a corresponding downlink port according to the setting of its internal downlink protocol packet address table.
  • the node server is further provided with an address information table, and the address information table records address occupation information, device identification information, and device resource information.
  • the node server may further include the following modules according to different situations of operation of the address information table:
  • the node server further includes:
  • the address information table initializing module is configured to initialize an address information table when the power is turned on, where the initializing includes:
  • the write device resource information is the port information of the current node server.
  • the node server further includes:
  • the first update module of the address information table is configured to update the address information table according to the setting of the downlink port that is directed by each downlink protocol packet in the downlink protocol packet address table of the node server, and the update includes:
  • the address of the access network to be allocated in each downlink protocol packet is respectively written in the blank entry of the address information table, and the address occupation information is marked as the address to be used.
  • the node server further includes:
  • the second update module of the address information table is configured to update the address information table according to the setting of the downlink port of the other access device that is directed to the port downlink protocol packet, where the update includes:
  • the address of the to-be-assigned access network in each port downlink protocol packet is respectively written, and the address occupation information is marked as the address to be used.
  • the node server further includes:
  • the third update module of the address information table is configured to update the address information table when receiving an incoming network command response sent by the access switch, where the update includes:
  • the device resource information is updated to be the port information of the current access switch, and the access network address information of the access network device to which the port is connected.
  • the node server further includes:
  • the address information table fourth update module is configured to update the address information table when receiving a network access command response sent by the lower-level access network device, where the update includes:
  • the address occupation information in the access network address corresponding entry of the lower-level access network device is marked as "the address is used;
  • the device resource information is updated to be the port information of the current lower-level access network device, and the access network address information of the access network device to which the port is connected.
  • the node server further includes:
  • the first downlink sending module of the state downlink protocol packet is configured to stop sending the state downlink protocol packet to the access switch when the state uplink protocol packet sent by the access switch is not received within a preset time;
  • a fourth update module of the address information table configured to update the address information table when the sending of the status downlink protocol packet to the access switch is stopped;
  • the node server further includes:
  • the second downlink sending module of the state downlink protocol packet is configured to stop sending the state downlink protocol packet to the lower-level access network device after receiving the state uplink protocol packet sent by the lower-level access network device within a preset time;
  • the fifth update module of the address information table is configured to update the address information table when the sending of the status downlink protocol packet to the lower-level access network device is stopped; the updating includes:
  • the address information table of the node server further includes uplink and downlink traffic information of the current access network device port.
  • the node server is further provided with an uplink protocol packet address table, and the node server further includes:
  • Table 1 initializes the configuration module, which is used to set in the internal uplink protocol packet address table when powering up, and all uplink protocol packets are directed to the CPU module.
  • the node server is further provided with a data packet address table, and the node server further includes:
  • the packet address table configuration module when used for power-on, sets the direction of all packets to be closed in its internal packet address table.
  • the node server further includes:
  • a first verification module configured to determine whether the registration information of the access switch exists, and if yes, determine that the access switch is legal; otherwise, determine that the access switch is invalid; if legal, trigger the first network access command sending module .
  • the node server further includes:
  • the node server of the present invention mainly includes a network interface module, a switching engine module, a CPU module, and a disk array module.
  • the access switch of the present invention mainly includes a network interface module (a downlink network interface module).
  • the uplink network interface module the switching engine module
  • the CPU module may be configured with an address table when receiving the packet, or may parse the packet and generate a response packet. Therefore, in order to enable a person skilled in the art to better understand the present invention, in the embodiment of the present invention, the modules involved in the network access system of the node server, the access switch, and the access network device of the present invention are mainly described from a functional perspective. In essence, however, these functional modules correspond to actual hardware modules. Compared with the prior art, the present invention has the following advantages:
  • multimedia services are significantly different from the basic conditions necessary for traditional data services (such as web text, images, email, FTP, and DNS services).
  • multimedia services are particularly sensitive to end-to-end latency and latency changes, but can tolerate occasional data loss.
  • the present invention allocates the address of each network device by sending a query packet to each communication port in a master control manner, and establishes a clear network topology on the master server side in the process of allocation.
  • the corresponding communication link can be directly allocated by the main control server (because it clarifies the device topology of the entire network), and no routing negotiation is required between the respective network devices (
  • the existing IP protocol scheme ensures a stable transmission rate and avoids delays.
  • each data packet of each service is transmitted through the same communication link, unlike the solution of the existing IP protocol, each data packet relies on self-negotiation to solve the routing problem in the data packet. Before sending, it does not know which path it will go through, that is, two packets of the same service may be transmitted to the target terminal through different paths. Thus, the present invention can ensure a stable transmission rate and avoid delay as compared with the present invention.
  • FIG. 1 is a schematic diagram showing the hardware structure of a node server according to the present invention.
  • FIG. 2 is a schematic structural diagram of hardware of an access switch according to the present invention.
  • FIG. 3 is a schematic diagram of a network access process of an access switch according to the present invention:
  • FIG. 4 is a schematic diagram of a connection between a node server and an access switch according to the present invention
  • FIG. 5 is a schematic diagram of a terminal network access process of the present invention:
  • FIG. 6 is a schematic diagram of connection of a node server, an access switch, and a terminal according to the present invention
  • FIG. 7 is a schematic diagram of a node switch network access process of the present invention.
  • FIG. 8 is a schematic diagram of connection of a metropolitan area server, a node switch, and a node server according to the present invention.
  • Figure 9 is a diagram showing the connection of a metropolitan area server, a node switch, a node server, and a terminal of the present invention.
  • the inventor of the present invention believes that the sufficient conditions for realizing the quality assurance of the whole network of the present invention are as follows:
  • the mechanism of "Best Efforts" in the core theory of IP Internet will inevitably lead to uneven network traffic and frequent packet loss.
  • the TCP protocol uses the network packet loss status to regulate the transmission traffic.
  • the inventor of the present invention believes that various current QoS methods are based on a false assumption. Based on this assumption, the solution to QoS is to provide priority for video traffic. But the fact is, because the network traffic required by different media forms is extremely uneven, as long as a few people use video services, the video traffic on the network will occupy the absolute main body.
  • the invention is like a highway, does not require police and traffic lights, cement-separated lanes and overpasses to ensure that the car is on a prescribed road. According to the experience of the California Department of Transportation, the way to avoid highway traffic jams is to close the entrance ramp.
  • Each path calculates and measures the flow, once the flow is near saturation, take a detour or reject new users.
  • the TV of the embodiment of the present invention can achieve a packet loss rate of one millionth of a million under a heavy load of 90%.
  • Computer files and streaming media are two distinct forms of media that are mutually exclusive.
  • the network theory and practice of the present invention reveals two outcomes: •
  • the present invention is a hundred times better than the IP Internet.
  • the PSTN telephone network uses a strict synchronization mechanism, and network congestion does not occur until the traffic is 100% full. In theory, after multiple merges are merged, they are evenly streamed. Practice has further proved that under the premise of uniform flow, network traffic can be close to the limit value without packet loss. Because it occupies more than 90% of the video traffic in future network traffic, it has its own characteristics. Therefore, the way of the Internet quality assurance of the present invention, which is mainly aimed at the video service, is to eliminate the unevenness of the source traffic, especially to prevent the packet loss phenomenon of the network switch under heavy load conditions.
  • a connection-oriented circuit is established by using an improved Ethernet, and a fixed-length data packet is uniformly used in the entire network. As long as the packet time interval is changed, a media stream of any bandwidth can be obtained.
  • the Internet of the present invention requires that the terminal design must have a current sharing capability. However, in the actual network environment, it is obviously impossible to hope that the user will consciously abide by the current sharing regulations. Therefore, in the embodiment of the present invention, the node server issues a pass to the network switch, and only allows the user data packet to pass through with a very fine time precision. For user terminals designed to meet the specified requirements, the pass is completely transparent.
  • the network practice has obtained satisfactory results, and the switch of the present invention can obtain a overloaded packet loss rate of more than one million in a condition of 90% bandwidth utilization.
  • the new network is a centralized control network structure.
  • the network can be a tree network, a star network, a ring network, etc., but on this basis, a centralized control node is needed in the network to control the entire network.
  • the new network is divided into two parts: the access network and the metropolitan area network.
  • the devices in the access network can be mainly divided into three categories: node servers, access switches, and terminals (including various set-top boxes, code boards, memories, etc.).
  • the node server is a node in the access network that functions as a centralized control, and can control the access switch and the terminal.
  • the node server can be directly connected to the access switch or directly connected to the terminal.
  • the devices in the metropolitan area network can also be divided into three categories: metropolitan area servers, node switches, and node servers.
  • the node server is the node server of the access network part, that is, the node server belongs to both the access network part and the metropolitan area network part.
  • the metropolitan area server is a node that functions as a centralized control function in the metropolitan area network, and can control the node switch and the node server.
  • the metro server can be directly connected to the node switch or directly to the node server. It can be seen that the whole new network is a layered and centralized control network structure, and the network controlled by the node server and the metropolitan area server can be various structures such as a tree type, a star type, and a ring type.
  • the devices in the novel network system of the present invention can be mainly classified into three categories: servers, switches (including Ethernet gateways), terminals (including various set top boxes, code boards, memories, etc.).
  • the new network can be divided into a metropolitan area network (or national network, global network, etc.) and an access network as a whole.
  • the devices in the access network can be mainly divided into three categories: node server, access switch (including Ethernet gateway), terminal (including various set-top boxes, code boards, memory, etc.).
  • each access network device is:
  • the network interface module 101, the switching engine module 102, the CPU module 103, and the disk array module are mainly included;
  • the packets from the network interface module 101, the CPU module 103, and the disk array module 104 all enter the switching engine module 102;
  • the switching engine module 102 performs an operation of checking the address table 105 for the incoming packet, thereby obtaining the packet guiding information;
  • the packet's navigation information stores the packet in the queue of the corresponding packet buffer 106; if the packet buffer 106 is nearly full, it is discarded;
  • the switching engine module 102 polls all packet buffer queues, and forwards if the following conditions are met: : 1) The port send buffer is not full; 2) The queue packet counter is greater than zero.
  • the disk array module 104 mainly implements control of the hard disk, including initialization, reading and writing, and the like of the hard disk.
  • the CPU module 103 is mainly responsible for protocol processing with the access switch and the terminal (not shown), and the address table 105 is The configuration of the downlink protocol packet address table, the uplink protocol packet address table, and the packet address table, and the configuration of the disk array module 104.
  • the network interface module (the downlink network interface module 201, the uplink network interface module 202), the switching engine module 203, and the CPU module 204 are mainly included;
  • the packet (uplink data) coming in by the downlink network interface module 201 enters the packet detecting module 205; the packet detecting module 205 detects whether the packet destination address (DA), the source address (SA), the packet type, and the packet length meet the requirements, if If yes, the corresponding stream identifier (stream-id) is assigned and enters the switching engine module 203, otherwise discarded; the incoming packet (downlink data) of the upstream network interface module 202 enters the switching engine module 203; the data entered by the CPU module 204 The packet enters the switching engine module 203; the switching engine module 203 performs an operation of checking the address table 206 for the incoming packet to obtain the packet guiding information; if the packet entering the switching engine module 203 is the downlink network interface to the upstream network interface, The packet is stored in the queue of the corresponding packet buffer 207 in conjunction with the stream identifier (stream-id); if the queue of the packet buffer 207 is nearly full, it is discarded; if the packet entering the switching engine
  • the switching engine module 203 polls all packet buffer queues. In the embodiment of the present invention, there are two cases: If the queue is the downlink network interface to the uplink network interface, the following conditions are met: 1) the port transmission buffer is not full; 2) the queue packet counter is greater than zero; 3) obtaining the token generated by the rate control module ;
  • the queue is not the downlink network interface to the upstream network interface, the following conditions are met: 1) The port transmission buffer is not full; 2) The queue packet counter is greater than zero.
  • the rate control module 208 is configured by the CPU module 408 to generate a token for all downlink network interfaces to the packet buffer queue of the uplink network interface in a programmable interval to control the bit rate of the uplink forwarding.
  • the CPU module 248 is primarily responsible for protocol processing with the node server, configuration of the address table 206, and configuration of the rate control module 208.
  • the utility model mainly comprises a network interface module, a service processing module and a CPU module; for example, the set top box mainly comprises a network interface module, an audio and video codec engine module and a CPU module; the coding board mainly comprises a network interface module, an audio and video coding engine module and a CPU module; It mainly includes network interface module, CPU module and disk array module.
  • the equipment of the metropolitan area network can be mainly divided into two categories: node server, node switch, and metropolitan area server.
  • the node switch mainly includes a network interface module, a switching engine module, and a CPU module;
  • the metropolitan area server mainly includes a network interface module, a switching engine module, and a CPU module.
  • the data packet of the access network mainly includes the following parts: destination address (DA), source address (SA), reserved byte, pay l oad (PDU), CRC.
  • DA destination address
  • SA source address
  • PDU pay l oad
  • the data packets of the access network mainly include the following parts:
  • the destination address (DA) consists of 8 bytes (byte), the first byte indicates the type of packet (such as various protocol packets, multicast packets, unicast packets, etc.), and there are up to 256 possibilities.
  • the second byte to the sixth byte are the metropolitan area network addresses, and the seventh and eighth bytes are the access network addresses;
  • the source address (SA) is also composed of 8 bytes, and the definition is the same as the destination address (DA);
  • the reserved byte consists of 2 bytes
  • the CRC consists of 4 bytes and its calculation method follows the standard Ethernet CRC algorithm.
  • the topology of the metropolitan area network is a pattern. There may be two or even more than two connections between the two devices, that is, the node switch and the node server, the node switch, and the node switch, the node switch, and the node server may exceed 2 kinds of connections.
  • the metropolitan area network address of the metropolitan area network device is unique.
  • the parameter: label is introduced in the embodiment of the present invention to uniquely describe a metropolitan area network device.
  • the definition of the label in this specification is similar to the definition of the label of MPLS (Mu lt i-Protoco l Labe l Swi t ch, multi-protocol label switching). Assuming that there are two connections between device A and device B, then the packet is slave. A to device B has 2 tags, and the packet has 2 tags from device B to device A. The tag is tagged and tagged. Assume that the tag of the packet entering device A (inbound tag) is 0x0000, and the tag (outbound tag) when the packet leaves device A may become 0x0001.
  • the access network process of the metropolitan area network is the network access process under centralized control, which means that the address allocation and label allocation of the metropolitan area network are dominated by the metropolitan area server, and the node switch and the node server are passively executed.
  • the label assignment with MPLS is different.
  • the label assignment of MPLS is the result of mutual negotiation between the switch and the server.
  • the metropolitan area network data packet mainly includes the following parts: That is, destination address (DA), source address (SA), reserved byte (Reserved), label, payload (PDU), CRC.
  • the format of the label can be defined as follows:
  • the label is 32 bits, wherein the high 16 bits are reserved, and only the lower 16 bits are used, and its position is between the reserved bytes of the data packet and the payload.
  • a downlink protocol packet (a protocol packet sent by the node server to the access switch and the terminal); an uplink protocol packet (by the access switch, the terminal) a protocol packet that responds to the node server); a unicast packet;
  • the address of the access network is 16 bits in total, so the total number of access switches and terminals that can be accessed is 65536. It is assumed that the datagram type of the downlink protocol packet is "1000 0000" (binary), that is, 0x80 (hexadecimal). The datagram type of the uplink protocol packet is "00001000” (binary), that is, 0x08 (hexadecimal), and the datagram type of the unicast packet is "0001 0000" (binary), that is, 0x10 (hexadecimal) System), the datagram type of the multicast packet is "01111000" (binary), that is, 0x78 (hexadecimal). By combining the same items, the address table of 4 bar and 8 bit length can be mapped to a 2 bit long address table. E.g:
  • the output of the address table indicates the packet-oriented port.
  • one of the access switches BX-008 has one uplink 100 Mbps network port, eight downstream 100 Mbps network ports, and one CPU module interface. If the eight downstream 100M network ports are defined as ports 0 to 7, respectively, the CPU module interface is defined as port 8, and one upstream 100M port is defined as port 9, which requires a total of 256K X lObit addresses.
  • the output of the address table is "00 0000 0001" for packet-oriented port 0, "11 0000 0000" for packet-oriented port 8, port 9, and so on.
  • each access switch that is allowed to access the network must be registered in the node server, and the unregistered access switch cannot access the network.
  • the process of accessing the access switch into the network involves the following steps:
  • the node server sends a query packet to each port, and the access switch sends a response packet after receiving the query packet, where the response packet includes registration information of the current access switch.
  • the node server After receiving the response packet sent by the access switch, the node server knows which port is connected to an access switch, and then finds the access switch information in the registration information table of the node server to the access switch. Sending a network access command (informing the access network address), the access switch enters the network after receiving the network access command, and sends a network access command response to the node server; S3. After receiving the network access command sent by the access switch, the node server knows that the access switch has entered the network, and periodically sends a status inquiry packet to the port to check whether the access switch works normally, and also to the The downlink port of the access switch sends a port query packet to check whether other access network devices are connected to the access switch.
  • a status query response is sent to the node server after receiving the device status query command.
  • the node server does not receive the status query response within a certain period of time, it considers that the access switch has been removed from the network, no longer sends the status query packet, and continues to send the query packet to the port.
  • MSS-400 server After the MSS-400 server is powered on, initialize the hardware and obtain the default metropolitan area network address (assumed to be 0x00 0x0000 0x0000). Import the configuration file from the hard disk to the CPU memory (such as the registration information of the switch, the registration information of the terminal, etc.), MSS. -400 server configures its own access network address to 0x0000;
  • MSS-400 server initialization 0, 1, 2, 3 table:
  • the MSS-400 server knows that it has 8 downlink ports, so it configures the entries of the eight 0 tables:
  • the packet is received into its CPU module. After the CPU parses the query packet, it generates a response packet (the response packet contains the registration information of the switch) and sends it to the MSS-400 server.
  • the DA of the response packet is 0x0800 0x0000 0x0000 0x0000.
  • SA is 0x0000 0x0000 0x0000 0x0001;
  • the MSS-400 server After receiving the response packet sent by the BX-008-0 switch, the MSS-400 server compares the source address (SA) of the response packet with the device type to know that its port 0 is connected to an access switch, and then The information of the access switch is found in the registration information table of the node server, and the network access command is sent to the access switch (informing that the access network address is 0x0001);
  • the BX-008-0 switch After receiving the network access command, the BX-008-0 switch knows that its access network address is 0x0001 and enters the network. Therefore, its 0th table "00 0000 0000 0001" is set to "01 0000 0000", and 0 The remaining entries in the table are configured as "00 0000 0000", that is, only the downstream protocol packet of the switch is imported into the CPU, and the incoming network command response is sent to the MSS-400 server.
  • the MSS-400 server receives the incoming command response from the BX-008-0 switch and knows that the BX-008-0 switch has entered the network. After that, it sends a device status query command to this port every second, checking BX-008-0. Whether the switch works normally, and also sends the port downlink protocol packet to the downlink port of the BX-008-0 switch to check whether other access network devices are connected under the access switch. In this case, the MSS-400 server Will do the following configuration in its table 0:
  • the destination address (DA) is 0x8000 0x0000 0x0000 0x000a of the port downlink protocol packet oriented port 0;
  • the destination address (DA) is 0x8000 0x0000 0x0000 0x000b of the port downlink protocol packet oriented port 0;
  • the destination address (DA) Is 0x8000 0x0000 0x0000 0x000c port downlink protocol packet is directed to port 0;
  • ie destination address (DA) is 0x8000 0x0000 0x0000 OxOOOd port downstream protocol packet oriented port 0;
  • ie destination address (DA) is 0x8000 0x0000 0x0000 0x000e port downstream protocol packet oriented port 0;
  • ie destination address (DA) is 0x8000 0x0000 0x0000 0x000f port downstream protocol packet oriented port 0;
  • the MSS-400 server will allocate the packet through the port including the port allocation information, and inform the BX-008-0 switch to do the following configuration in its table No. 0:
  • ie destination address (DA) is 0x8000 0x0000 0x0000 0x000a port downstream protocol packet is directed to port 1;
  • ie destination address (DA) is 0x8000 0x0000 0x0000 0x000b port downstream protocol packet oriented port 2;
  • ie destination address (DA) is 0x8000 0x0000 0x0000 OxOOOd port downstream protocol packet is directed to port 4;
  • the destination address (DA) is 0x8000 0x0000 0x0000 0x000e of the port downlink protocol packet to port 5;
  • ie destination address (DA) is 0x8000 0x0000 0x0000 0x0010 port downstream protocol packet is directed to port 7;
  • MSS-400 server destination address is 0x8000 0x0000 0x0000 0x0009, 0x8000 0x0000 0x0000 0x000a, 0x8000 0x0000 0x0000 0x000b, 0x8000 0x0000 0x0000 0x000c, 0x8000 0x0000 0x0000 OxOOOd, 0x8000 0x0000 0x0000 OxOOOe, 0x8000 0x0000 0x0000 OxOOOf, 0x8000 0x0000 0x0000 0x0010
  • SA is 0x0000 0x0000 0x0000 0x0000
  • the port downlink protocol packet will be directed to the MSS-400 server port No. 0.
  • the port downlink protocol packet will be directed to the BX in turn.
  • S ll, BX-008-1 switch receives port downlink protocol packet from port 1 of BX-008-0 switch (that is, port downlink protocol with destination address 0x8000 0x0000 0x0000 0x000a)
  • the port uplink protocol packet (the port uplink protocol packet contains the registration information of the switch)
  • the DA of the packet is 0x0800 0x0000 0x0000 0x0000
  • the SA is 0x0000 0x0000 0x0000 0x0000 0x000a;
  • the MSS-400 server After receiving the port uplink protocol packet sent by the BX-008-1 switch, the MSS-400 server compares the source address (SA) of the uplink protocol packet with the device type to know that port 1 of BX-008-0 is connected. The station accesses the switch, and then finds the connection switch information in the registration information table inside the server, and sends an incoming network command to the access switch (informing its access network address is 0x000a);
  • the BX-008-1 switch After receiving the network access command, the BX-008-1 switch knows that its access network address is 0x000a and enters the network. Configure its 0th table "00 0000 0000 0000 1010" as "01 0000 0000", the rest of the 0th table. The entry is configured as "00 0000 0000", that is, only the downlink protocol packet of the switch is imported into the CPU, and the network access command response is sent to the server;
  • the MSS-400 server After receiving the network access command sent by the switch, the MSS-400 server knows that the BX-008-1 switch has entered the network, and then sends a device status query command to the port every second to check whether the BX-008-1 switch works normally. At the same time, send the port downlink protocol packet to the downlink port of the BX-008-1 switch to check whether other access network devices are connected to the access switch. If the current access switch works normally, it will send a status query response to the server after receiving the device status query command. When the server does not receive the status query response within 6 seconds, it considers that the access switch has been removed from the network, no longer sends the device status query command, and continues to send the query packet to the port. 3.1.3 Terminal Access Process:
  • each terminal that is allowed to access the network must be registered in the node server, and the terminal that is not registered cannot access the network.
  • the process of the terminal accessing the network involves the following steps:
  • the node server sends a query packet to each port, and the terminal sends a response packet after receiving the query packet, where the response packet includes the registration information of the terminal;
  • the node server After receiving the response packet sent by the terminal, the node server knows which port is connected to which terminal (set top box, code board or memory), and then finds the information of the terminal in the registration information table inside the node server, to the terminal. Sending a network access command (informing the access network address of the terminal), the terminal enters the network after receiving the network access command, and sends a network access command response to the node server;
  • the node server After receiving the network access command sent by the terminal, the node server knows that the terminal has entered the network, and periodically sends a status inquiry packet to the port to check whether the terminal works normally. If the terminal works normally, it will send a status query response to the node server after receiving the status inquiry packet. When the node server does not receive the status query response within a certain period of time, it recognizes that the terminal has been removed from the network, does not send the status query packet, and continues to send the query packet to the port.
  • the address of the access network can be set to 16 bits, and all access network devices have unique access network addresses (including set-top boxes, access switches, Memory, even the node server itself).
  • an address information table can be maintained in the CPU module of the node server.
  • the size of the table is 2 to the 16th power, that is, 64K, and the entries of each table. It consists of the following:
  • Address occupation descriptor "00" indicates that the address is not used, "01” indicates that the address is inactive (the node server sends the port downlink protocol packet with this address, but does not receive the incoming line protocol packet), "10” " Indicates that this address is used (set by the node server after receiving the incoming line protocol package);
  • device resource description information for example, if the device is an access switch, the access network address of the device connected to the network port, and the uplink and downlink traffic of each network port; if the device is a memory, its The access network address of the device connected to the network port, the count of its read and write channels, and the upstream and downstream traffic counts of the network port; etc., all of this information is used to provide decision-making basis for the service process, and will be modified in each service flow. these messages.
  • Port 0 of BX-008-0 is connected to a set-top box STB-0
  • port 1 of BX-008-1 is connected to a set-top box STB-1.
  • MSS-400 server After the MSS-400 server is powered on, initialize the hardware and obtain the default metropolitan area network address (assumed to be 0x00 0x0000 0x0000). Import the configuration file from the hard disk to the CPU memory (such as the registration information of the switch, the registration information of the terminal, etc.), MSS. -400 server initialization address information table, all cleared (indicating that all addresses are not used), MSS-400 server configures its own access network address to 0x0000, that is, the 0x0000 item of the address information table is configured as follows:
  • This node server has eight downstream 100M network ports defined as port 0 to port 7, one CPU module interface is defined as port 8, and one disk array interface is defined as port 9. 1 uplink Gigabit optical port is defined as port 10, the node server model is MSS-400, the access network address of the device connected to its network port is not allocated, and the uplink and downlink traffic count of each network port is 0;
  • the next available address in the address information table is 0x0001;
  • MSS-400 server initialization 0, 1, 2, 3 table:
  • the next available address of the address information table is 0x0009;
  • BX-008-0, BX-008-1 switch initializes the hardware after power-on
  • the BX-008-0 switch After receiving the query packet, the BX-008-0 switch receives the query packet to its CPU module according to the configuration of the No. 0 table, and the CPU module parses the query packet and generates a response packet.
  • the response contains the registration information of the access switch) and sends it to the MSS-400 server.
  • the DA of the packet is 0x0800 0x0000 0x0000 0x0000, SA 0x0000 0x0000 0x0000 0x0001;
  • the MSS-400 server After receiving the response packet sent by the BX-008-0 switch, the MSS-400 server compares the source address (SA) of the response packet with the device type to know that the port 0 is connected to an access switch, and then The information of the access switch is found in the registration information table of the node server, and the network access command is sent to the access switch (informing that the access network address is 0x0001);
  • the BX-008-0 switch After receiving the network access command, the BX-008-0 switch knows that its access network address is 0x0001 and enters the network. Therefore, its 0th table "00 0000 0000 0001" is set to "01 0000 0000", table 0 The other entries are configured as "00 0000 0000", that is, only the downlink protocol packet of the switch is imported into the CPU, and the network access command response is sent to the server;
  • the MSS-400 server receives the incoming network command response from the BX-008-0 switch. It knows that the BX-008-0 switch has entered the network, so the 0x0001 item in the address information table inside the server is configured as:
  • This access switch has eight downstream 100M network ports defined in turn. For port 0 to port 7, one CPU module interface is defined as port 8, and one uplink 100 Mbps port is defined as port 9, the access switch model is BX-008, and its upstream network port is connected.
  • the access network address of the device is 0x0000 (ie MSS-400), the access network address of the device connected to the downlink network port is not allocated, and the uplink and downlink traffic counts of each network port are 0;
  • the MSS-400 server will do the following configuration in its table 0:
  • the destination address (DA) is 0x8000 0x0000 0x0000 0x000a of the port downlink protocol packet oriented port 0;
  • the destination address (DA) is 0x8000 0x0000 0x0000 0x000b of the port downlink protocol packet oriented port 0;
  • ie destination address (DA) is 0x8000 0x0000 0x0000 0x000c port downstream protocol packet oriented port 0;
  • ie destination address (DA) is 0x8000 0x0000 0x0000 0x000d port downstream protocol packet oriented port 0;
  • ie destination address (DA) is 0x8000 0x0000 0x0000 0x000e port downstream protocol packet oriented port 0;
  • ie destination address (DA) is 0x8000 0x0000 0x0000 0x000f port downstream protocol packet oriented port 0;
  • the MSS-400 server will distribute the packet through the port that includes the port assignment information.
  • the BX-008-0 switch has the following configuration in its table No. 0:
  • ie destination address (DA) is sufficient 0x8000 0x0000 0x0000 0x0009 port downstream protocol packet oriented port 0;
  • ie destination address (DA) is sufficient 0x8000 0x0000 0x0000 0x000a port downstream protocol packet is directed to port 1;
  • ie destination address (DA) is sufficient 0x8000 0x0000 0x0000 0x000f port downstream protocol packet is directed to port 6;
  • the port downlink protocol packet will be sequentially directed to the MSS-400 server port 0, according to the No. 0 table configuration of the BX-008-0 switch, the port downlink protocol The packet will be directed to ports 0 to 7 of the BX-008-0 switch in turn; and, items 0x0009 to 0x0010 of the address information table in the MSS-400 server are configured to:
  • the next available address is 0x0011;
  • S ll, STB-0 receives the port downlink protocol packet from the port 0 of the BX-008-0 switch (that is, the port downlink protocol packet whose destination address is 0x8000 0x0000 0x0000 0x0009), and sends the port uplink protocol packet after receiving the port downlink protocol packet.
  • the DA of the packet is 0x0800 0x0000 0x0000 0x0000, SA 0x0000 0x0000 0x0000 0x0009 (port 0 of the switch);
  • the MSS-400 server After receiving the port uplink protocol packet sent by the STB-0 switch, the MSS-400 server knows that the source address (SA) of the uplink protocol packet and the device type know that the port No. 0 of the BX-008-0 is connected to a terminal. Then, the terminal information is found in the registration information table in the server, and the network access command is sent to the terminal (the terminal's access network address is 0x0009);
  • STB-0 After receiving the network access command, STB-0 knows that its access network address is 0x0009 and enters the network, and sends a network access command response to the server.
  • the MSS-400 server receives the incoming network command response sent by STB-0 and knows that the STB-0 switch has entered the network. Therefore, the 0x0009 item in the address information table is configured as:
  • This terminal has an audio and video codec engine, 1 100M network port, the terminal model is STB, and the access network address of the device connected to its network port is 0x0001 (ie BX-008-0) , the uplink and downlink traffic count of its network port is 0;
  • the 0x0001 item of the address information table is configured to:
  • the access switch has eight downlink 100M network ports defined as port 0 to port 7, and one CPU module interface is defined as port 8, and one uplink 100 Mbps port is defined as 9 No. port, the access switch model is BX-008, the access network address of the device connected to the uplink network port is 0x0000 (ie MSS-400), and the access network address of the device connected to the downlink network port 0 is 0x0009.
  • the rest are unassigned, and the uplink and downlink traffic counts of each of its network ports are 0; Afterwards, the MSS-400 server sends a device status query command to this port every second to check whether STB-0 is working normally. When the server does not receive the status query response within 6 seconds, it thinks that STB-0 has been removed from the network. Send the device status query command and continue to send the query packet to the port.
  • BX-008-1 will also enter the network, and the access network address is 0x0002; STB-1 will also enter the network, and the access network address is 0x0012.
  • the PDU is the information exchange mode between the user terminal and the server.
  • the two use the original socket to transmit the PDU.
  • the data format is as follows:
  • the switch receives the response sent by the 8A01 to the server.
  • the code board receives the response sent by the 8A01 to the server.
  • the types of data packets defined in the metropolitan area network are four, which are: the sub-city query tag package (the labeled protocol package sent by the metropolitan area server to the node switch and the node server);
  • the metropolitan area response tag packet (the tagged protocol packet that is sent by the node switch and the node server to the metropolitan area server);
  • the address of the metropolitan area network is 40bit in total. In this paper, it is divided into three layers, which are 8bit, 16bit, and 16bit, and are defined as national network, wide area network, and metropolitan area network. Data transmission between terminals of the same metropolitan area network and the same access network is controlled by the node server of the access network.
  • the terminal STB_0 is located in the access network A
  • the terminal STB_1 is located in the access network B
  • the access network A and the access network B belong to the same metropolitan area network C.
  • the implementation process is as follows:
  • STB_0 makes a request and STB_1 makes a video call;
  • the node server of the access network A finds that it belongs to the access network according to the number of the STB_1.
  • the metropolitan area network C's metropolitan area server finds that it belongs to the access network according to the number of STB_1.
  • the node server of the access network B finds that the STB_1 is in the access network B according to the number of the STB_1, and the node server of the access network B sends a call menu to the STB_1;
  • STB_1 can choose to accept or reject, and the response is sent to the node server of access network B, which is assumed to be accepted;
  • the node server of the access network B sends a response to the metropolitan area server of the metropolitan area network C;
  • the metropolitan area network server of the metropolitan area network C sends a response to the node server of the access network A;
  • the node server of access network A sends a response to STB_0.
  • the terminal only interacts with the node server of the access network, the node server interacts with the metropolitan area server of the metropolitan area network, and so on, the metropolitan area server and the local wide area network Wide-area server interaction.
  • the datagram type of the metro query packet is "1001 0000" (binary), which is 0x90 (hexadecimal).
  • the datagram type of the metropolitan area response packet is "00001001" (binary), which is 0x09 (16).
  • Binary the datagram type of the unicast tag packet is "0001 0000” (binary), which is 0x10 (hexadecimal), and the datagram type of the multicast tag packet is "0111 1000" (binary). That is 0x78 (hexadecimal), which requires 4 lookup tables, for example
  • Tag lookup table for unicast tag packets defined as table 6 with a size of 64K;
  • the tag lookup table of the multicast tag data packet defined as the table No. 7, with a size of 64K; the output of the metro domain query tag packet, the metropolitan area response tag packet, the unicast tag data packet, and the tag lookup table of the multicast tag data packet
  • there is also a 16-bit outgoing tag for example, one of the node switches, the MX-4, has four Gigabit optical ports and one CPU module interface. If 4 Gigabit optical ports are defined as ports 0 to 3, and the CPU module interface is defined as port 4, a total of 64k X 21bit (5bit + 16bit) metro query tag packet address lookup table, 64k X 21bit is required.
  • the output of the metropolitan area query tag packet lookup table with the entry tag of 0x0001 is "1 0000 0000 00000000", indicating that the packet is directed to port 4 (CPU port), the outbound tag is 0x0000, and the inbound tag is 0x0001.
  • the output of the packet lookup table is "0 0011 0000 0011 0000 0000", indicating that the packet is directed to port 0, port 1, the out tag is 0x0300, and so on.
  • port 0 enters a packet. Its header data is 0x1056 0x1500 0x0000 0x55aa 0x0056 0x1500 0001 0xaa55 0x0000 0x00001, where DA is 0x1056 0x1500 0x0000 0x55aa, SA is 0x0056 0x1500 0001 0xaa55 Reserved byte is 0x0000, label is 0x0001, Then its packet type is 0x10, and then check the table rule at this time to check the No.
  • the metropolitan area server sends a query packet to each port, and the node switch sends a response packet after receiving the query packet (the response includes the device type and device identifier of the switch, which is the inherent information of each switch);
  • the metropolitan area server After receiving the response from the node switch, the metropolitan area server knows that the node is connected to a node switch, and then finds the node switch information in the registration information table inside the metropolitan area server, and sends a network access command to the node switch. (telling the metropolitan area network address and label of the switch), the node switch enters the network after receiving the network access command, and sends a network access command response to the metropolitan area server;
  • the metropolitan area server After receiving the network access command response from the switch, the metropolitan area server knows that the node switch has entered the network, and then sends a status inquiry packet to the port every second to check whether the node switch works normally, and also to the The other ports of the node switch send the port query packet to check whether other devices are connected under the node switch. If the node switch works normally, a status query response is sent to the metropolitan area server after receiving the status query packet. When the metropolitan area server does not receive the status query response within a certain period of time (such as within 6 seconds), it considers that the node switch has been removed from the network, no longer sends the status query packet, and continues to send the query packet to the port.
  • a certain period of time such as within 6 seconds
  • the device information table All devices in the metropolitan area network are described by the device information table. Two device types with a te and six device IDs can be used to uniquely identify a device. The same device type has different device information tables to describe, such as a node switch information table and a node server information table.
  • the entries of the device information table are composed as follows:
  • Device identification 6 bytes. When the device is registered, it is written to the hard disk or flash of the metropolitan area server. After the metropolitan area server is powered on, it is imported into the CPU.
  • Device address 2 bytes, the metropolitan area network address assigned by the device.
  • the address of the metropolitan area network is 16bit. All the devices in the metropolitan area network have unique metropolitan area network addresses (including metropolitan area servers, node switches, and node servers).
  • the CPU module of the metropolitan area server maintains a 16 times of 2 times.
  • the table of the square that is, the table of 64K, is called the metropolitan area address information table, and the entries of each table are composed as follows:
  • Address occupation descriptor 2 bytes, 0x0000 means that this address is not used, 0x0001 means this address is inactive (the metropolitan area server sends the network access command packet with this address, but does not receive the network access command response), 0x0002 means this address Used (the metropolitan area server is set after receiving the incoming response packet);
  • Device type 2 bytes, for example 0x0000 means the metropolitan area server MS-1000, 0x0001 means one of the node switches MX-4, 0x0002 represents one of the node servers MSS-400;
  • Device resource description information a number of bytes, for example, the device is a node switch, and includes a metropolitan area network address of a device connected to its network port, and an uplink and downlink traffic count of each network port; if the device is a node server, Then it includes the access network address of the device connected to its network port, the uplink and downlink traffic count of its network port, and so on; all of this information is used to provide decision-making basis for the service process, and this information is modified in each service flow. .
  • the metropolitan area query tag describes the connection from the metropolitan area server to the node switch or node server
  • the metropolitan area response label describes the connection from the node switch or node server to the metropolitan area server.
  • the CPU module of the metropolitan area server maintains another 2nd 16th power table, that is, a 64K table, which is called a metropolitan area protocol label information table, and the entries of each table are composed as follows:
  • Tag occupation descriptor 2 bytes, 0x0000 means that the tag is not used, 0x0001 means the tag is inactive (the metro server sends the port query packet with this tag, but does not receive the incoming response packet), 0x0002 indicates the tag Used (the metropolitan area server is set after receiving the incoming response packet);
  • Tag descriptor 2 bytes, the metropolitan area network address of the device corresponding to the tag;
  • Label routing description information 4 bytes, used to describe the metropolitan area network address and port number of the previous hop switch of the metropolitan area query label packet. The first 2 bytes represent the metropolitan area network address of the previous hop switch, after Two bytes represent the port number of the previous hop switch.
  • the metro server has 4 Gigabit optical ports and 1 CPU module interface. If the four Gigabit optical ports are defined as ports 0 to 3, the CPU module interface is defined as port 4, the metropolitan area server model is MS-1000, and the MS-1000 port 0 and port 1 are respectively connected. One MX-4-0 port 2 and port 3, MX-4-0 port 0 is connected to one MSS-400-0, and port 1 is connected to one MSS-400-l.
  • the network interaction process is as follows:
  • the hardware is initialized, and the configuration file is imported from the hard disk to the CPU memory (for example, registration information of the node switch, registration information of the node server, etc.), and the MS-1000 server initializes the metropolitan area address information table and the metropolitan area.
  • the protocol label information table all cleared (indicating that all addresses and labels are unused), the MS-1000 server configures its own metropolitan area network address to be 0x0000, that is, the 0x0000 item of the metropolitan area address information table is configured as follows:
  • This metro server has 4 Gigabit optical ports defined as ports 0 to 3, and the CPU module interface is defined as port 4, this node server model is MS-1000, its network port The MAN address of the connected device is not allocated, and the uplink and downlink traffic of each network port is counted as 0; The next available address is 0x0001, the next city i or protocol label is 0x0000;
  • Configuration table No. 4 is "0 0000 0000 0000 0000" to " 0 0000 1111 1111 1111", that is, all metropolitan area query tag packet transmissions are closed;
  • the MS-1000 server configuration knows that it has 4 Gigabit optical ports, and the next Metropolitan Area Protocol label is 0x0000, so it configures four Table 4 entries as
  • the next metropolitan protocol label is 0x0004;
  • MS-100G server sends header information to 0x9000 0x0000 0x0000 0x0000 0x0000 0x0000 0x0000 0x0000 0x0000 0x0000, 0x9000 0x0000 0x0000 0x0000 0x0000 0x0000 0x0000 0x0000 0x0000 0x0000 0x0000 0x0000 0x0000 0x0000 0x0000 0x0000 0x0000 0x0000 0x0000 0x0000 0x0000 0x0000 0x0000 0x0000 0x0000 0x0000 0x0000 0x0002 0x0000 0x0000 0x0000 0x0000 0x0000 0x0000 0x0000 0x0000 0x0000 0x0000 0x0000 0x0003
  • Port inquiry packet since the packet is exchanged according to the label, it does not matter even if the DA is the same. According to the No. 4 table, the
  • the 0x0000 item of the tag information table is configured as follows:
  • Tag Occupation Descriptor 0x0001 indicates that this tag is inactive; • Tag Descriptor: No modification;
  • the 0x0001 item of the tag information table is configured as follows:
  • Tag Occupation Descriptor 0x0001 indicates that this tag is inactive
  • the 0x0002 item of the tag information table is configured as follows:
  • Tag Occupation Descriptor 0x0001 indicates that this tag is inactive
  • Item 0x0003 of the tag information table is configured as follows:
  • Tag Occupation Descriptor 0x0001 indicates that this tag is inactive
  • the next available tag is 0x0004;
  • Configuration No. 5 is "0 0000 0000 0000 0000" to " 0 0000 1111 1111 1111", that is, all metropolitan area response label packet transmissions are closed;
  • topology map MX-4-G switch port 3 receives the query packet of the metropolitan area protocol label 0x0001, then
  • Send 2 response packets (the response contains the device type of the switch, the device ID, and the port number of the received query packet).
  • the header of a packet is 0x0900 0x0000 0x0000 0x0000 0x0000 0x0000 0x0000 0x0000 0x0000 0x0000 0x0000 0x0000 0x0000 0x0000 0x0000 0x0000 0x0000 , in the packet Indicate that the port number of the query packet received is 2;
  • the node switch compares the device identifier in the response packet with the device identification item in the device information table of the node switch, until the same is the same to indicate that the device is registered, and the device status item is checked. 0x0000, know that this device is not connected to the network;
  • the MS-1000 server According to the 5th field number in the response packet PDU (indicating the switch port number of the received query packet), the MS-1000 server knows that port 0 is connected to port 2 of a switch.
  • the packet header is 0x9000 0x0000 0x0001 0x0000 0x0000 0x0000 0x0000 0x0000 0x0000 0x0000 0x0000;
  • the 0x0001 of the address information table is configured as follows: • Address occupation descriptor: 0x0001 indicates that this address is inactive (the metropolitan area server sends the network access command packet with this address, but does not receive the network access command response);
  • the entries of the corresponding device information table are configured as follows:
  • Device status: 0x0001 indicates that the device is waiting for access to the network (the metropolitan area server sent the network access command packet, but did not receive the network access command response);
  • the node switch According to the device type in the response packet, it is known that the node switch compares the device identifier in the response packet with the device identification item in the device information table of the node switch, until the same is the same to indicate that the device is registered, and the device status item is checked. 0x0001, know that this device is waiting to enter the network;
  • the last hop switch is the MS-1000 server (address 0x0000) and the port number is 1;
  • the MS-1000 server According to the 5th field number in the response packet PDU (indicating the switch port number of the received query packet), the MS-1000 server knows that port 1 is connected to port 3 of a switch.
  • the packet header is 0x90000x0000 0x0001 0x00000x0000 0x0000 0x00000x0000 0x0000 0x0000 0x0001;
  • the 0x0001 of the address information table is configured as follows:
  • Address occupation descriptor: 0x0001 indicates that this address is inactive (the metropolitan area server sends the network access command packet with this address, but does not receive the network access command response);
  • the entries of the corresponding device information table are configured as follows: • Equipment identification: no modification;
  • Device status: 0x0001 indicates that the device is waiting for access to the network (the metropolitan area server sent the network access command packet, but did not receive the network access command response);
  • MX-4-0 switch No. 2 port receives the access protocol packet of the metropolitan area protocol label 0x0000 (the head of the packet is 0x9000 0x0000 0x0001 0x0000 0x0000 0x0000 0x0000 0x0000 0x0000 0x0000 0x0000 0x0000), after comparing the device type and device Identifies that I know that my metropolitan area network address is 0x0001 and I enter the network. At the same time, I send a network access command response to the server.
  • the headers of the packets are 0x0900 0x0000 0x0000 0x0000 0x0000 0x0000 0x0000 0x0000 0x0001 0x0000 0x0000 0x0000 0x0000;
  • MX-4-0 switch port 3 receives the network protocol packet 0x0001 of the metropolitan area protocol label (the head of the packet is 0x9000 0x0000 0x0001 0x0000 0x0000 0x0000 0x0000 0x0000 0x0000 0x0000 0x0000 0x0001), after comparing the device type, The device ID, knowing that its MAN address is 0x0001, enters the network, and sends a network access command response to the server.
  • the headers of the packets are 0x0900 0x0000 0x0000 0x0000 0x0000 0x0000 0x0000 0x0001 0x0000 0x0000 0x0000 0x0001;
  • S 1 MS-1000 server No. 0 port receives the metropolitan area protocol label 0x0000 network access command response packet
  • MS-1000 server According to the network access command response packet PDU field number 9 (labeling the switch port number of the received query packet) MS-1000 server knows that port 0 is connected to port 2 of a switch;
  • the metropolitan area address of the switch is 0x0001.
  • the 0x0001 of the address information table is configured as follows:
  • Device resource description information 4 Gigabit optical ports are defined as ports 0 to 3, and the CPU module interface is defined as port 4; its port 2 is connected to a MS-1000 with a metro address of 0x0000. On port 0, the metropolitan area network address of the device connected to other network ports is unknown, and the uplink and downlink traffic counts of each of its network ports are 0.
  • the 0x0000 item of the tag information table is configured as follows:
  • Tag Occupation Descriptor 0x0002 indicates that this tag is used
  • the entries of the corresponding device information table are configured as follows:
  • Device status: 0x0002 indicates that the device has entered the network (the metropolitan area server has sent the network access command packet and received the network access command response);
  • the 0x0000 item of the metro address information table is configured as follows:
  • This metro server has 4 Gigabit optical ports defined as ports 0 to 3, and the CPU module interface is defined as port 4, this node server model is MS-1000, its 0 The port is connected to port MX-4-0 of the metropolitan area address 0x0001. The metropolitan area network address of the device connected to the other network port is not allocated, and the uplink and downlink traffic count of each network port is 0.
  • the device status query command will be sent to port 0 periodically (for example, every second). If the MS-1000 server does not receive the status query response within a certain period of time (such as 6 seconds), the device status query command will not be sent, and the port will continue to be sent to the port. 0 sends the query packet.
  • the port 1 of the MS-1000 server After receiving the network access command response packet of the metropolitan area protocol label 0x0001, the port 1 of the MS-1000 server receives the network access command response packet. 1) According to the metropolitan area protocol label 0x0001 in the network access command response packet, check the label routing description information of the 0x0001 item of the label information table to know that the switch knows that the last hop switch is the MS-1000 server (address is 0x0000), the port number Is number 1;
  • the MS-1000 server According to the ninth field number of the incoming packet PDU (indicating the switch port number of the received query packet), the MS-1000 server knows that port 1 is connected to port 3 of a switch.
  • the metropolitan area address of the switch is 0x0001.
  • the 0x0001 of the address information table is configured as follows:
  • Device resource description information 4 Gigabit optical ports are defined as ports 0 to 3, and the CPU module interface is defined as port 4; its port 2 is connected to a MS-1000 with a metro address of 0x0000. Port 0, its port 3 is connected to port 1 of MS-1000 with metro address 0x0000. The address of the metropolitan area network of the device connected to other network ports is unknown, and the uplink and downlink traffic of each network port is counted. Is 0.
  • the 0x0001 item of the tag information table is configured as follows:
  • Tag Occupation Descriptor 0x0002 indicates that this tag is used
  • the entries of the corresponding device information table are configured as follows:
  • Device status: 0x0002 indicates that the device has entered the network (the metropolitan area server has sent the network access command packet and received the network access command response);
  • the 0x0000 item of the metro address information table is configured as follows:
  • This metro server has 4 Gigabit optical ports defined as ports 0 to 3, and the CPU module interface is defined as port 4, this node server model is MS-1000, its 0 The port is connected to the MX-4-0 port 2 of the metropolitan area address 0x0001, and the port 1 is connected to the MX-4-0 port 3 of the metropolitan area address 0x0001, and the devices connected to other network ports The metropolitan area network address is not allocated, and the uplink and downlink traffic counts of each of its network ports are 0;
  • the device status query command will be sent to port 1 at regular intervals (for example, every second). If the MS-1000 server does not receive the status query response within a certain period of time (such as 6 seconds), it will not send the device status query command and continue to the port. 1 Send the query package.
  • MS-1000 server knows that port 0 is connected to port 2 of MX-4-0 with metro address 0x0001, port 1 is connected to port 3 of MX-4-0, MX-4-0 Ports 0 and 1 are unknown, and the next metropolitan protocol label is 0x0004. So it configures the four Table 4 entries as
  • the next metropolitan area protocol label is 0x0008.
  • the MS-1000 notifies the MX-4-0 by using the label 0x0000 or 0x0001, and configures the entries of the MX-4-0 table 4:
  • MS-1000 server sends header information to 0x9000 0x0000 0x0000 0x0000 0x0000 0x0000 0x0000 0x0000 0x0000 0x0000 0x0004, 0x9000 0x0000
  • Item 0x0004 of the tag information table is configured as follows:
  • Tag Occupation Descriptor 0x0001 indicates that this tag is inactive
  • Label route description information 0x0001 (the metropolitan area network address of the last-hop switch is the metropolitan area network address of MX-4-0), 0x0000 (port 0 of MX-4-0).
  • Item 0x0005 of the tag information table is configured as follows:
  • Tag Occupation Descriptor 0x0001 indicates that this tag is inactive; • Tag Descriptor: No modification;
  • Label route description information 0x0001 (the metropolitan area network address of the last-hop switch is the metropolitan area network address of MX-4-0), 0x0001 (port 1 of MX-4-0).
  • Item 0x0006 of the tag information table is configured as follows:
  • Tag Occupation Descriptor 0x0001 indicates that this tag is inactive
  • Label route description information 0x0001 (the metropolitan area network address of the last-hop switch is the metropolitan area network address of MX-4-0), 0x0000 (port 0 of MX-4-0).
  • Item 0x0007 of the tag information table is configured as follows:
  • Tag Occupation Descriptor 0x0001 indicates that this tag is inactive
  • Label route description information 0x0001 (the metropolitan area network address of the previous hop switch is the metropolitan area network address of MX-4-0), 0x0001 (port 1 of MS-1000).
  • the next available tag is 0x0008;
  • the switch initializes the hardware after power-on. Since the node server is the initiator or the terminating end of the tag, there is no need to replace the tag;
  • the MSS-400-0 switch receives the query packet of the metropolitan area protocol label 0x0004 and 0x0006 on the port 10, then:
  • Send 2 response packets (the response contains the device type of the switch, the device ID, and the port number of the received query packet).
  • the header of a packet is 0x0900 0x0000 0x0000 0x0000 0x0000 0x0000 0x0000 0x0000 0x0000 0x0000 0x0004 , in the packet Indicate that the port number of the query packet received is 10;
  • the header of another packet 0x0900 0x0000 0x0000 0x0000 0x0000 0x0000 0x0000 0x0000 0x0000 0x0000 0x0000 0x0000 0x0006, and the port number of the received query packet is marked as 10 in the packet;
  • the MSS-400-1 switch port 10 receives the query packet of the metropolitan area protocol labels 0x0005 and 0x0007, then:
  • Send 2 response packets (the response contains the device type of the switch, the device ID, and the port number of the received query packet).
  • the header of a packet is 0x0900 0x0000 0x0000 0x0000 0x0000 0x0000 0x0000 0x0000 0x0000 0x0000 0x0000 0x0005, in the packet Indicate that the port number of the query packet received is 10;
  • this metro server has 4 Gigabit optical ports and 1 CPU module interface. If the four Gigabit optical ports are defined as ports 0 to 3, the CPU module interface is defined as port 4, the metropolitan area server model is MS-1000, and the MS-1000 port 0 and port 1 are respectively connected. One MX-4-0 port 2 and port 3, MX-4-0 port 0 is connected to one MSS-400-0, and port 1 is connected to one MSS-400-l.
  • Terminal STB-G is connected to port 0 of MSS-400-G
  • terminal STB_ 1 is connected to port 1 of MSS-400-1
  • their address after access is STB_ 0 ( 0x0000 0x0000 0x0002 x0009 ), STB-1 ( 0x0000 0x0000 0x0003 0x0012)
  • the metropolitan area network address of MX-4-0 is 0x0001
  • STB_ 0 sends an application to MSS-400-0 for visual communication with STB_1.
  • the steps are as follows: 51, STB-0 issued a service request instruction packet, the package DA 0x0800 0x0000 0x0002 0x0000, SA 0x0000 0x0000 0x0002 0x0009, reserved 0x0000, PDU part see Appendix, the service parameter is SERVICE_TYPE_TELEPHONE_REQUEST or SERVICE_TYPE_TELEPH0NE_DIRECT o
  • the service application instruction package is directed to MSS-400-0 according to the configuration of Table 1, MSS-400-0 judges the application for receiving visual communication according to the contents of the package; according to SA, it is STB_0, (assuming the bandwidth of the videophone) It is 6Mbit/s for uplink and downlink, and 80Mbit/s for uplink and downlink bandwidth of STB_0 and MSS-400-0. Check that the uplink and downlink bandwidths of STB_0 and MSS-400-0 are found to meet the service requirements, and then continue to jump to S4. Otherwise, Jump to S3.
  • MSS-400-0 sends a menu to the calling party STB_0, indicating that the service is rejected; to the STB_0: packet DA 0x80000x00000x0002 0x0009, SA 0x0000 0x0000 0x0002 0x0000, reserved 0x0000, PDU section see Appendix menu data format.
  • the MSS-400-0 knows that the called party is not in its own access network according to the called number CAM table (content-address mapping table), so the MSS-400-0 sends a service request instruction packet to the metropolitan area server MSS-1000.
  • Package DA 0x0900 0x0000 0x0000 0x0000, SA 0x00000x00000x00020x0009, reserved 0x0000, cut, negotiated label is 0x0005
  • the service parameter is SERVICE_TYPE_TELEPHONE_REQUEST or SERVICE_TYPE_TELEPH0NE_DIRECT o
  • MSS-1000 receives the MSS-400-0 service application package and judges the application for receiving visual communication according to the content of the package; according to SA, it is the terminal under MSS-400-0, (assuming MX-4-0 And the MSS-400-0 has 800 Mbit/s left and right bandwidth;) According to the called number, the CAM table (content-address mapping table) knows that the called party is under the access network of MSS-400-1 (false
  • MX-4-0 and MSS-400-1 have 800 Mbit/s left and right bandwidth. Check the upstream and downstream bandwidths of MX-4-0, MSS-400-0, and MSS-400-1 to meet the service requirements. Continue to jump to S7, otherwise jump to S6.
  • MSS-1000 sends a service rejection packet to MSS-400-0.
  • Package DA 0x9000 0x00000x00020x0009
  • SA 0x00000x00000x00000x0000
  • reserved 0x0000 the protocol label is 0x0005
  • the PDU part is ignored
  • the MSS-400-0 receives the service rejection packet and jumps to S3.
  • the MSS-1000 sends a service request packet to the MSS-400-1, and issues a service request instruction packet.
  • the data format definition of the network access process) the service parameter is SERVICE_TYPE_TELEPHONE_REQUEST or SERVICE_TYPE_TELEPH0NE_DIRECT o
  • MSS-400-1 receives the service application package of MSS-1000, judges the application for receiving visual communication according to the content of the package; and checks the CAM table (content-address mapping table) according to the called number to know that the called party is STB_1 (Assume that the uplink and downlink bandwidths of STB_1 and MSS-400-1 are 80 Mbit/s left.) Check that the uplink and downlink bandwidths of STB_1 and MSS-400-1 meet the service requirements, and then continue to jump to S10, otherwise skip to S9.
  • CAM table content-address mapping table
  • MSS-400-1 sends a menu called, respectively, waiting for the called party to answer;
  • STB-1 receives the menu, issues the application SERVICE_TYPE_PERMISSION to accept the communication, the package DA 0x0800 0x0000 0x0003 0x0000, SA 0x0000 0x0000 0x0003 0x0012, reserved 0x0000, (the PDU part see the data format definition of the city i or network access network process), service parameters Is SERVICE_TYPE_PERMISSI0N o
  • MSS-400-1 receives the response packet of STB-l, the sending service allows packet to MSS-1000, the packet DA 0x90000x00000x00000x0000, SA 0x00000x0000 0x0003 0x0000, reserved 0x0000, the protocol label is 0x0006, the PDU part is ignored.
  • the MSS-1000 When the MSS-1000 receives the service permission packet, it assigns a unicast label (assuming the MSS-400-0 to MSS-400-1 inbound label, the outgoing label is 0x0000, MSS-400-1 to MSS-400-0) Inbound label, outgoing label is 0x0001); The MSS-1000 sends a label distribution package to the MX-4-0.
  • the MSS-1000 sends a label distribution package to the MSS-400-0.
  • the package DA 0x9000 0x0000 0x0002 0x0000, SA 0x0000 0x0000 0x0000, reserved 0x0000, the protocol label is 0x0005, and the PDU part contains the inbound label, the outgoing label and the navigation port. Contains bindings for DAs, SAs, and tags;
  • the MSS-1000 sends a label distribution package to the MSS-400-1.
  • the package DA 0x9000 0x0000 0x0003 0x0000, SA 0x0000 0x0000 0x0000, reserved 0x0000, the protocol label is 0x0006, and the PDU part contains the inbound label, the outgoing label, and the navigation port. Contains bindings for DAs, SAs, and tags;
  • MX-4-0 receives the label allocation package to update its No. 6 table
  • the 0x0000 item is: the outgoing label is 0x0000
  • the guiding port is port 1
  • the 0x0001 item is: the outgoing label is 0x0001, the guiding port is 0 port.
  • MSS-400-0 receives the label allocation package to update its DA, SA and label binding CAM table (address-tag binding table), that is, the 0x0000 item of the CAM table is: DA 0x1000 0x0000 0x0003 0x0012 SA 0x0000 0x0000 0x0002 0x0009;
  • the 0x0000 items are: the outgoing label is 0x0000, and the leading port is port 10;
  • MSS-400-0 configures its own No. 2 table as follows:
  • S16, MSS-400-1 receives the label allocation packet to update its DA, SA and label binding CAM table, that is, the 0x0001 item of the CAM table is: DA 0x10000x00000x00020x0009 SA 0x0000 0x0000 0x0003 0x0012;
  • the 0x0001 entries are: the outgoing label is 0x0001, and the leading port is port 10;
  • MSS-400-1 configures its own No. 2 table as follows:
  • the following codec command packets are respectively directed to STB-0, STB-L STB-0, and STB-1 to start encoding and decoding according to the contents of the packet, and receive and transmit unicast data.
  • the DA of the packet sent by STB-0 to STB-1 is 0x1000 0x0000 0x0003 0x0012 SA is 0x0000 0x0000 0x0002 0x0009;
  • the packet enters MSS-400-0, and the switching engine of MSS-400-0 checks that the metropolitan area network address of the DA is not the access network. Check the CAM table bound to the DA, SA, and label to obtain the unicast label. 0x0000, then check the 0x0000 item of the 6th table to get the label 0x0000, the guiding port is the 10th port, and add the label 0x0000 to the transmitting end of the 10th port, that is, the head of the packet is 0x1000 0x0000 0x0003 0x0012 0x0000 0x0000 0x0002 0x0009 0x0000 0x0000 0x0000;
  • the package enters MX-4-0, and the exchange engine of MX-4-0 looks up the table according to the combined address field.
  • the table is checked, that is, the address of the table is "110 0000 0000 0000", and the data is based on MX-
  • the unicast label packet with the unicast label of 0-00000 of the 4-0 table is directed to port 1 and the outgoing label is 0x0000, that is, the head of the packet is 0x1000 0x0000 0x0003 0x0012 0x0000 0x0000 0x0002 0x0009 0x0000 0x0000 0x0000;
  • the receiving module that enters port 10 of MSS-400-1 removes the label and enters the switching engine.
  • the switching engine of MSS-400-1 looks up the table according to the combined address field. The table is checked, that is, the address of the table is "10 0000 0000 0001 0010", according to the configuration of MSS-400-1 No. 2 table, knowing that the output of this entry is "00 0000 0010" means that the downstream port 1 is opened, so the packet enters STB-1;
  • the DA of the same STB-1 packet sent to STB-0 is 0x1000 0x0000 0x0002 0x0009 SA 0x0000 0x0000 0x0003 0x0012 ;
  • the packet enters the MSS-400-1, and the switching engine of the MSS-400-1 checks that the MAN address of the DA is not the access network.
  • the package enters the MX-4-0, MX-4-0 switching engine according to the combined address field lookup table, check the table No. 6, that is, the address of the table is "110 0000 0000 0001", according to MX-4-
  • the unicast label packet with the unicast label of 0 in the No. 6 table of 0 is directed to port 0.
  • the outbound label is 0x0001, that is, the header of the packet is 0x1000 0x0000 0x0002 0x0009 0x0000 0x0000 0x0003 0x0012 0x0000 0x0000 0x0001;
  • the receiving module that enters port 10 of MSS-400-0 removes the label and enters the switching engine.
  • the switching engine of MSS-400-0 searches the table according to the combined address field, and checks the table No. 2, that is, the address of the table is "10 0000 0000 0000 1001", according to the configuration of MSS-400-G No. 2 table, knowing that the output of this entry is "00 0000 0001" means that the downstream port No. 0 is opened, so the packet enters STB-0.
  • the PDU is the information exchange mode between the user terminal and the server. Both use the raw socket ( Raw Socke t ) to deliver the PDU.
  • the data format is as follows:
  • the node switch and the node server receive the response sent by the 9A01 to the server.
  • the manner in which the node server is connected to the metropolitan area network is only a preferred embodiment of the present invention.
  • the node server adopts any prior art network access mode.
  • the node server may be in the manner of an IP node to enter the metropolitan area network, that is, the metropolitan area network is an IP network structure, and may be arbitrarily used by those skilled in the art according to actual conditions, and the present invention does not need to be limited thereto.
  • the node server can also independently control a local area network, and the node server at this time does not need to access the network, and the access switch and the terminal under the access node can access the local area network controlled by the node server.
  • the advantages of the embodiments of the present invention are further described below by comparison with the IP Internet:
  • the address of the IP Internet is told by the user equipment to the network; the novel network address of the present invention is told by the network to the user equipment.
  • the novel network of the present invention creates an ordered structure "Digital Address System (D/SCAF) with color.
  • D/SCAF Digital Address System
  • the novel network address of the present invention is not only unique, but also has a positionable and determinable function, like a personal identification number. Similarly, other features of the user port's geographic location, device nature, service rights, etc. are implicit.
  • the novel network switch of the present invention specifies the behavior rules of the packet according to these characteristics, and implements data offloading of different natures.
  • Each service issues an independent pass to block the hacker attack and the spread of the virus; the IP Internet can freely enter and exit, and the user has its own firewall; the new network of the present invention must apply for a pass for each service.
  • the premise that the node server can issue independent passes for each service is that the node server knows the address of each service terminal and knows the path topology that each service packet needs to pass.
  • the communication protocol Since the communication protocol is executed at the user terminal, it may be tampered with. Since routing information is broadcast on the network, it may be eavesdropped. The inherent flaws in address spoofing, anonymous attacks, mail bombs, teardrops, covert monitoring, port scanning, internal intrusion, and altered information in the network provide space for hackers. Internet pollution such as spam is difficult to prevent.
  • IP Internet users can set up any IP address to impersonate others, they can send probes to any device on the network to spy on other people's information, and can also send arbitrary interference packets (pollution water) to the network.
  • many smart people have invented various firewalls and tried to stay alone. However, installing a firewall is voluntary, the effect of the firewall is temporary and relative, and the IP Internet itself will never be clean. This is the second security failure of the IP Internet.
  • the network switch After the novel network user of the present invention enters the network, the network switch only allows the user to issue a limited service request to the node server (because the addresses of the respective access network devices are all allocated by the node server), and all other data packets are closed. If the node server approves the user application, it sends a network passport to the switch where the user is located. If each data packet sent by the user terminal does not meet the audit conditions of the network switch, it will be discarded, completely eliminating hacker attacks. The pass is automatically revoked after each service. The pass mechanism is performed by the switch and is not within the user controllable range: Review the source address of the user data packet: Prevent users from sending any fake or anonymous data packets (automatically set after entering the network)
  • Audit target address Users can only send packets to the server-specified object (determined when the service is applied)
  • Audit data traffic User sent data traffic must comply with server regulations (determined at service request)
  • the novel network of the invention does not need negative means such as firewall, antivirus, encryption, internal and external network isolation, etc.
  • the novel network of the invention completely blocks the path of hacker attack and virus diffusion, and is a security network which can be relied on in essence.
  • the network device is completely isolated from the user data, cutting off the lifeline of viruses and Trojans;
  • the IP internet device can arbitrarily disassemble the user data packet; the novel network device of the present invention is completely isolated from the user data. That is, in data transmission, the new network device (for example, switch, gateway, etc.) does not disassemble the user data packet, but directly searches for the mapping table according to the address of the data packet, and forwards it from the corresponding port. That is, the switch of the present invention does not have its own function of calculating and selecting routes.
  • the new network device for example, switch, gateway, etc.
  • the computer created by Von Neumann puts the program instructions and operational data in the same place, that is, a program can modify other programs and data in the machine.
  • This computer model which has been used to this day, has left a chance for Trojan horses, worms, viruses, backdoors, and so on. With the rapid accumulation of viruses, anti-virus software and patches are always slow and passive.
  • IP Internet TCP/IP protocol The core technology of the Internet TCP/IP protocol is best effort, store and forward, and error detection retransmission.
  • network servers and routers In order to realize the mission of the Internet, network servers and routers must have the ability to resolve user data packets. This leaves a way for hacker viruses. Network security has become smarter than anyone else, and there is no peace. This is the third genetic flaw in the IP Internet.
  • the CPU in all the servers and switch devices of the novel network of the present invention access any one of the user data packets. That is to say, the entire novel network of the present invention is only a transparent pipeline for establishing a complete isolation and traffic behavior specification between the service provider and the receiver terminal device. Regardless of what data is sent or received by the user terminal, it is irrelevant to the network. The lifeline of viruses and Trojans has been structurally cut off. Therefore, the novel network of the present invention eliminates the possibility of unrelated persons on the network stealing user data. Similarly, those who want to be hackers or drug makers have no object at all.
  • the IP Internet is a free market with no middlemen; the new network of the present invention is a department store with a middleman. For the network, both consumers and content providers belong to the network user category, but the size is different.
  • the IP Internet is an unmanaged free market where direct communication (P2P) is possible between any user. That is to say, whether or not to manage is the user has the final say, whether or not the charge is a single large user (supplier) has the final say, whether or not to comply with the regulations is also a single large user (vampire website) has the final say. Operators can only accept admission fees at most. It is only impossible to implement legal, ethical, security and business rules. It is impossible now and in the future. This is the fourth structural disability of the IP Internet.
  • the novel network of the present invention creates a service node concept and forms a managed department store business model. Between users, or between consumers and suppliers, there is absolutely no free access, and all connections must be approved by the node server (intermediate). This is a necessary condition for effective management of network services. In order to become a new type of network user, you must first negotiate your role with the network operator. From ordinary consumers, online stores, school hospitals, government departments, to TV stations, all of them belong to the operators' customers, just like the above departments are all telephones. The company's customers are the same. Although it seems that each character is only sending and receiving video content, how to send and receive must strictly abide by their agreed behavior rules. With insurmountable specifications, the relationships between various types of users can be truly divided into C2C, B2C, B2B, etc., or collectively referred to as managed user-to-user communication (MP2P).
  • MP2P managed user-to-user communication
  • node servers (intermediaries) can serve is that the node server can know the address of each service terminal and know the path topology that each service packet needs to pass.
  • the IP Internet pursues the first communication and post-management mode; the novel network of the present invention adopts the first management and the post-communication mode.
  • IP Internet defines management as an additional additional service built on the application layer. Therefore, management naturally becomes a dispensable display. This is the fifth intractable nature of the IP Internet.
  • the novel network user terminal of the present invention can only select one of the designated services within the scope of the node server license.
  • the protocol signaling during the service establishment process is performed by the node server (without the user's hand).
  • the user terminal only passively answers the server's questions, accepts or rejects the service, and cannot participate in the protocol process.
  • the data packet can only be sent in the manner specified by the passport. Any packet that deviates from the passport is always discarded in the underlying switch.
  • the basic idea of the novel network protocol of the present invention is to implement a business model centered on service content, rather than simply completing a simple data exchange. In this mode, security is an inherent property of the new network, not an additional service item attached to the network. Of course, business authority review, resource verification, and billing procedures can all be easily included in the management contract.
  • the device embodiment of the access network device, a node server, and an access switch are basically corresponding to the foregoing method embodiments. For details, refer to the related description in the foregoing method embodiment. I won't go into details.

Landscapes

  • Engineering & Computer Science (AREA)
  • Computer Networks & Wireless Communication (AREA)
  • Signal Processing (AREA)
  • Data Exchanges In Wide-Area Networks (AREA)
  • Small-Scale Networks (AREA)

Abstract

一种接入网设备的入网方法,包括:接入交换机上电,在其内部的下行协议包地址表中设置所有下行协议包导向CPU模块;所述接入交换机接收节点服务器发送的下行协议包,将所述下行协议包导向该接入交换机的CPU模块,所述CPU模块生成上行协议包,并发送给节点服务器;所述下行协议包中包括一个待分配的接入网地址;节点服务器向该接入交换机发送入网命令,所述入网命令中包括该接入交换机的接入网地址,所述接入网地址即为该待分配接入网地址;所述接入交换机更新其内部的下行协议包地址表为,仅目的地址为自己接入网地址的协议包导向CPU模块。本发明可以保证数据包传输的稳定传输速率,避免延时,从而保证网络传输品质。

Description

一种接入网设备的入网方法及系统
技术领域
本发明涉及新型网络的技术领域, 特别是涉及一种接入网设备的 入网方法、 一种接入网设备的入网系统、 一种节点服务器及一种接入 交换机。 背景技术
新型网络(包括互联网)使得不同的个人和机构之间能够交换信 息和其他信息资源。 网络通常包括通路、 传输、 信令以及网络管理等 技术。 这些技术已广泛地见诸于各类文献。 对此作了概括介绍的有:
Steven Shepherd 所 著 的 《 Telecommunications Convergence 》 ( McGraw-Hill, 2000 ), Annabel Z. Dodd所著的《The Essential Guide, to Telecommimications》第三版 ( Prentice Hall PRT, 2001 ),或 Ray Horak 所著的《 Communications Systems and Networks》第二版 ( M&T Books, 2000 )。这些技术以往取得的进展已经充分地增进了信息传输的速度和 质量, 并降低了其费用。
连接终端到一个广域传输网络的通路技术 (如终端装置和网络边 缘的局域环路) 已经从 14.4、 28.8 和 56K 的调制解调器发展到包括 ISDN, Tl、 线缆调制解调器、 DSL、 以太网和无线连接在内的技术。
现今用在广域网中的传输技术包括: 同步光纤网 ( SONET )、 密 集波分复用 (DWDM )、 帧中继、 异步传输模式 (ATM ) 和弹性分组 环 (RPR )。
在所有不同的信令技术中 (如在网络中用来建立、 维持和终结通 信的协议和方法), 互联网协议(IP ) 的应用最为广泛。 事实上, 几乎 所有的通信和网络专家认为集声音(如电话)、 视频和数据网于一体的 一个基于 IP协议的网络(如互联网 )将是不可避免的。 就像一位作者 所阐述的那样: "有一件事是清楚的, 那就是以 IP 为基础的整合各类 网络于一体的列车已经驶离了车站, 有些乘客对此次旅行极具热情, 而另一些则很不情愿地被拖拽而行, 并哭、 叫、 踢打着列举 IP的种种 缺陷。 但是不管它有着何种缺陷, IP已被采纳为一种行业标准, 除了 它以外没有任何其他的技术具有如此大的潜力和发展的空间。 ,,(摘自
1998年 8月 10日《Network World》上的' ΊΡ Convergence: Building the Future" , 作者 Susan Breidenbach )。
随着 Internet的业务爆炸式增长,其应用范围已扩展到社会的各个 领域和各个行业。 从电信业来看, 传统的电信业务已越来越多地采用 IP传输, 即所谓的 Everything Over IP。 现有的电信网的框架将从电路 交换及其组网技术,逐步转向以分组交换特别是 IP 为基础的新框架, 电信网承载的业务将从以电话为主, 转向以数据业务为主。
TCP/IP网络协议
TCP/IP ( Transmission Control Protocol/Internet Protocol, 传输控制 协议 /网间网协议) 是目前世界上应用最为广泛的协议, 它的流行与 Internet 的迅猛发展密切相关一 TCP/IP 最初是为互联网的原型 ARPANET 所设计的, 目的是提供一整套方便实用、 能应用于多种网 络上的协议, 事实证明 TCP/IP做到了这一点, 它使网络互联变得容易 起来, 并且使越来越多的网络加入其中, 成为 Internet的事实标准。
* 应用层一应用层是所有用户所面向的应用程序的统称。 TCP/IP 协议族在这一层面有着^ ί艮多协议来支持不同的应用, 许多大家所熟悉 的基于 Internet 的应用的实现就离不开这些协议。 如我们进行万维网 ( WWW )访问用到了 HTTP协议、 文件传输用 FTP协议、 电子邮件 发送用 SMTP、 域名的解析用 DNS协议、 远程登录用 Telnet协议等 等, 都是属于 TCP/IP应用层的; 就用户而言, 看到的是由一个个软件 所构筑的大多为图形化的操作界面,而实际后台运行的便是上述协议。
* 传输层一这一层的的功能主要是提供应用程序间的通信, TCP/IP协议族在这一层的协议有 TCP和 UDP。
* 网络层一是 TCP/IP协议族中非常关键的一层, 主要定义了 IP 地址格式, 从而能够使得不同应用类型的数据在 Internet 上通畅地传 输, IP协议就是一个网络层协议。
* 网络接口层一这是 TCP/IP软件的最低层, 负责接收 IP数据包 并通过网络发送之, 或者从网络上接收物理帧, 抽出 IP数据报, 交给 IP层。
IP是怎样实现网络互连的? 各个厂家生产的网络系统和设备, 如 以太网、 分组交换网等, 它们相互之间不能互通, 不能互通的主要原 因是因为它们所传送数据的基本单元 (技术上称之为"帧") 的格式不 同。 IP协议实际上是一套由软件程序组成的协议软件, 它把各种不同 "帧"统一转换成" IP 数据包"格式, 这种转换是因特网的一个最重要的 特点, 使所有各种计算机都能在因特网上实现互通, 即具有 "开放性" 的特点。
那么, "数据包" 是什么? 它又有什么特点呢? 数据包也是分组交 换的一种形式, 就是把所传送的数据分段打成 "包", 再传送出去。 但 是, 它属于"无连接型", 是把打成的每个"包"(分组)都作为一个"独 立的报文"传送出去, 所以叫做"数据包"。 这样, 在开始通信之前就不 需要先连接好一条电路, 各个数据包不一定都通过同一条路径传输, 所以叫做"无连接型"。 这一特点非常重要, 在文本信息传输的情况下, 它大大提高了网络的坚固性和安全性。
每个数据包都有报头和报文这两个部分, 报头中有目的地址等必 要内容, 使每个数据包不经过同样的路径都能准确地到达目的地。 在 目的地重新组合还原成原来发送的数据。这就要 IP具有分组打包和集 合组装的功能。
在实际传送过程中, 数据包还要能根据所经过网络规定的分组大 小来改变数据包的长度, IP数据包的最大长度可达 65535个字节。
服务品质保证(QoS )是 IP 互联网的一个主要问题。 尽管长期以 来无数个研究报告试图解决这一难题,如果我们将 QoS 主要里程碑按 时间排列, 不难看出互联网 QoS 是不断降低要求, 并不断失败的无奈 历史。 从" Inte Serv" ( 1990 ) 到" Diff Serv" ( 1997 ) , 再到" Lightload" ( 2001 ) , 各种看似有效的 QoS 局部改善方案加起来, 距离全网范围 品质保证的目标还是像水中的月亮。 QoS 看起来很近,其实遥不可达。
早在 IP 互联网初期阶段, 视讯应用已经成为网络月良务的目标, 如 MBone 。 由于缺乏有效的品质保证, 长期无法开展有商业价值的视频 通讯服务, 削弱了 IP 互联网的盈利能力。 因此, 解决网络传输品质难 题, 具有很大的商业价值。 网络传输品质具体表现为丟包和误码。 电 脑文件对于传输中的错误不敏感, 就算传输过程中大部分数据包都丟 掉了, 只要有 TCP 的重发机制, 电脑还是会认为网络是可用的。但是, 若丟包和误码率高于千分之一, 对同步视讯将会造成视音品质下降。 经验数据告诉我们, 高品质视频通讯甚至要求丟包和误码少于十万分 之一。 当前网络环境的测试数据显示, 绝大部分丟包发生在路由器内 部, 在光纤传输中产生的误码几乎可以忽略不计。
*为什么 "Inte Serv"不成功?
"Inte Serv"建立在独立流资源预留的基础上, 采用 Resource Reservation SetupProtocol ( RSVP ) 。 在一个大规模网络环境中, 如果 能在两个视讯终端之间预留一部分带宽资源, 为该视讯业务专用。 听 起来很好, 但实际上行不通。
首先, 这个方案要求全网设备改造, 等于重新建网, 实际操作几 乎不可能。
其次, 就算实现了全网改造, 比如能够在每一台交换机内, 为 2Mbps 的视讯业务保留 2Mbps带宽, 能否解决品质保证呢? 答案是否 定的。
所谓 RSVP 的 2Mbps 带宽只能对宏观而言, 如果 1 秒钟的数据集 中在前半秒发送,就会造成问题,形成周期性的突发流量。由于, IP 互 联网的核心理念是尽力而为, 在每一个网络节点, 交换机总是试图以 最快速度转发数据。 当一个视讯流通过多级交换机后必然导致流量分 布不均匀。 多个不均匀的非同步流合在一起, 在一段时间内将产生更 大的不均匀, 也就是说, 网络流量一定有周期性的阻塞。 随着视讯用 户数增加, 周期性的阻塞没有上限, 当超过交换机内部储存量, 直接 导致丟包后果。
*为什么 "Diff Serv"不成功?
在" Inte Serv"问世 7 年后, 一种新方法" Diff Serv"开始流行。 "Diff Serv"试图提供一种优于尽力而为的网络服务。 这一方法不需要复杂的 全网资源预留, 实施很简单。 只要在每个数据包中打上"优先级,,标记, 网络交换机首先处理带有"优先级,,的视讯数据。 其基本原理好比银行 为 VIP 客户发放金卡, 能够有效减少高端客户的排队时间。 这个方法 听起来也很好, 但实际上还是行不通。
我们不能忽视一个简单的事实, 单一视讯业务流量远远大于传统 非视讯业务(百倍以上) 。
只要有少量视讯用户, 网络上看到的几乎都是视讯数据包。 如果 大部分数据包都有金卡, 也就谈不上 VIP 了。 另外由于 IP互联入网并 不是强制性的, 尽管 QoS 为用户制定了一套独善其身的道德标准, 但 要求别人都自觉执行根本不现实。
因此, "Diff Serv,,除了在少数企业专网中有用, 难以在大规模公 网中有效推广。
*为什么 "Light load"不成功?
自从 IP 互联网逐步普及以来,人们不间断地寻找解决网络品质保 证的良方。 网络技术专家们经过 10 多年搜肠刮肚, 两大 QoS 方案均 不理想。 在对于解决 QoS 失去信心的大环境下, 一些不愿留名的人提 出了不是办法的办法, 即" Light load"。 其基本设想是所谓的轻载网络, 认为只要给足带宽, 光纤入户, 就不担心网络拥塞。
轻载网络的设想可行吗? 答案还是否定的。
当前的网络技术专家们似乎没有意识到一个基本原理, 网络丟包 现象的根源是流量不均勾性造成的。 从宏观上看, 在一个时间段发送 略快一点, 必然导致另一时间段的拥挤, 只要网络流量不均匀, 网络 可能达到的峰值流量就没有上限,在短时间内可以占满任意大的带宽。
其实, 只要有 2Mbps 带宽就可以传输相当不错的视讯节目, 若有 8Mbps 带宽, 就可以传输 HDTV 品质的视讯内容。 然而, 如果我们在 普通网站上随意点看一段文字或一幅照片, 现今的网站服务器多数使 用千兆网口,其瞬间流量是 HDTV 的数十倍。如果有许多个类似网站, 刚巧碰撞在一起, 在某个短时间产生的突发流量会超过全网用户使用 HDTV 所需, 能够占满任意宽的网络。 统计分析显示, 这种碰撞是很 频繁的。
IP 互联网试图采用储存器来吸收瞬间流量,其后果是增加了传输 时延。 由于储存能力有限, 而突发流量没有上限。 因此, 采用储存方 法只能改善本设备丟包的机会, 在本节点吸收的突发流量将对下一个 节点造成更大的压力。 视讯流量源源不断, 交换机储存方式加剧了突 发流向薄弱节点汇聚, 网络丟包不可避免。
当前的网络建设者们, 采用轻载加上 "Diff Serv"技术, 可以应付 窄带的 VoIP 语音业务。 这是因为语音在网络总流量中不占主要部分, 一旦发生拥挤, 牺牲电脑文件, 对语音优先。 但是, 对于高带宽的视 频通讯而言, 局部扩容只能收到暂时改善的效果。 如果其他地方也扩 容, 网络流量的不均勾性跟着水涨船高, 导致原先已扩容部分的效果 下降。 如果全网都平均扩容的话, 传输品质又将恢复到原先没有扩容 前的样子。 也就是说, 整体扩容是无效的。
当前的设备厂商推荐每户数十, 乃至上百兆的超宽带接入网, 就 算每家都有了光纤到户, 还是难以向消费者展示品质保证的视频通讯 服务。 再复杂的 QoS 手段充其量只能"改善, ΊΡ 互联网的传输品质, 而无法 "保证"网络传输品质。 发明内容
本发明实施例所要解决的技术问题是提供一种接入网设备的入网 方法, 用以保证数据包传输的稳定传输速率, 避免延时, 从而保证网 络传输品质。
本发明实施例还提供了一种接入网设备的入网系统、 一种节点服 务器及一种接入交换机, 用以保证上述方法在实际中的实现及应用。 为了解决上述技术问题, 本发明实施例公开了一种接入网设备的 入网方法, 包括以下步骤: 接入交换机上电, 在其内部的下行协议包地址表中设置所有下行 协议包导向 CPU模块;
所述接入交换机接收节点服务器发送的下行协议包, 依据所述下 行协议包地址表的设置, 将所述下行协议包导向该接入交换机的 CPU 模块, 所述 CPU模块生成上行协议包, 并发送给节点服务器; 所述下 行协议包中包括一个待分配的接入网地址;
节点服务器向该接入交换机发送入网命令, 所述入网命令中包括 该接入交换机的接入网地址, 所述接入网地址即为该接入交换机所接 收下行协议包中的待分配接入网地址;
所述接入交换机更新其内部的下行协议包地址表为, 仅目的地址 为自己接入网地址的协议包导向 CPU模块。
经过上述步骤, 所述接入交换机就入网了。 在本发明实施例中, 所入的 "网"是指新型网。 这种新型网是一种集中控制的网络结构, 该 网络可以是树型网、 星型网、 环状网等等类型, 但在此基础上网络中 需要有集中控制节点来控制整个网络。
新型网分为接入网和城域网两部分。 接入网部分的设备主要可以 分为 3类: 节点服务器, 接入交换机, 终端 (包括各种机顶盒、 编码 板、存储器等)。其中,节点服务器是接入网中起集中控制功能的节点, 可控制接入交换机和终端。 节点服务器可直接与接入交换机相连, 也 可以直接与终端相连。 类似的, 城域网部分的设备也可以分为 3类: 城域服务器, 节点交换机, 节点服务器。 其中, 节点服务器即为接入 网部分的节点服务器, 即节点服务器既属于接入网部分, 又属于城域 网部分。 城域服务器是城域网中起集中控制功能的节点, 可控制节点 交换机和节点服务器。 城域服务器可直接连接节点交换机, 也可直接 连接节点服务器。 由此可见, 整个新型网络是一种分层集中控制的网 络结构, 而节点服务器和城域服务器下控制的网络可以是树型、 星型、 环状等各种结构。
在具体实现中,往往接入交换机下还连接有其它下级接入网设备, 在这种情况下,节点服务器会向已入网的接入交换机发送端口分配包, 即作为本发明的一种优选实施例, 当已入网的接入交换机接收到节点 服务器发送的端口分配包时, 所述的方法还包括以下步骤:
已入网的接入交换机将目的地址为自己接入网地址的端口分配包 导向 CPU模块;
依据包中的端口分配信息, 在其内部的下行协议包地址表中, 设 置各个端口下行协议包所导向的下行端口。
进一步而言, 当已入网的接入交换机接收到节点服务器发送的端 口下行协议包时, 所述的方法还包括以下步骤:
所述接入交换机依据其内部下行协议包地址表的设置, 将所述端 口下行协议包导向对应的下行端口; 所述端口下行协议包中包括一个 待分配的接入网地址;
若节点服务器接收到连接在所述接入交换机下行端口的某个下级 接入网设备发送的端口上行协议包, 则向该下级接入网设备发送入网 命令, 所述入网命令中包括该下级接入网设备的接入网地址, 所述接 入网地址即为该下级接入网设备所接收的端口下行协议包中待分配的 接入网地址。
在本实施例中,所述下级接入网设备可以包括接入交换机或终端。 当所述下级接入网设备为接入交换机时, 所述接入交换机会依据 入网命令更新其内部的下行协议包地址表为, 设置目的地址为自己接 入网地址的协议包导向 CPU模块, 并且还会向节点服务器发送入网命 令应答。
当所述下级接入网设备为终端时, 所述终端会向节点服务器发送 入网命令应答。
当节点服务器收到接入交换机发送的入网命令应答后, 就知道这 台接入交换机已经入网了, 以后将定时向该接入交换机发送状态下行 协议包, 即作为本发明的一种优选实施例, 当已入网的接入交换机接 收到节点服务器定时发送的状态下行协议包时, 所述的方法还包括以 下步骤: 已入网的接入交换机依据其内部下行协议包地址表的设置, 将目 的地址为自己接入网地址的状态下行协议包导向 CPU模块;
所述 CPU模块生成状态上行协议包, 并发送给节点服务器。
对于连接在某个接入交换机下的下级接入网设备, 节点服务器也 会定时向所述下级接入网设备发送状态下行协议包, 即作为本发明的 一种优选实施例, 当已入网的接入交换机接收到节点服务器定时发送 的状态下行协议包时, 所述的方法还包括以下步骤:
已入网的接入交换机依据其内部下行协议包地址表的设置, 将目 的地址为下级接入网设备的接入网地址的状态下行协议包导向对应的 端口; 所述状态下行协议包经由所述端口被传送至对应的下级接入网 设备;
所述下级接入网设备针对所接收到的状态下行协议包生成状态上 行协议包, 并发送给节点服务器。
在具体实现中, 所述接入交换机内部还可以设置上行协议包地址 表, 在这种情况下, 所述的方法还包括以下步骤:
接入交换机上电时, 在其内部的上行协议包地址表中设置所有上 行协议包导向的上行端口。
作为又一种实现方式, 所述接入交换机内部还可以设置数据包地 址表, 在这种情况下, 所述的方法还包括以下步骤:
接入交换机上电时, 在其内部的数据包地址表中设置所有数据包 的导向关闭。
本发明实施例还可以包括节点服务器入网的过程, 当节点服务器 上电后, 将导入下级接入网设备的注册信息至 CPU , 获得城域网地址, 并配置自己的接入网地址。
作为本发明的一种优选实施例, 所述节点服务器入网的过程可以 包括以下步骤:
所述节点服务器入城域网, 所述城域网是具有集中控制功能的网 络, 包括城域服务器、 节点交换机和节点服务器, 其中所述节点交换 机连接在城域服务器和节点服务器之间; 所述节点服务器入城域网的 步骤包括:
城域网设备接入城域网, 由城域网中具有集中控制功能的城域服 务器为入网的设备分配协议标签和城域网地址;
其中, 所述城域网设备包括节点交换机和节点服务器, 所述协议 标签用于描述城域网设备与城域服务器之间的连接; 当同一个城域网 设备与城域服务器之间有多个连接时, 城域服务器为每个连接分配不 同的协议标签;
针对每个跨越城域网的服务申请, 城域服务器分配对应服务的数 据标签 ,所述数据标签用于描述服务所涉及的节点服务器之间的连接。
优选的是, 所述标签分为入标签和出标签, 入标签指数据包进入 城域服务器或节点交换机的标签, 出标签指该数据包离开城域服务器 或节点交换机的标签;
同一个数据包的入标签和出标签为不同的标签,或者为相同标签。 更具体而言, 城域网设备接入城域网, 由城域服务器分配协议标 签和城域网地址的步骤包括以下子步骤:
城域服务器向其所有下行端口发送城域查询标签包, 每个城域查 询标签包中包含一个城域服务器分配的待用协议标签;
节点服务器上电后, 收到城域服务器发送的城域查询标签包, 然 后向城域服务器返回城域应答标签包, 所述城域应答标签包中包含城 域网设备的序列号和收到城域查询标签包的端口号;
城域服务器收到城域应答标签包后, 根据包中的序列号验证城域 网设备是否注册, 如果已注册, 则向城域网设备收到城域查询标签包 的端口发送入网命令, 所述入网命令中包含城域服务器为城域网设备 分配的城域网地址和所述待用协议标签;
城域网设备的相应端口收到入网命令后, 返回入网命令应答, 城 域网设备接入城域网。 作为本发明的另一种优选实施例,所述节点服务器入网可以 IP节 点的方式入城域网, 即所述城域网为 IP网络结构。 其具体实现延用现 在 IP入网的方式即可, 本发明在此就不贅述了。
为实现更好地集中控制管理,本发明实施例还可以包括以下步骤: 节点服务器在其内部的下行协议包地址表中设置各下行协议包分 别导向的下行端口;
所述节点服务器依据所述下行协议包地址表的设置, 通过相应的 下行端口发送下行协议包。
更为优选的是, 本发明实施例还可以包括以下步骤:
节点服务器生成包括有端口分配信息的端口分配包;
节点服务器依据其内部下行协议包地址表的设置, 通过相应的下 行端口发送端口分配包。
为更好地分配及管理下级接入网设备的地址, 所述节点服务器内 部还设置有地址信息表, 所述地址信息表中记录有地址占用信息、 设 备标识信息和设备资源信息。 对于所述地址信息表的操作可以包括如 下情形:
情形一、 节点服务器上电时, 初始化所述地址信息表, 所述初始 化操作包括:
在地址信息表的空白表项中写入节点服务器的接入网地址, 并将 地址占用信息标记为该地址已用;
写入设备标识信息为当前节点服务器;
写入设备资源信息为当前节点服务器的端口信息。
情形二、 依据节点服务器在其内部的下行协议包地址表中对各下 行协议包所导向的下行端口的设置, 更新所述地址信息表, 所述更新 操作包括:
在地址信息表的空白表项中分别写入各个下行协议包中的待分配 接入网地址, 并将地址占用信息标记为该地址待用。
情形三、 依据对端口下行协议包所导向的其他接入设备的下行端 口的设置, 更新所述地址信息表, 所述更新操作包括: 在地址信息表的空白表项中分别写入各个端口下行协议包中的待 分配接入网地址, 并将地址占用信息标记为该地址待用。
情形四、当节点服务器接收到接入交换机发送的入网命令应答时, 更新所述地址信息表, 所述更新操作包括:
将所述接入交换机的接入网地址对应表项中的地址占用信息标记 为该地址已用;
更新所述设备标识信息为当前接入交换机;
更新所述设备资源信息为当前接入交换机的端口信息, 以及, 其 端口所连接的接入网设备的接入网地址信息。
情形五、 当节点服务器接收到终端发送的入网命令应答时, 更新 所述地址信息表, 所述更新操作包括:
将所述终端的接入网地址对应表项中的地址占用信息标记为该地 址已用;
更新所述设备标识信息为当前终端;
更新所述设备资源信息为当前终端的端口信息, 以及, 其端口所 连接的接入网设备的接入网地址信息。
情形六、 若节点服务器在预设时间内未收到所述接入交换机发送 的状态上行协议包,则会停止向所述接入交换机发送状态下行协议包, 并更新所述地址信息表; 所述更新操作包括:
将所述接入交换机的接入网地址对应表项中的地址占用信息标记 为该地址未用;
清除所述设备标识信息和设备资源信息。
情形七、 若节点服务器在预设时间内未收到所述下级接入网设备 发送的状态上行协议包, 则停止向所述下级接入网设备发送状态下行 协议包, 并更新所述地址信息表; 所述更新操作包括:
将所述下级接入网设备的接入网地址对应表项中的地址占用信息 标己为该地址未用;
清除所述设备标识信息和设备资源信息。 为便于控制和管理下级接入网设备和数据传送的通信路径, 所述 节点服务器的地址信息表中还可以包括当前接入网设备端口的上下行 流量信息。
更为优选的是, 所述节点服务器内部还可以设置有上行协议包地 址表, 所述的方法还包括以下步骤:
节点服务器上电时, 在其内部的上行协议包地址表中设置, 所有 上行协议包导向 CPU模块。
更为优选的是,所述节点服务器内部还可以设置有数据包地址表, 所述的方法还包括以下步骤:
节点服务器上电时, 在其内部的数据包地址表中设置所有数据包 的导向关闭。
对于请求接入的下级接入网设备, 节点服务器通常都会先判断其 是否合法, 并在合法时才对其发送入网命令, 即在本发明的一种优选 实施例中, 所述的方法还包括以下步骤:
节点服务器的 CPU模块依据入网命令应答包, 判断是否存在所述 接入交换机的注册信息, 若是, 则判定该接入交换机合法; 否则, 判 断该接入交换机不合法;
若合法, 则发送入网命令。
或者,
节点服务器的 CPU模块依据入网命令应答包, 判断是否存在所述 接入网设备的注册信息, 若是, 则判定该接入网设备合法; 否则, 判 断该接入网设备不合法;
若合法, 则发送入网命令。
一般而言, 所述注册信息可以包括设备类型和设备标识。 本发明实施例还包括一种节点服务器, 包括以下模块:
下行协议包发送模块, 用于向接入交换机发送下行协议包, 所述 下行协议包中包括一个待分配的接入网地址; 上行协议包接收模块, 用于接收接入交换机针对所述下行协议包 回复的上行协议包;
第一入网命令发送模块, 用于依据所接收的上行协议包, 向相应 的接入交换机发送入网命令, 所述入网命令中包括分配给该接入交换 机的接入网地址, 所述接入网地址为该接入交换机所接收下行协议包 中的待分配接入网地址。
为控制和管理连接在接入交换机下的下级接入网设备, 作为本发 明的一种优选实施例, 所述节点服务器还可以包括以下模块:
端口分配包发送模块, 用于向已入网的接入交换机发送端口分配 包, 所述端口分配包中包括端口分配信息, 所述端口分配信息为各个 端口下行协议包导向所述接入交换机各个下行端口的信息。
更为优选的是, 所述的节点服务器还可以包括以下模块: 端口下行协议包发送模块, 用于向已入网的接入交换机发送端口 下行协议包; 所述端口下行协议包中包括一个待分配的接入网地址; 端口上行协议包接收模块, 用于接收连接在所述接入交换机下行 端口的某个下级接入网设备发送的端口上行协议包;
第二入网命令发送模块,用于向该下级接入网设备发送入网命令 , 所述入网命令中包括该下级接入网设备的接入网地址, 所述接入网地 址即为该下级接入网设备所接收的端口下行协议包中待分配的接入网 地址。
在本实施例中, 所述节点服务器还可以包括:
入网应答接收模块, 用于接收所述下级接入网设备发送的入网命 令应答。
在本发明实施例中, 所述下级接入网设备可以包括接入交换机或 终端。
为检测直接连接在所述节点服务器下的接入交换机的运行情况, 所述节点服务器还包括以下模块:
状态下行协议包发送模块, 用于定时向已入网的接入交换机发送 状态下行协议包; 状态上行协议包第一接收模块, 用于接收所述接入交换机针对所 述状态下行协议包回复的状态上行协议包。
为检测连接在所述接入交换机下的下级接入网设备的运行情况, 所述节点服务器还包括以下模块:
状态上行协议包第二接收模块, 用于接收下级接入网设备针对所 述状态下行协议包回复的状态上行协议包。
在具体实现中, 所述的节点服务器还包括以下模块:
初始化模块, 用于在上电后, 导入下级接入网设备的注册信息至 CPU , 获得城域网地址, 并配置自己的接入网地址。
为接入城域网, 所述节点服务器还可以包括以下模块:
协议标签和地址获取模块, 用于接入城域网后, 从城域服务器获 得分配的协议标签和城域网地址, 所述协议标签用于描述节点服务器 与城域服务器之间的连接; 当同一个节点服务器与上级连接设备有多 个连接时, 获得对应每个连接的不同协议标签, 所述上级连接设备包 括节点交换机和城域服务器;所述城域网是具有集中控制功能的网络, 包括城域服务器、 节点交换机和节点服务器, 其中所述节点交换机连 接在城域服务器和节点服务器之间;
数据标签获取模块, 用于针对每个跨越城域网的服务申请, 从城 域服务器获得分配的对应当次服务的数据标签, 所述数据标签用于描 述服务所涉及的节点服务器之间的连接;
标签添加模块, 用于将节点服务器发给城域网的协议包或数据包 添加相应的协议标签或数据标签;
标签删除模块, 用于将从城域网接收的协议包或数据包去掉相应 的协议标签或数据标签。
优选的是, 所述标签分为入标签和出标签, 入标签指数据包进入 城域服务器或节点交换机的标签, 出标签指该数据包离开城域服务器 或节点交换机的标签;
同一个数据包的入标签和出标签为不同的标签,或者为相同标签; 所述标签包括协议标签和数据标签。 更为优选的是, 所述的节点服务器还可以包括:
地址-标签映射表, 用于针对每个跨越城域网的服务, 记录跨越城 域网的两个终端的接入网地址与出标签的绑定关系; 其中, 接入网地 址为每个节点服务器为其下连接的入网设备分配的地址;
则所述标签添加模块依据地址-标签映射表,查找节点服务器发给 城域网的协议包或数据包对应的出标签,并添加查找到的出标签发送。
在这种情况下, 所述的节点服务器还可以包括:
协议包标签表, 用于节点服务器上电时, 设置所有的城域协议包 导向到 CPU模块, 所述城域协议包包括城域服务器发送的城域查询标 签包和城域服务申请包。
和 /或, 所述节点服务器还可以包括以下模块:
应答包标签表,用于将城域应答标签包分别导向到相应上行端口; 应答包标签表初始化模块, 用于节点服务器上电时, 设置所有城 域应答标签包的导向关闭;
应答包标签表配置模块, 用于收到城域服务器发送的城域查询标 签包后, 修改自身的应答包标签表, 将所述协议标签对应的城域应答 标签包导向到接收该城域查询标签包的上行端口。
作为另一种接入城域网的实现方式, 所述节点服务器可以包括以 下模块:
IP入网模块, 用于以 IP节点的方式入城域网, 所述城域网为 IP 网络结构。
优选的是, 所述节点服务器还可以包括
0 号表初始化配置模块, 用于在上电时, 初始化其内部的下行协 议包地址表为各下行协议包的导向关闭。
为实现更好地集中控制管理, 所述节点服务器还可以包括:
0 号表第一设置模块, 用于在其内部的下行协议包地址表中设置 各下行协议包分别导向的下行端口;
所述下行协议包发送模块依据所述下行协议包地址表的设置, 通 过相应的下行端口发送下行协议包。 更为优选的, 所述节点服务器还可以包括:
端口分配包生成模块, 用于生成包括有所述端口分配信息的端口 分配包;
所述端口分配包发送模块依据其内部下行协议包地址表的设置, 通过相应的下行端口发送所述端口分配包。
为更好地分配及管理下级接入网设备的地址, 所述节点服务器内 部还设置有地址信息表, 所述地址信息表中记录有地址占用信息、 设 备标识信息和设备资源信息。
根据对所述地址信息表的操作的不同情形, 所述节点服务器还可 以包括以下模块:
情形一、 所述节点服务器还包括:
地址信息表初始化模块, 用于在上电时, 初始化地址信息表, 所 述初始化包括:
在地址信息表的空白表项中写入节点服务器的接入网地址, 并将 地址占用信息标记为该地址已用;
写入设备标识信息为当前节点服务器;
写入设备资源信息为当前节点服务器的端口信息。
情形二、 节点服务器还包括:
地址信息表第一更新模块, 用于依据节点服务器在其内部的下行 协议包地址表中对各下行协议包所导向的下行端口的设置, 更新所述 地址信息表, 所述更新包括:
在地址信息表的空白表项中分别写入各个下行协议包中的待分配 接入网地址, 并将地址占用信息标记为该地址待用。
情形三、 节点服务器还包括:
地址信息表第二更新模块, 用于依据对端口下行协议包所导向的 其他接入设备的下行端口的设置, 更新所述地址信息表, 所述更新包 括:
在地址信息表的空白表项中分别写入各个端口下行协议包中的待 分配接入网地址, 并将地址占用信息标记为该地址待用。 情形四、 节点服务器还包括:
地址信息表第三更新模块, 用于在接收到接入交换机发送的入网 命令应答时, 更新所述地址信息表, 所述更新包括:
将所述接入交换机的接入网地址对应表项中的地址占用信息标记 为该地址已用;
更新所述设备标识信息为当前接入交换机;
更新所述设备资源信息为当前接入交换机的端口信息, 以及, 其 端口所连接的接入网设备的接入网地址信息。
情形五、 节点服务器还包括:
地址信息表第四更新模块, 用于在接收到下级接入网设备发送的 入网命令应答时, 更新所述地址信息表, 所述更新包括:
将所述下级接入网设备的接入网地址对应表项中的地址占用信息 标" ^己为该地址已用;
更新所述设备标识信息为当前下级接入网设备;
更新所述设备资源信息为当前下级接入网设备的端口信息,以及, 其端口所连接的接入网设备的接入网地址信息。
情形六、 节点服务器还包括:
状态下行协议包第一停发模块, 用于在预设时间内未收到所述接 入交换机发送的状态上行协议包时, 停止向所述接入交换机发送状态 下行协议包;
地址信息表第四更新模块, 用于在停止向所述接入交换机发送状 态下行协议包时, 更新所述地址信息表; 所述更新包括:
将所述接入交换机的接入网地址对应表项中的地址占用信息标记 为该地址未用;
清除所述设备标识信息和设备资源信息。
情形七、 节点服务器还包括:
状态下行协议包第二停发模块, 用于在预设时间内未收到所述下 级接入网设备发送的状态上行协议包, 则停止向所述下级接入网设备 发送状态下行协议包; 地址信息表第五更新模块, 用于在停止向所述下级接入网设备发 送状态下行协议包时, 更新所述地址信息表; 所述更新包括:
将所述下级接入网设备的接入网地址对应表项中的地址占用信息 标己为该地址未用;
清除所述设备标识信息和设备资源信息。
为便于控制和管理下级接入网设备和数据传送的通信路径, 所述 地址信息表中还包括当前接入网设备端口的上下行流量信息。
更为优选的是,所述节点服务器内部还设置有上行协议包地址表, 所述节点服务器还包括:
1 号表初始化配置模块, 用于在上电时, 在其内部的上行协议包 地址表中设置, 所有上行协议包导向 CPU模块。
更为优选的是, 所述节点服务器内部还设置有数据包地址表, 所 述节点服务器还包括:
数据包地址表配置模块, 用于上电时, 在其内部的数据包地址表 中设置所有数据包的导向关闭。
对于请求接入的下级接入网设备, 节点服务器通常都会先判断其 是否合法, 并在合法时才对其发送入网命令, 即在本发明的一种优选 实施例中, 所述节点服务器还包括:
第一验证模块, 用于判断是否存在所述接入交换机的注册信息, 若是, 则判定该接入交换机合法; 否则, 判断该接入交换机不合法; 若合法, 则触发第一入网命令发送模块。
或者,
第二验证模块, 用于判断是否存在所述接入网设备的注册信息, 若是, 则判定该接入网设备合法; 否则, 判断该接入网设备不合法; 若合法, 则触发第二入网命令发送模块。 本发明实施例还公开了一种接入交换机, 包括以下模块:
0 号表初始化配置模块, 用于在上电时, 在其内部的下行协议包 地址表中设置所有下行协议包导向 CPU模块; 下行协议包接收模块, 用于接收节点服务器发送的下行协议包, 并依据所述下行协议包地址表的设置, 将所述下行协议包导向该接入 交换机的 CPU模块,所述下行协议包中包括一个待分配的接入网地址; 上行协议包回复模块, 用于由所述 CPU模块生成上行协议包, 并 发送给节点服务器;
第一入网命令接收模块, 用于接收节点服务器发送的入网命令, 所述入网命令中包括该接入交换机的接入网地址, 所述接入网地址即 为该接入交换机所接收下行协议包中的待分配接入网地址;
0 号表第一设置模块, 用于更新其内部的下行协议包地址表为, 仅目的地址为自己接入网地址的协议包导向 CPU模块。
在入网后, 所述接入交换机还包括以下模块:
端口分配包接收模块, 用于在入网后, 接收节点服务器发送的端 口分配包; 所述端口分配包中包括端口分配信息, 所述端口分配信息 为各个端口下行协议包导向接入交换机各个下行端口的信息;
第一导向模块, 用于将目的地址为自己接入网地址的端口分配包 导向 CPU模块;
0 号表第二设置模块, 用于依据所述端口分配信息, 在其内部的 下行协议包地址表中, 设置各个端口下行协议包所导向的下行端口。
更为优选的, 所述接入交换机还包括以下模块:
端口下行协议包接收模块, 用于在入网后, 接收到节点服务器发 送的端口下行协议包; 所述端口下行协议包中包括一个待分配的接入 网地址;
第二导向模块, 用于依据其内部下行协议包地址表的设置, 将所 述端口下行协议包导向对应的下行端口。
更为优选的, 所述接入交换机还包括入网命令应答模块, 用于向 节点服务器发送入网命令应答。
在接收节点服务器定时发送的状态下行协议包时, 所述接入交换 机还包括: 状态下行协议包接收模块, 用于在入网后, 接收节点服务器定时 发送的状态下行协议包;
第三导向模块, 用于依据其内部下行协议包地址表的设置, 将目 的地址为自己接入网地址的状态下行协议包导向 CPU模块;
状态上行协议包回复模块, 用于由所述 CPU模块生成状态上行协 议包, 并发送给节点服务器。
优选的, 所述接入交换机还包括:
第四导向模块, 用于依据其内部下行协议包地址表的设置, 将目 的地址为下级接入网设备的接入网地址的状态下行协议包导向对应的 端口; 所述状态下行协议包经由所述端口被传送至对应的下级接入网 设备。
优选的, 所述接入交换机内部还设置有上行协议包地址表, 所述 的接入交换机还包括:
1 号表初始化配置模块, 用于在上电时, 在其内部的上行协议包 地址表中设置所有上行协议包导向的上行端口。
优选的, 所述接入交换机内部还设置有数据包地址表, 所述的接 入交换机还包括:
数据包地址表配置模块, 用于在上电时, 在其内部的数据包地址 表中设置所有数据包的导向关闭。 本发明实施例还公开了一种接入网设备的入网系统, 所述接入网 设备包括节点服务器及连接在其下级的接入交换机, 所述节点服务器 包括向接入交换机发送下行协议包的下行协议包发送模块, 以及, 依 据接入交换机回复的上行协议包发送入网命令的第一入网命令发送模 块;
所述接入交换机包括:
0 号表初始化配置模块, 用于在上电时, 在其内部的下行协议包 地址表中设置所有下行协议包导向 CPU模块; 下行协议包接收模块, 用于依据所述下行协议包地址表的设置, 将接收到的下行协议包导向该接入交换机的 CPU模块, 所述下行协议 包中包括一个待分配的接入网地址;
上行协议包回复模块, 用于由所述 CPU模块生成上行协议包, 并 发送给节点服务器;
第一入网命令接收模块, 用于接收节点服务器发送的入网命令, 所述入网命令中包括该接入交换机的接入网地址, 所述接入网地址即 为该接入交换机所接收下行协议包中的待分配接入网地址;
0 号表第一设置模块, 用于更新其内部的下行协议包地址表为, 仅目的地址为自己接入网地址的协议包导向 CPU模块。
在具体实现中,往往接入交换机下还连接有其它下级接入网设备, 在这种情况下,节点服务器会向已入网的接入交换机发送端口分配包, 即作为本发明的一种优选实施例, 所述节点服务器还包括用于向已入 网的接入交换机发送端口分配包的端口分配包发送模块, 所述端口分 配包中包括端口分配信息, 所述端口分配信息为各个端口下行协议包 导向所述接入交换机各个下行端口的信息;
在这种情况下, 所述接入交换机还包括:
第一导向模块, 用于将目的地址为自己接入网地址的端口分配包 导向 CPU模块;
0 号表第二设置模块, 用于依据所述端口分配信息, 在其内部的 下行协议包地址表中, 设置各个端口下行协议包所导向的下行端口。
进一步而言, 所述节点服务器还包括用于向已入网的接入交换机 发送端口下行协议包的端口下行协议包发送模块, 所述端口下行协议 包中包括一个待分配的接入网地址;
在这种情况下, 所述接入交换机还包括:
第二导向模块, 用于依据其内部下行协议包地址表的设置, 将所 述端口下行协议包导向对应的下行端口。 在具体实现中, 所述接入网设备还包括连接在已入网接入交换机 下行端口的下级接入网设备, 在这种情况下, 所述节点服务器还包括 用于向所述下级接入网设备发送入网命令的第二入网命令发送模块; 所述下级接入网设备包括:
端口上行协议包回复模块, 用于针对接收到的端口下行协议包生 成端口上行协议包, 并发送给节点服务器;
第二入网命令接收模块, 用于接收节点服务器发送的入网命令, 所述入网命令中包括该下级接入交换机的接入网地址, 所述接入网地 址即为该下级接入交换机所接收端口下行协议包中的待分配接入网地 址。
在本实施例中, 所述下级接入网设备包括接入交换机或终端。 当所述下级接入网设备为接入交换机时 ,所述接入交换机还包括: 0 号表第三设置模块, 用于依据所述入网命令更新其内部的下行 协议包地址表为, 设置目的地址为自己接入网地址的协议包导向 CPU 模块。
当节点服务器收到接入交换机发送的入网命令应答后, 就知道这 台接入交换机已经入网了, 以后将定时向该接入交换机发送状态下行 协议包, 即作为本发明的一种优选实施例, 所述节点服务器还包括用 于定时向已入网的接入交换机发送状态下行协议包的状态下行协议包 发送模块;
在这种情况下, 所述的接入交换机还包括:
第三导向模块, 用于依据其内部下行协议包地址表的设置, 将目 的地址为自己接入网地址的状态下行协议包导向 CPU模块;
状态上行协议包回复模块, 用于由所述 CPU模块生成状态上行协 议包, 并发送给节点服务器。
对于连接在某个接入交换机下的下级接入网设备, 节点服务器也 会定时向所述下级接入网设备发送状态下行协议包, 即作为本发明的 一种优选实施例, 所述接入交换机还包括: 第四导向模块, 用于依据其内部下行协议包地址表的设置, 将目 的地址为下级接入网设备的接入网地址的状态下行协议包导向对应的 端口; 所述状态下行协议包经由所述端口被传送至对应的下级接入网 设备。
当所述下级接入网设备为下级接入交换机时; 所述下级接入交换 机还包括:
状态上行协议包回复模块, 用于针对所接收到的状态下行协议包 生成状态上行协议包, 并发送给节点服务器。
当所述下级接入网设备为终端时; 所述终端还包括:
状态上行协议包回复模块, 用于针对所接收到的状态下行协议包 生成状态上行协议包, 并发送给节点服务器。
优选的, 所述接入交换机内部还设置有上行协议包地址表, 所述 接入交换机还包括:
1 号表初始化配置模块, 用于在上电时, 在其内部的上行协议包 地址表中设置所有上行协议包导向的上行端口。
优选的, 所述接入交换机内部还设置有数据包地址表, 所述接入 交换机还包括:
数据包地址表配置模块, 用于在上电时, 在其内部的数据包地址 表中设置所有数据包的导向关闭。
优选的, 在节点服务器入网时, 所述节点服务器还包括: 初始化模块, 用于在上电后, 导入下级接入网设备的注册信息至
CPU , 获得城域网地址, 并配置自己的接入网地址。
作为本发明的一种优选实施例, 所述节点服务器还包括用于接入 城域网的以下模块:
协议标签和地址获取模块, 用于接入城域网后, 从城域服务器获 得分配的协议标签和城域网地址, 所述协议标签用于描述节点服务器 与城域服务器之间的连接; 当同一个节点服务器与上级连接设备有多 个连接时, 获得对应每个连接的不同协议标签, 所述上级连接设备包 括节点交换机和城域服务器;所述城域网是具有集中控制功能的网络, 包括城域服务器、 节点交换机和节点服务器, 其中所述节点交换机连 接在城域服务器和节点服务器之间;
数据标签获取模块, 用于针对每个跨越城域网的服务申请, 从城 域服务器获得分配的对应当次服务的数据标签, 所述数据标签用于描 述服务所涉及的节点服务器之间的连接;
标签添加模块, 用于将节点服务器发给城域网的协议包或数据包 添加相应的协议标签或数据标签;
标签删除模块, 用于将从城域网接收的协议包或数据包去掉相应 的协议标签或数据标签。
优选的, 所述标签分为入标签和出标签, 入标签指数据包进入城 域服务器或节点交换机的标签, 出标签指该数据包离开城域服务器或 节点交换机的标签;
同一个数据包的入标签和出标签为不同的标签,或者为相同标签; 所述标签包括协议标签和数据标签。
优选的, 所述节点服务器还包括:
地址-标签映射表, 用于针对每个跨越城域网的服务, 记录跨越城 域网的两个终端的接入网地址与出标签的绑定关系; 其中, 接入网地 址为每个节点服务器为其下连接的入网设备分配的地址;
则所述标签添加模块依据地址-标签映射表,查找节点服务器发给 城域网的协议包或数据包对应的出标签,并添加查找到的出标签发送。
优选的, 所述节点服务器还包括:
协议包标签表, 用于节点服务器上电时, 设置所有的城域协议包 导向到 CPU模块, 所述城域协议包包括城域服务器发送的城域查询标 签包和城域服务申请包。
优选的, 所述节点服务器还包括:
应答包标签表,用于将城域应答标签包分别导向到相应上行端口; 应答包标签表初始化模块, 用于节点服务器上电时, 设置所有城 域应答标签包的导向关闭; 应答包标签表配置模块, 用于收到城域服务器发送的城域查询标 签包后, 修改自身的应答包标签表, 将所述协议标签对应的城域应答 标签包导向到接收该城域查询标签包的上行端口。
作为本发明的另一种优选实施例, 所述节点服务器还包括用于接 入城域网的 IP入网模块, 用于以 IP节点的方式入城域网, 所述城域 网为 IP网络结构。
优选的, 所述节点服务器还包括:
0 号表初始化配置模块, 用于在上电时, 初始化其内部的下行协 议包地址表为各下行协议包的导向关闭。
优选的, 所述节点服务器还包括:
0 号表第一设置模块, 用于在其内部的下行协议包地址表中设置 各下行协议包分别导向的下行端口;
所述下行协议包发送模块依据所述下行协议包地址表的设置, 通 过相应的下行端口发送下行协议包。
优选的, 所述节点服务器还包括:
端口分配包生成模块, 用于生成包括有所述端口分配信息的端口 分配包;
所述端口分配包发送模块依据其内部下行协议包地址表的设置, 通过相应的下行端口发送所述端口分配包。
为更好地分配及管理下级接入网设备的地址, 所述节点服务器内 部还设置有地址信息表, 所述地址信息表中记录有地址占用信息、 设 备标识信息和设备资源信息。
根据对所述地址信息表的操作的不同情形, 所述节点服务器还可 以包括以下模块:
情形一、 所述节点服务器还包括:
地址信息表初始化模块, 用于在上电时, 初始化地址信息表, 所 述初始化包括:
在地址信息表的空白表项中写入节点服务器的接入网地址, 并将 地址占用信息标记为该地址已用; 写入设备标识信息为当前节点服务器;
写入设备资源信息为当前节点服务器的端口信息。
情形二、 所述节点服务器还包括:
地址信息表第一更新模块, 用于依据节点服务器在其内部的下行 协议包地址表中对各下行协议包所导向的下行端口的设置, 更新所述 地址信息表, 所述更新包括:
在地址信息表的空白表项中分别写入各个下行协议包中的待分配 接入网地址, 并将地址占用信息标记为该地址待用。
情形三、 所述节点服务器还包括:
地址信息表第二更新模块, 用于依据对端口下行协议包所导向的 其他接入设备的下行端口的设置, 更新所述地址信息表, 所述更新包 括:
在地址信息表的空白表项中分别写入各个端口下行协议包中的待 分配接入网地址, 并将地址占用信息标记为该地址待用。
情形四、 所述节点服务器还包括:
地址信息表第三更新模块, 用于在接收到接入交换机发送的入网 命令应答时, 更新所述地址信息表, 所述更新包括:
将所述接入交换机的接入网地址对应表项中的地址占用信息标记 为该地址已用;
更新所述设备标识信息为当前接入交换机;
更新所述设备资源信息为当前接入交换机的端口信息, 以及, 其 端口所连接的接入网设备的接入网地址信息。
情形五、 所述节点服务器还包括:
地址信息表第四更新模块, 用于在接收到下级接入网设备发送的 入网命令应答时, 更新所述地址信息表, 所述更新包括:
将所述下级接入网设备的接入网地址对应表项中的地址占用信息 标" ^己为该地址已用;
更新所述设备标识信息为当前下级接入网设备; 更新所述设备资源信息为当前下级接入网设备的端口信息,以及, 其端口所连接的接入网设备的接入网地址信息。
情形六、 所述节点服务器还包括:
状态下行协议包第一停发模块, 用于在预设时间内未收到所述接 入交换机发送的状态上行协议包时, 停止向所述接入交换机发送状态 下行协议包;
地址信息表第四更新模块, 用于在停止向所述接入交换机发送状 态下行协议包时, 更新所述地址信息表; 所述更新包括:
将所述接入交换机的接入网地址对应表项中的地址占用信息标记 为该地址未用;
清除所述设备标识信息和设备资源信息。
情形七、 所述节点服务器还包括:
状态下行协议包第二停发模块, 用于在预设时间内未收到所述下 级接入网设备发送的状态上行协议包, 则停止向所述下级接入网设备 发送状态下行协议包;
地址信息表第五更新模块, 用于在停止向所述下级接入网设备发 送状态下行协议包时, 更新所述地址信息表; 所述更新包括:
将所述下级接入网设备的接入网地址对应表项中的地址占用信息 标己为该地址未用;
清除所述设备标识信息和设备资源信息。
优选的, 所述节点服务器的地址信息表中还包括当前接入网设备 端口的上下行流量信息。
优选的, 所述节点服务器内部还设置有上行协议包地址表, 所述 节点服务器还包括:
1 号表初始化配置模块, 用于在上电时, 在其内部的上行协议包 地址表中设置, 所有上行协议包导向 CPU模块。
优选的, 所述节点服务器内部还设置有数据包地址表, 所述节点 服务器还包括: 数据包地址表配置模块, 用于上电时, 在其内部的数据包地址表 中设置所有数据包的导向关闭。
优选的, 所述节点服务器还包括:
第一验证模块, 用于判断是否存在所述接入交换机的注册信息, 若是, 则判定该接入交换机合法; 否则, 判断该接入交换机不合法; 若合法, 则触发第一入网命令发送模块。
优选的, 所述节点服务器还包括:
第二验证模块, 用于判断是否存在所述接入网设备的注册信息, 若是, 则判定该接入网设备合法; 否则, 判断该接入网设备不合法; 若合法, 则触发第二入网命令发送模块。 需要说明的是, 从实际硬件的角度划分, 本发明的节点服务器主 要包括网络接口模块、 交换引擎模块、 CPU 模块、 磁盘阵列模块; 本 发明的接入交换机主要包括网络接口模块 (下行网络接口模块、 上行 网络接口模块)、 交换引擎模块和 CPU模块; 由于上述硬件模块针对不 同的处理情况, 所执行的功能并不相同, 比如, 交换引擎模块对于不 同的包(协议包、 数据包等), 所查的地址表就不一样, 获得的包的导 向信息就不一样; 或如, CPU 模块在收到包的情况下, 可能是配地址 表, 也可能是解析包并生成一个应答包。 因而, 为使本领域技术人员 更好地理解本发明, 在本发明实施例中, 主要从功能角度描述了本发 明的节点服务器、接入交换机及接入网设备的入网系统所涉及的模块, 然而实质上, 这些功能模块与实际硬件模块是对应的。 与现有技术相比, 本发明具有以下优点:
多媒体服务所必须的基本条件和传统的数据性服务所必须的基本 条件 (如网页文本、 图像、 电子邮件、 FTP和 DNS服务)有着显著差 别。 特别是多媒体服务对于端对端延时和延时变化特别敏感, 但是却 能容忍偶然的数据丟失, 这些在服务需求上的截然不同表明了原先已 设计成用来传输数据的新型网络架构是不适合被用来提供多媒体服务 的。
本发明以主控的方式, 通过向各个通信端口发送查询包, 对各个 网络设备的地址进行分配, 并且在分配的过程中在主控服务器端建立 了清晰的网络拓朴。 这样, 在具体数据包的传输过程中, 就可以直接 由主控服务器分配相应的通信链路即可 (因为其明晰整个网络的设备 拓朴), 不需要各个网络设备之间再进行路由协商 (现有 IP协议的方 案), 这样可以保证稳定的传输速率, 避免延时。
其次, 在一次业务数据的传输中, 各个该业务的各个数据包通过 相同的通信链路进行传输, 而不像现有 IP协议的方案, 每个数据包依 靠自行协商解决路由问题, 在数据包发出之前并不知道其会经过哪条 路径,即同一业务的两个数据包可能通过不同的路径传输至目标终端。 这样, 本发明与之相比, 可以保证稳定的传输速率, 避免延时。 附图说明
图 1是本发明的一种节点服务器的硬件结构示意图;
图 2是本发明的一种接入交换机的硬件结构示意图;
图 3是本发明的一种接入交换机入网过程的示意图:
图 4是本发明的一种节点服务器与接入交换机的连接示意图; 图 5是本发明的一种终端入网过程的示意图:
图 6是本发明的一种节点服务器、 接入交换机、 终端的连接示意 图;
图 7是本发明的一种节点交换机入网过程的示意图:
图 8是本发明的一种城域服务器、 节点交换机、 节点服务器的连 接示意图;
图 9是本发明的一种城域服务器、 节点交换机、 节点服务器、 终 端的连接示意图。
具体实施方式 为使本发明的上述目的、 特征和优点能够更加明显易懂, 下面结 合附图和具体实施方式对本发明作进一步详细的说明。
一、 以下简单介绍本发明的核心构思:
本专利发明人认为, 本发明实现全网品质保证的充分条件有以下 几点:
第一, IP 互联网核心理论中关于"尽力而为"( Best Efforts ) 的机 制必然导致网络流量不均匀和频繁的丟包。 实际上, TCP 协议正是利 用网络丟包状态来调节发送流量。
第二, IP 互联网核心理论中关于"储存转发"( Store & Forward ) 的机制在吸收本地突发流量的同时, 将造成下一个节点网络流量更大 的不均匀。
第三, IP 互联网核心理论中关于"检错重发"(Error Detection & Retransmission ) 的机制在同步视频通讯中, 将造成不可容忍的延时, 因此没有实用价值。
第四, 连续性的网络流量不均匀或突发流量必然导致周期性交换 机(路由器)丟包。
由此可见, 由于电脑文件突发流量本质是离散性, 没有后继, 上 述 IP 互联网核心理论曾经使互联网高效率传输文件。 但是, 面对连续 性同步流媒体传输中的品质保证, 昔日的功臣却成了破坏网络传输品 质的元凶。 我们已经从前面论述中得出结论, 任何资源预留、 优先级 别和轻载方案都不能从根本上解决同步流媒体的品质保证。
既然前述方法无一可行, 那么, 解决网络传输品质保证的出路在 哪里?
本专利发明人认为, 当前各种 QoS 方法, 都建立在一种错误的假 设上。 根据这种假设, QoS 的解决方法是为视讯流量提供优先处理的 特权。 但事实是, 由于不同媒体形式所需的网络流量极度不匀, 只要 有少数人使用视讯服务, 网络上的视讯流量将占据绝对主体。
如果换一角度看, 专门为大部分网络流量提供好的品质, 等效于 专门为少部分非视讯流量提供差的品质。 既然, 大部分网络流量必须 要求品质保证, 那么, 剩下少数不要求品质保证的业务流量也都给于 品质保证何尝不可。假设 1000 位旅客订飞机票时都要求头等舱, 只有 少数几位可以接受经济舱, 那么, 航空公司的自然措施是取消经济舱。 因为, 为了满足极少数差异化的经济舱, 航空公司所花的代价远大于 给这些旅客提供免费升舱。 实际上, 网络传输品质完全不保证, 或者 完全有保证都很简单, 难就难在部分保证和部分不保证, 尤其还不知 道这"部分"两字的界线划在哪里。 因此, 只要为全部网络业务都提供 品质保证, QoS 问题就不存在了。
IP 互联网初期好比是乡间小路,在民风淳朴的小镇不需要交通警 察。 但是到了繁华的大都市, 有些热闹路段的红绿灯和交通警察都控 制不了混乱局面, 出行赴约难以确定时间, 就像今天的 IP 互联网。
本发明好比是高速公路, 不需要警察和红绿灯, 水泥隔开的车道 和立交桥确保汽车在规定的道路行驶。 根据加州交通局的经验, 避免 高速公路堵车的办法是关闭入口匝道。
力口州高速公路的设计思路有三个特点:
•在公路入口匝道设置开关, 控制宏观车流量。
*保持车速稳定, 提高道路通车率。
•采用水泥结构的道路分隔和立交桥,而不是警察和红绿灯来规范 车辆行驶。
本发明实施例遵循电话网的原理, 采取类似上述高速公路的三项 措施:
*每条通路都计算和实测流量, 一旦流量接近饱和, 采取绕道或拒 绝新用户加入。
*严格均流发送, 本发明实施例 TV 能够在 90%重载流量下, 达到 百万分之一的丟包率。
•上行数据匹配和流量控制, 从结构上确保用户严格遵守交通规 则, 因为品质保证措施不可能指望用户自觉执行。
电脑文件与流媒体是两种截然不同的媒体形式, 处理方式相互排 斥。 本发明的网络理论和实践揭示了两项成果: •本发明实施例百倍于 IP 互联网的性价比优势
•发展高品质对称电视不干扰现有 IP 互联网业务的实施方法 尤其在大流量的骨干网络, 电脑文件和流媒体通过不同波长合用 一根光纤。 如果一定要统一到单一网络, 如接入网, 那么应该是将电 脑文件统一到视讯流媒体网络。 本发明实施例提供了完整的透明承载
IP互联网的解决方案。
将流媒体与文件分开处理只是第一步, 如何保证独立的流媒体网 络品质?
前面说过, PSTN 电话网络采用严格的同步机制, 当流量百分之 百用满之前, 不会出现网络阻塞现象。 从理论上讲, 多个均勾流合并 以后, 还是均匀流。 实践进一步证明, 在均匀流的前提下, 网络流量 可以接近于极限值, 而不发生丟包现象。 由于占据未来网络流量中九 成以上的视讯媒体流, 本身具备均勾流特征。 因此, 以视讯业务为主 要目标的本发明互联网品质保证的途径自然是消除信源流量不均匀, 尤其在意从根本上防止重载条件下网络交换机的丟包现象。
本发明实施例是用改良以太网建立面向连接的电路, 全网统一采 用定长度数据包。 只要改变发包时间间隔, 就可以得到任意带宽的媒 体流。 为了保证网络均流特征, 本发明互联网要求终端设计必须具备 均流能力。 但是, 在实际网络环境中, 显然不可能寄希望于用户自觉 遵守均流规定。 因此, 本发明实施例节点服务器向网络交换机发放通 行证, 只允许用户数据包在很细的时间精度下均勾通过。 对于符合规 定要求设计的用户终端, 通行证是完全透明的。
在上述前提下, 网络实践得出令人满意的结果, 本发明的交换机 能够在 90%带宽利用率的条件下, 获得优于百万分之一的重载丟包率。
综上所述, 品质保证是下一代网络不可回避的问题, 流媒体网络 是不同于传统电脑文件的另外一个物种。 因此, 修改 IP互联网去适应 视讯业务是没有前途的, 创立新的网络是唯一的出路。 二、 以下对本发明提出的一种新型网进行介绍: 新型网是一种集中控制的网络结构, 该网络可以是树型网、 星型 网、 环状网等等类型, 但在此基础上网络中需要有集中控制节点来控 制整个网络。
新型网分为接入网和城域网两部分。 接入网部分的设备主要可以 分为 3类: 节点服务器, 接入交换机, 终端 (包括各种机顶盒、 编码 板、存储器等)。其中,节点服务器是接入网中起集中控制功能的节点, 可控制接入交换机和终端。 节点服务器可直接与接入交换机相连, 也 可以直接与终端相连。 类似的, 城域网部分的设备也可以分为 3类: 城域服务器, 节点交换机, 节点服务器。 其中, 节点服务器即为接入 网部分的节点服务器, 即节点服务器既属于接入网部分, 又属于城域 网部分。 城域服务器是城域网中起集中控制功能的节点, 可控制节点 交换机和节点服务器。 城域服务器可直接连接节点交换机, 也可直接 连接节点服务器。 由此可见, 整个新型网络是一种分层集中控制的网 络结构, 而节点服务器和城域服务器下控制的网络可以是树型、 星型、 环状等各种结构。
1、 新型网设备分类
1. 1 本发明的新型网系统中的设备主要可以分为 3类: 服务器, 交换机(包括以太网网关), 终端 (包括各种机顶盒, 编码板, 存储 器等)。 新型网整体上可以分为城域网 (或者国家网、 全球网等)和接 入网。
1. 2 其中接入网部分的设备主要可以分为 3类: 节点服务器, 接 入交换机(包括以太网网关), 终端 (包括各种机顶盒, 编码板, 存 储器等)。
各接入网设备的具体硬件结构为:
节点服务器:
如图 1所示,主要包括网络接口模块 1 01、交换引擎模块 1 02、 CPU 模块 1 03、 磁盘阵列模块; 其中, 网络接口模块 101, CPU模块 103、 磁盘阵列模块 104进来 的包均进入交换引擎模块 102; 交换引擎模块 102对进来的包进行查 地址表 105的操作, 从而获得包的导向信息; 并根据包的导向信息把 该包存入对应的包緩存器 106的队列; 如果包緩存器 106的队列接近 满, 则丟弃; 交换引擎模块 102轮询所有包緩存器队列, 如果满足以 下条件进行转发: 1 ) 该端口发送緩存未满; 2 ) 该队列包计数器大于 零。 磁盘阵列模块 104主要实现对硬盘的控制, 包括对硬盘的初始化、 读写等操作; CPU模块 103主要负责与接入交换机、 终端 (图中未示 出)之间的协议处理, 对地址表 105 (包括下行协议包地址表、 上行 协议包地址表、 数据包地址表) 的配置, 以及, 对磁盘阵列模块 104 的配置。
接入交换机:
如图 2所示, 主要包括网络接口模块 (下行网络接口模块 201、 上行网络接口模块 202 )、 交换引擎模块 203和 CPU模块 204;
其中, 下行网络接口模块 201进来的包 (上行数据) 进入包检测 模块 205; 包检测模块 205检测包的目地地址 ( DA )、 源地址 ( SA )、 数据包类型及包长度是否符合要求, 如果符合, 则分配相应的流标识 符 ( stream-id), 并进入交换引擎模块 203, 否则丟弃; 上行网络接 口模块 202 进来的包 (下行数据) 进入交换引擎模块 203; CPU模块 204进来的数据包进入交换引擎模块 203;交换引擎模块 203对进来的 包进行查地址表 206的操作, 从而获得包的导向信息; 如果进入交换 引擎模块 203的包是下行网络接口往上行网络接口去的, 则结合流标 识符( stream-id)把该包存入对应的包緩存器 207的队列; 如果该包 緩存器 207的队列接近满, 则丟弃; 如果进入交换引擎模块 203的包 不是下行网络接口往上行网络接口去的, 则根据包的导向信息, 把该 数据包存入对应的包緩存器 207的队列; 如果该包緩存器 207的队列 接近满, 则丟弃。
交换引擎模块 203轮询所有包緩存器队列, 在本发明实施例中分 两种情形: 如果该队列是下行网络接口往上行网络接口去的, 则满足以下条 件进行转发: 1 )该端口发送緩存未满; 2 )该队列包计数器大于零; 3 ) 获得码率控制模块产生的令牌;
如果该队列不是下行网络接口往上行网络接口去的, 则满足以下 条件进行转发: 1 )该端口发送緩存未满; 2 )该队列包计数器大于零。
码率控制模块 208是由 CPU模块 2 04来配置的, 在可编程的间隔 内对所有下行网络接口往上行网络接口去的包緩存器队列产生令牌, 用以控制上行转发的码率。
CPU模块 2 04主要负责与节点服务器之间的协议处理, 对地址表 206的配置, 以及, 对码率控制模块 208的配置。
终端:
主要包括网络接口模块、 业务处理模块和 CPU模块; 例如, 机顶 盒主要包括网络接口模块、 视音频编解码引擎模块、 CPU 模块; 编码 板主要包括网络接口模块、 视音频编码引擎模块、 CPU 模块; 存储器 主要包括网络接口模块、 CPU模块和磁盘阵列模块。
1 . 3 城域网部分的设备主要可以分为 2类: 节点服务器, 节点交 换机, 城域服务器。 其中, 节点交换机主要包括网络接口模块、 交换 引擎模块和 CPU模块; 城域服务器主要包括网络接口模块、 交换引擎 模块和 CPU模块构成。
2、 新型网数据包定义
2. 1 接入网数据包定义
接入网的数据包主要包括以下几部分: 目的地址 (DA )、 源地址 ( SA )、 保留字节、 pay l oad ( PDU )、 CRC。
如下表所示, 接入网的数据包主要包括以下几部分:
DA SA Reserved Payload CRC 目的地址 (DA ) 由 8个字节 (byte ) 组成, 第一个字节表示数据 包的类型 (例如各种协议包、 组播数据包、 单播数据包等), 最多有 256 种可能, 第二字节到第六字节为城域网地址, 第七、 第八字节为 接入网地址;
源地址( SA )也是由 8个字节( byte )组成, 定义与目的地址( DA ) 相同;
保留字节由 2个字节组成;
pay load 部分根据不同的数据报的类型有不同的长度, 如果是各 种协议包的话是 64个字节,如果是单组播数据包话是 32 + 1024 = 1056
CRC有 4个字节组成, 其计算方法遵循标准的以太网 CRC算法。
2. 2 城域网数据包定义
城域网的拓朴是图型, 两个设备之间可能有 2种、 甚至 2种以上 的连接, 即节点交换机和节点服务器、 节点交换机和节点交换机、 节 点交换机和节点服务器之间都可能超过 2种连接。 但是, 城域网设备 的城域网地址却是唯一的,为了精确描述城域网设备之间的连接关系 , 在本发明实施例中引入参数: 标签, 来唯一描述一个城域网设备。
本说明书中标签的定义和 MPLS ( Mu l t i-Protoco l Labe l Swi t ch , 多协议标签交换) 的标签的定义类似, 假设设备 A和设备 B之间有两 个连接, 那么数据包从设备 A到设备 B就有 2个标签, 数据包从设备 B到设备 A也有 2个标签。 标签分入标签、 出标签, 假设数据包进入 设备 A的标签(入标签)是 0x0000 , 这个数据包离开设备 A时的标签 (出标签)可能就变成了 0x0001。 城域网的入网流程是集中控制下的 入网过程, 也就意味着城域网的地址分配、 标签分配都是由城域服务 器主导的, 节点交换机、 节点服务器都是被动的执行而已, 这一点与 MPLS的标签分配是不同的, MPLS的标签分配是交换机、 服务器互相协 商的结果。
如下表所示, 城域网的数据包主要包括以下几部分:
Figure imgf000040_0001
即目的地址 (DA)、 源地址 (SA)、 保留字节 (Reserved), 标签、 payload ( PDU ), CRC。 其中, 标签的格式可以参考如下定义: 标签是 32bit, 其中高 16bit保留, 只用低 16bit, 它的位置是在数据包的保 留字节和 payload之间。
3、 新型网的实现
以下讨论节点服务器与接入交换机、 节点服务器与终端的入网流 程。 为了简化设计, 定义在接入网中的数据包类型为 4种, 分别是: 下行协议包 (由节点服务器发往接入交换机、 终端的协议包); 上行协议包 (由接入交换机、 终端回应给节点服务器的协议包); 单播数据包;
组播数据包;
接入网的地址总共是 16bit, 所以总共可以接入的接入交换机、 终端数为 65536, 假设下行协议包的数据报类型为" 1000 0000" (二进 制),也就是 0x80(十六进制),上行协议包的数据报类型为 "00001000" (二进制),也就是 0x08(十六进制),单播数据包的数据报类型为 "0001 0000" (二进制), 也就是 0x10 (十六进制), 组播数据包的数据报类 型为" 01111000" (二进制), 也就是 0x78 (十六进制), 通过合并同类 项, 可以 4巴 8bit长的地址表映射为 2bit长的地址表, 例如:
"1000 0000"=> "00", 下行协议包的地址表, 在本发明实施例中 定义为 0号表;
"0000 1000"=> "01", 上行协议包的地址表, 在本发明实施例中 定义为 1号表;
"0001 0000"=> "10", 单播数据包的地址表, 在本发明实施例中 定义为 2号表; "0111 1000"=> "11", 组播数据包的地址表, 在本发明实施例中 定义为 3号表;
结合 16bit的接入网地址, 在实际中只需 4张 64K = 4 X 65536 , 也就是 256K的地址表,地址表的输出就表示数据包导向的端口。例如, 其中的一种接入交换机 BX-008, 它具有 1个上行的百兆网口, 8个下 行百兆网口, 1个 CPU模块接口。 如果 8个下行百兆网口依次定义为 0 号端口到 7号端口, CPU模块接口定义为 8号端口, 1个上行的百兆网 口定义为 9号端口, 则总共需要 256K X lObit的地址表, 例如地址表 的输出为" 00 0000 0001"表示数据包导向的 0号端口, "11 0000 0000" 表示数据包导向的 8号、 9号端口, 以此类推。
殳设 9号端口进来一个数据包它的目的地址( DA )是 0x8056 0x1500 0x00000x55aa, 那么它的数据包类型为 0x80, 接入网地址为 0x55aa, 才艮据查表规则这时查 0号表, 即地址为" 00 0101 0101 1010 1010", 此地址对应的地址表的输出为" 01 0000 0000", 表示数据包导向 8号 端口。
3.1接入网设备的入网流程
3.1.1接入交换机的入网过程:
首先每台允许入网的接入交换机都必须在节点服务器里注册, 没 有注册的接入交换机无法入网。 如图 3所示, 所述接入交换机入网的 过程涉及以下步骤:
51、 节点服务器向每个端口发送查询包, 接入交换机收到查询包 后发送应答包, 应答包中包含当前接入交换机的注册信息;
52、 节点服务器收到接入交换机发出的应答包后, 就知道哪个端 口下接了一台接入交换机, 然后在节点服务器内部的注册信息表里找 到该接入交换机信息, 向该接入交换机发送入网命令 (告知接入网地 址), 该接入交换机收到入网命令后就入网了, 同时向节点服务器发送 入网命令应答; S 3、 节点服务器收到接入交换机发出的入网命令应答就知道该接 入交换机已经入网了, 以后定时向这个端口发送状态查询包, 检查这 台接入交换机是否正常工作, 同时还要向该接入交换机的下行端口发 送端口查询包, 检查是否有其他接入网设备接在该接入交换机下面。 如果当前接入交换机正常工作, 收到设备状态查询指令后会发送状态 查询应答给节点服务器。 当节点服务器一段时间之内没有收到状态查 询应答, 就认为该接入交换机已经被移出网络, 不再发送状态查询包, 而继续向本端口发送查询包。
3.1.2节点服务器与接入交换机在入网过程中的交互示例: 为了简单起见, 假设节点服务器并未和节点交换机相连, 忽略城 域网的入网过程, 为了方便讨论, 假设此节点服务器有 8个下行百兆 网口依次定义为 0号端口到 Ί号端口, 1个 CPU模块接口定义为 8号 端口, 1个磁盘阵列模块接口定义为 9号端口, 1个上行千兆光口定义 为 10号端口, 此节点服务器型号为 MSS-400。 如图 4所示, MSS-400 的 0 号端口接了 1 台 BX-008-0 , BX-008-0 的 1 号端口接了 1 台 BX-008- l o
51、 MSS-400服务器上电后初始化硬件,获得默认城域网地址(假 设为 0x00 0x0000 0x0000 ), 从硬盘导入配置文件到 CPU内存 (例如 交换机的注册信息、终端的注册信息等等), MSS-400服务器配置自己 的接入网地址为 0x0000;
52、 MSS-400服务器初始化 0、 1、 2、 3号表:
•配置 0号表为 "000 0000 0000" , 即所有查询包传送关闭; •配置 1号表为 "001 0000 0000" , 即所有的应答包导向 CPU;
•配置 2号、 3号表为 "000 0000 0000" , 即所有单组播数据包传送 关闭;
53、 MSS-400服务器知道自己有 8个下行端口, 所以它配置 8个 0号表的表项分别为:
• "00 0000 0000 0000 0001" => "000 0000 0001" ,即目的地址( DA ) 畢^ o ' ^ ^ o-800-xa 、9S
^^mf - ^ ^ ',,οοοο oooo ooo„^f e、备 ζ ¾·
• o ^ 丁 ^者
'"000000000ΐ,, "xxxx xxxx xxxx xxxx I0"H ΐ ¾·
iildD ^者 畢 '"oooo oooo io,, "χχχχ χχχχ χχχχ χχχχ oo"H 0 ¾·
'^ "^^^^^丁 讲 ΐ-800-xa 、o-800-xa 、ss
'( 0000X00000X00000X00000X0 VS ) ^ ^畢 ^ 8000X0 ooooxo 0000X00008X0
Figure imgf000043_0001
ooooxo 0008X0 ^000X00000X00000X00008X0 ^000X00000X00000X00008X0 、 OOO^O 0000X0 OOOOXO ΟΟΟδΧΟ ^OOO^O 0000X00000X00008X0 000X0 ooooxo ooooxo 0008X0 ( να )
Figure imgf000043_0002
oot-ssw
• ^^r L ^者 ^畢 8000X00000X00000^00008^0 ^ (YQ) Tf?i^ ^ '"0000 000 ΐ 000,, <= ,,000 ΐ oooo oooo 0000 00" ·
-
Figure imgf000043_0003
( va ) ίψ^ @ tig '"00000010000„ <= ,,ΠΐΟ OOOO OOOO OOOO 00" ·
■ o
Figure imgf000043_0004
( va ) ίψ^ @ tig '"00000100000„ <= ,,ΟΠΟ OOOO OOOO OOOO 00" ·
■ o ^ - ^畢 0000X00000X00008X0 (va) ίψ^ @ tig '"0000 1000 000,, <= ,,ΐθΐθ 0000 0000 0000 00" ·
■ o ^ - £ ^-^ ^畢 toooxo ooooxo οοοοχο 0008X0 (va) ίψ^ ^ 、0001 0000 000,, <= ,,ΟΟΐΟ 0000 0000 0000 00" ·
-
Figure imgf000043_0005
0000X00000X00008X0 ( va ) ίψ^ @ tig '"0010 OOOO 000„ <= ,,ΠΟΟ OOOO OOOO OOOO 00" ·
■ o ^ - ΐ ^畢 Z000X00000X0 OOOOXO 0008X0 ( va ) ίψ^ @ tig '"01000000 ΟΟΟ,, <= ,,ΟΐΟΟ OOOO OOOO OOOO 00,,·
■ o
Figure imgf000043_0006
-IV- l66LL0/U0ZSiD/lDd TCS910/Z10Z OAV 询包被接收至其 CPU模块, 由 CPU解析该查询包后生成应答包 (应 答包中包含本交换机的注册信息), 并发送给 MSS-400服务器, 该应 答包的 DA是 0x0800 0x0000 0x0000 0x0000 , SA是 0x0000 0x0000 0x0000 0x0001 ;
57、 MSS-400服务器收到 BX-008-0 交换机发出的应答包后, 对 比应答包的源地址(SA )及设备类型就知道它的 0号端口下接了一台 接入交换机, 然后在节点服务器内部的注册信息表里找到这台接入交 换机的信息, 向这台接入交换机发送入网命令(告知其接入网地址为 0x0001 );
58、 BX-008-0交换机收到入网命令后, 知道自己的接入网地址是 0x0001就入网了, 于是配置其 0号表" 00 0000 0000 0000 0001" 为" 01 0000 0000", 并 0号表其余表项配置为 "00 0000 0000", 即只有本交换 机的下行协议包导入 CPU, 同时向 MSS-400服务器发送入网命令应答
(入网命令应答包);
59、 MSS-400服务器收到 BX-008-0 交换机发出的入网命令应答 就知道 BX-008-0交换机已经入网了,以后每秒钟向这个端口发送设备 状态查询指令, 检查 BX-008-0 交换机是否正常工作, 同时还要向 BX-008-0交换机的下行端口发送端口下行协议包, 检查是否有其它接 入网设备接在本接入交换机下面, 在这种情况下, MSS-400服务器会 在其 0号表中做如下配置:
• "00 0000 0000 0000 1001" => "000 0000 0001", 即目的地址 ( DA )是 0x8000 0x0000 0x0000 0x0009的端口下行协议包导向 0号 端口;
• "00 0000 0000 0000 1010" => "000 0000 0001", 即目的地址 ( DA )是 0x8000 0x0000 0x0000 0x000a的端口下行协议包导向 0号端 口;
• "00 0000 0000 0000 1011" => "000 0000 0001",即目的地址( DA ) 是 0x8000 0x0000 0x0000 0x000b的端口下行协议包导向 0号端口;
• "00 0000 0000 0000 1100" => "000 0000 0001",即目的地址( DA ) 是 0x8000 0x0000 0x0000 0x000c的端口下行协议包导向 0号端口;
• "00 0000 0000 0000 1101" => "000 0000 0001",即目的地址( DA ) 是 0x8000 0x0000 0x0000 OxOOOd的端口下行协议包导向 0号端口;
• "00 0000 0000 0000 1110" => "000 0000 0001 ",即目的地址( DA ) 是 0x8000 0x0000 0x0000 0x000e的端口下行协议包导向 0号端口;
• "00 0000 0000 0000 1111" => "000 0000 0001",即目的地址( DA ) 是 0x8000 0x0000 0x0000 0x000f的端口下行协议包导向 0号端口;
• "00 0000 0000 0001 0000" => "000 0000 0001 ", 即目的地址 ( DA )是 0x8000 0x0000 0x0000 0x0010的端口下行协议包导向 0号 端口;
MSS-400 服务器会通过包括端口分配信息的端口分配包, 通知 BX-008-0交换机在其 0号表中做如下配置:
• "00 0000 0000 0000 1001" => "00 0000 0001", 即目的地址 ( DA ) 是 0x8000 0x0000 0x0000 0x0009的端口下行协议包导向 0号端口;
• "00 0000 0000 0000 1010" => "00 0000 0010", 即目的地址 ( DA ) 是 0x8000 0x0000 0x0000 0x000a的端口下行协议包导向 1号端口;
• "00 0000 0000 0000 1011" => "00 0000 0100", 即目的地址 ( DA ) 是 0x8000 0x0000 0x0000 0x000b的端口下行协议包导向 2号端口;
• "00 0000 0000 0000 1100" => "00 0000 1000", 即目的地址 ( DA ) 是 0x8000 0x0000 0x0000 0x000c的端口下行协议包导向 3号端口;
• "00 0000 0000 0000 1101" => "00 0001 0000", 即目的地址 ( DA ) 是 0x8000 0x0000 0x0000 OxOOOd的端口下行协议包导向 4号端口;
• "00 0000 0000 0000 1110" => "00 0010 0000", 即目的地址 ( DA ) 是 0x8000 0x0000 0x0000 0x000e的端口下行协议包导向 5号端口;
• "00 0000 0000 0000 1111 " => "00 0100 0000", 即目的地址 ( DA ) 是 0x8000 0x0000 0x0000 OxOOOf的端口下行协议包导向 6号端口;
• "00 0000 0000 0001 0000" => "00 1000 0000", 即目的地址 ( DA ) 是 0x8000 0x0000 0x0000 0x0010的端口下行协议包导向 7号端口;
S 10、 MSS-400服务器发目的地址( DA )是 0x8000 0x0000 0x0000 0x0009、 0x8000 0x0000 0x0000 0x000a, 0x8000 0x0000 0x0000 0x000b , 0x8000 0x0000 0x0000 0x000c , 0x8000 0x0000 0x0000 OxOOOd, 0x8000 0x0000 0x0000 OxOOOe , 0x8000 0x0000 0x0000 OxOOOf, 0x8000 0x0000 0x0000 0x0010 的端口下行协议包 (SA都为 0x0000 0x0000 0x0000 0x0000 ), 根据 MSS-400服务器中 0号表的配置, 端口下行协议包会 依次导向 MSS-400服务器 0号端口, 根据 BX-008-0交换机中 0号表 的配置, 端口下行协议包会依次导向 BX-008-0交换机 0到 Ί号端口; S l l、 BX-008- 1交换机从 BX-008-0交换机的 1号端口收到端口下 行协议包(即目的地址是 0x8000 0x0000 0x0000 0x000a的端口下行协 议包)后发送端口上行协议包 (该端口上行协议包中包含本交换机的 注册信息),包的 DA是 0x0800 0x0000 0x0000 0x0000 , SA是 0x0000 0x0000 0x0000 0x000a;
512、 MSS-400服务器收到 BX-008- 1交换机发出的端口上行协议 包后, 对比上行协议包的源地址 (SA ) 及设备类型就知道 BX-008-0 的 1号端口下接了一台接入交换机, 然后在服务器内部的注册信息表 里找到这台接交换机信息, 并向该接入交换机发送入网命令(告知其 接入网地址为 0x000a );
513、 BX-008-1 交换机收到入网命令后, 知道自己的接入网地址 是 0x000a就入网了, 配置其 0号表" 00 0000 0000 0000 1010" 为" 01 0000 0000" , 0号表其余表项配置为 "00 0000 0000" , 即只有本交换机 的下行协议包导入 CPU, 同时向服务器发送入网命令应答;
514、 MSS-400 服务器收到交换机发出的入网命令应答就知道 BX-008- 1交换机已经入网了, 以后每秒钟向这个端口发送设备状态查 询指令, 检查 BX-008- 1 交换机是否正常工作, 同时还要向 BX-008- 1 交换机的下行端口发送端口下行协议包, 检查是否有其他接入网设备 接在本接入交换机下面。 如果当前接入交换机正常工作, 收到设备状 态查询指令后会发送状态查询应答给服务器。 当服务器 6秒之内没有 收到状态查询应答, 就认为这台接入交换机已经被移出网络, 不再发 送设备状态查询指令, 继续向本端口发送查询包。 3.1.3终端的入网过程:
首先每台允许入网的终端都必须在节点服务器中注册, 没有注册 的终端无法入网。 如图 5所示, 所述终端入网的过程涉及以下步骤:
Sl、 节点服务器向每个端口发送查询包, 终端收到查询包后发送 应答包, 该应答包中包含终端的注册信息;
52、 节点服务器收到终端发出的应答包后就知道哪个端口下接了 哪种终端(机顶盒, 编码板还是存储器), 然后在节点服务器内部的注 册信息表里找到该终端的信息, 向该终端发送入网命令(告诉终端的 接入网地址), 终端收到入网命令后就入网了, 同时向节点服务器发送 入网命令应答;
53、 节点服务器收到终端发出的入网命令应答就知道本终端已经 入网了, 以后定时向这个端口发送状态查询包, 检查终端是否正常工 作。 如果终端正常工作, 收到状态查询包后会发送状态查询应答给节 点服务器。 当节点服务器一段时间之内没有收到状态查询应答, 就认 为本终端已经被移出网络, 不再发送状态查询包, 继续向本端口发送 查询包。
3.1.4节点服务器与接入交换机、 终端在入网过程中的交互示例: 接入网的地址可以设置为 16bit, 所有接入网设备都有唯一的接入 网地址 (包括机顶盒、 接入交换机、 存储器, 甚至节点服务器本身)。 为方便管理所有接入网设备的接入网地址,在节点服务器的 CPU模块 中可以维护一张地址信息表, 该表的大小为 2的 16次方, 也即 64K, 每个表的表项由如下构成:
1 )地址占用描述符: "00"表示此地址未用, "01"表示此地址待用 (节点服务器用此地址发出了端口下行协议包, 但未收到入网上行协 议包), "10" 表示此地址已用 (节点服务器收到入网上行协议包后设 置);
2 )设备描述符: 例如, "000000"表示节点服务器, "000001"表示 其中一种接入交换机 BX-008 , "000010" 表示其中一种存储器, "000011"表示其中一种终端;
3 )设备资源描述信息: 例如, 该设备是接入交换机的话, 它的网 络端口连接的设备的接入网地址, 它的各个网络端口的上下行流量计 数; 如果该设备是存储器的话, 它的网络端口连接的设备的接入网地 址, 它的读写通道的计数以及网络端口的上下行流量计数; 等等, 所 有这些信息是为了服务流程提供决策依据, 而且每次的服务流程中都 会修改这些信息。
如图 6所示,假设一台节点服务器 MSS-400 , 它的 0号端口接了 一台接入交换机 BX-008-0 , 它的 1 号端口接了一台接入交换机 BX-008-1 , BX-008-0的 0号端口接了一台机顶盒 STB-0, BX— 008-1 的 1号端口接了一台机顶盒 STB-1。
51、 MSS-400服务器上电后初始化硬件,获得默认城域网地址(假 设为 0x00 0x0000 0x0000 ), 从硬盘导入配置文件到 CPU内存 (例如 交换机的注册信息、终端的注册信息等等), MSS-400服务器初始化地 址信息表, 全部清零(表示所有地址未用), MSS-400服务器配置自己 的接入网地址为 0x0000, 也即地址信息表的第 0x0000项被配置成如 下:
•地址占用描述符: "10" 表示此地址已用;
•设备描述符: "000000"表示节点服务器;
•设备资源描述信息:此节点服务器有 8个下行百兆网口依次定义 为 0号端口到 7号端口, 1个 CPU模块接口定义为 8号端口, 1个磁 盘阵列接口定义为 9号端口, 1个上行千兆光口定义为 10号端口, 此 节点服务器型号为 MSS-400, 它的网络端口连接的设备的接入网地址 未分配, 它的各个网络端口的上下行流量计数为 0;
地址信息表下一个可用地址为 0x0001 ;
52、 MSS-400服务器初始化 0、 1、 2、 3号表:
•配置 0号表为 "000 0000 0000", 即所有下行协议包传送关闭; •配置 1号表为 "001 0000 0000", 即所有上行协议包导向 CPU; •^ : ?^ ¾··
Figure imgf000049_0001
'( 0000X00000X00000X00000X0 VS ) ^ ^畢 ^ 8000X0 οοοοχο 0000X00008X0
Figure imgf000049_0002
οοοοχο 0008X0 ^000X00000X00000X00008X0 ^OOOXO 0000X00000X00008X0 、 OOO^O 0000X0 OOOOXO ΟΟΟδΧΟ ^OOO^O 0000X00000X00008X0 000X0 0000X00000X00008X0 ( V(I ) Τί?ί^ @ ί,Ψ ^ 00t7-SSW ' S
• ^^r L ^者 ^畢 8000X00000X00000^00008^0 ^ (YQ) Tf?i^ ^ '"0000 000 ΐ 000,, <= ,,000 ΐ 0000 0000 0000 00" ·
-
Figure imgf000049_0003
( va ) ίψ^ @ tig '"00000010000„ <= ,,ΠΐΟ OOOO OOOO OOOO 00" ·
■ o
Figure imgf000049_0004
( va ) ίψ^ @ tig '"00000100000„ <= ,,ΟΠΟ OOOO OOOO OOOO 00" ·
■ o ^ - ^畢 0000X00000X00008X0 (va) ίψ^ @ tig '"0000 1000 000,, <= ,,ΐθΐθ 0000 0000 0000 00" ·
■ o ^ - £ ^-^ ^畢 toooxo ooooxo οοοοχο 0008X0 (va) ίψ^ ^ 、0001 0000 000,, <= ,,ΟΟΐΟ 0000 0000 0000 00" ·
-
Figure imgf000049_0005
0000X00000X00008X0 ( va ) ίψ^ @ tig '"00100000000„ <= ,,ΠΟΟ OOOO OOOO OOOO 00" ·
■ o ^ - ΐ ^畢 Z000X00000X0 OOOOXO 0008X0 ( va ) ίψ^ @ tig '"01000000 ΟΟΟ,, <= ,,ΟΐΟΟ OOOO ΟΟΟΟ ΟΟΟΟ 00,,·
■ ο
Figure imgf000049_0006
( Va )Tf?i^ @ ^ '"10000000000„ <= ,,ΐΟΟΟ OOOO OOOO OOOO 00,,·
: ^^^ 0 ^ 8 ^¾ ¾ ^ ^ 'ΐ oooxo ^
Figure imgf000049_0007
e、备 ζ ¾·
-LP- l66LL0/U0ZSiD/lDd TCS910/Z10Z OAV •设备资源描述信息: 不做修改;
地址信息表的下一个可用地址为 0x0009;
55、 BX-008-0 , BX-008-1交换机上电后初始化硬件,
•配置其 0号表,,00 xxxx xxxx xxxx xxxx" 为" 01 0000 0000", 即 所有下行协议包包导向 CPU;
•配置其 1号表,,01 xxxx xxxx xxxx xxxx" 为" 10 0000 0000", 即所 有的上行协议包导向上行的百兆网口;
•配置其 2号、 3号表为 "00 0000 0000", 即所有单组播数据包传 送关闭;
56、 BX-008-0交换机收到查询包后, 根据其 0号表的配置, 该查 询包被接收至其 CPU模块, 由 CPU模块解析该查询包并生成应答包
(该应答中包含本接入交换机的注册信息)发送给 MSS-400服务器, 包的 DA是 0x0800 0x0000 0x0000 0x0000 , SA 0x0000 0x0000 0x0000 0x0001 ;
57、 MSS-400服务器收到 BX-008-0 交换机发出的应答包后, 对 比应答包的源地址 (SA )、 及设备类型就知道其 0 号端口下接了一台 接入交换机, 然后在节点服务器内部的注册信息表里找到这台接入交 换机的信息, 向该接入交换机发送入网命令 (告知其接入网地址为 0x0001 );
58、 BX-008-0交换机收到入网命令后, 知道自己的接入网地址是 0x0001就入网了, 于是配置其 0号表" 00 0000 0000 0000 0001" 为" 01 0000 0000", 0号表其余表项配置为 "00 0000 0000" , 即只有本交换机 的下行协议包导入 CPU, 同时向服务器发送入网命令应答;
59、 MSS-400服务器收到 BX-008-0 交换机发出的入网命令应答 就知道 BX-008-0交换机已经入网了,于是将服务器内部的地址信息表 的第 0x0001项被配置成:
•地址占用描述符: "10" 表示此地址已用;
•设备描述符: "000001"表示其中一种接入交换机 BX-008;
•设备资源描述信息:此接入交换机有 8个下行百兆网口依次定义 为 0号端口到 7号端口, 1个 CPU模块接口定义为 8号端口, 1个上 行百兆网口定义为 9号端口, 此接入交换机型号为 BX-008 , 它的上行 网络端口连接的设备的接入网地址是 0x0000 (即 MSS-400 ), 下行网 络端口连接的设备的接入网地址未分配, 它的各个网络端口的上下行 流量计数为 0;
以后每秒钟向这个端口发送设备状态查询指令, 检查 BX-008-0 交换机是否正常工作,同时还要向 BX-008-0交换机的下行端口发送端 口下行协议包, 检查是否有其他接入网设备接在本接入交换机下面。 在这种情况下, MSS-400服务器会在其 0号表中做如下配置:
• "00 0000 0000 0000 1001" => "000 0000 0001", 即目的地址 ( DA )是 0x8000 0x0000 0x0000 0x0009的端口下行协议包导向 0号 端口;
• "00 0000 0000 0000 1010" => "000 0000 0001", 即目的地址 ( DA )是 0x8000 0x0000 0x0000 0x000a的端口下行协议包导向 0号端 口;
• "00 0000 0000 0000 1011" => "000 0000 0001",即目的地址( DA ) 是 0x8000 0x0000 0x0000 0x000b的端口下行协议包导向 0号端口;
• "00 0000 0000 0000 1100" => "000 0000 0001",即目的地址( DA ) 是 0x8000 0x0000 0x0000 0x000c的端口下行协议包导向 0号端口;
• "00 0000 0000 0000 1101" => "000 0000 0001",即目的地址( DA ) 是 0x8000 0x0000 0x0000 0x000d的端口下行协议包导向 0号端口;
• "00 0000 0000 0000 1110" => "000 0000 0001",即目的地址( DA ) 是 0x8000 0x0000 0x0000 0x000e的端口下行协议包导向 0号端口;
• "00 0000 0000 0000 1111" => "000 0000 0001",即目的地址( DA ) 是 0x8000 0x0000 0x0000 0x000f的端口下行协议包导向 0号端口;
• "00 0000 0000 0001 0000" => "000 0000 0001", 即目的地址 ( DA )是 0x8000 0x0000 0x0000 0x0010的端口下行协议包导向 0号 端口;
MSS-400 服务器会通过包括端口分配信息的端口分配包, 通知 BX-008-0交换机在其 0号表中做如下配置:
• "00 0000 0000 0000 1001" => "00 0000 0001", 即目的地址 ( DA ) 足 0x8000 0x0000 0x0000 0x0009的端口下行协议包导向 0号端口;
• "00 0000 0000 0000 1010" => "00 0000 0010", 即目的地址 ( DA ) 足 0x8000 0x0000 0x0000 0x000a的端口下行协议包导向 1号端口;
• "00 0000 0000 0000 1011" => "00 0000 0100", 即目的地址 ( DA ) 足 0x8000 0x0000 0x0000 0x000b的端口下行协议包导向 2号端口;
• "00 0000 0000 0000 1100" => "00 0000 1000", 即目的地址 ( DA ) 足 0x8000 0x0000 0x0000 0x000c的端口下行协议包导向 3号端口;
• "00 0000 0000 0000 1101" => "00 0001 0000", 即目的地址 ( DA ) 足 0x8000 0x0000 0x0000 0x000d的端口下行协议包导向 4号端口;
• "00 0000 0000 0000 1110" => "00 0010 0000", 即目的地址 ( DA ) 足 0x8000 0x0000 0x0000 0x000e的端口下行协议包导向 5号端口;
• "00 0000 0000 0000 1111" => "00 0100 0000", 即目的地址 ( DA ) 足 0x8000 0x0000 0x0000 0x000f的端口下行协议包导向 6号端口;
• "00 0000 0000 0001 0000" => "00 1000 0000", 即目的地址 ( DA ) 足 0x8000 0x0000 0x0000 0x0010的端口下行协议包导向 7号端口; S 10、 MSS-400服务器发目的地址( DA )是 0x8000 0x0000 0x0000 0x0009、 0x8000 0x0000 0x0000 0x000a, 0x8000 0x0000 0x0000 0x000b , 0x8000 0x0000 0x0000 0x000c, 0x8000 0x0000 0x0000 0x000d, 0x8000 0x0000 0x0000 OxOOOe, 0x8000 0x0000 0x0000 OxOOOf, 0x8000 0x0000 0x0000 0x0010 的端口下行协议包 (SA都为 0x0000 0x0000 0x0000 0x0000 ), 根据 MSS-400服务器中 0号表的配置, 所述端口下行协议 包会依次导向 MSS-400服务器 0号端口, 根据 BX-008-0交换机中 0 号表配置,端口下行协议包会依次导向 BX-008-0交换机 0到 7号端口; 并且, MSS-400服务器中的地址信息表的第 0x0009至 0x0010项被配 置成:
•地址占用描述符: "01"表示此地址待用;
•设备描述符: 不做修改; •设备资源描述信息: 不做修改;
下一个可用地址为 0x0011 ;
S l l、 STB-0从 BX-008-0交换机的 0号端口收到端口下行协议包 (即目的地址是 0x8000 0x0000 0x0000 0x0009的端口下行协议包 )收 到端口下行协议包后发送端口上行协议包 (包含本终端的注册信息), 包的 DA是 0x0800 0x0000 0x0000 0x0000 , SA 0x0000 0x0000 0x0000 0x0009 (交换机的 0号端口);
512、 MSS-400服务器收到 STB-0交换机发出的端口上行协议包 后, 对比上行协议包的源地址( SA )及设备类型就知道 BX-008-0的 0 号端口下接了一台终端, 然后在服务器内部的注册信息表里找到终端 信息, 向终端发送入网命令(告诉终端的接入网地址为 0x0009 );
513、 STB-0 收到入网命令后, 知道自己的接入网地址是 0x0009 就入网了, 同时向服务器发送入网命令应答;
514、 MSS-400 服务器收到 STB-0 发出的入网命令应答就知道 STB-0交换机已经入网了, 于是将地址信息表的第 0x0009项配置成:
•地址占用描述符: "10" 表示此地址已用;
•设备描述符: "000011"表示其中一种终端;
•设备资源描述信息:此终端有视音频编解码引擎, 1个百兆网口, 此终端型号为 STB , 它的网络端口连接的设备的接入网地址是 0x0001 (即 BX-008-0 ), 它的网络端口的上下行流量计数为 0;
地址信息表的第 0x0001项被配置成:
•地址占用描述符: 不做修改;
•设备描述符: 不做修改;
•设备资源描述信息:此接入交换机有 8个下行百兆网口依次定义 为 0号端口到 7号端口, 1个 CPU模块接口定义为 8号端口, 1个上 行百兆网口定义为 9号端口, 此接入交换机型号为 BX-008 , 它的上行 网络端口连接的设备的接入网地址是 0x0000 (即 MSS-400 ), 下行网 络端口 0连接的设备的接入网地址是 0x0009, 其余未分配, 它的各个 网络端口的上下行流量计数为 0; 以后 MSS-400服务器每秒钟向这个端口发送设备状态查询指令, 检查 STB-0是否正常工作,当服务器 6秒之内没有收到状态查询应答, 就认为 STB-0已经被移出网络, 不再发送设备状态查询指令, 继续向 本端口发送查询包。
参照上述第 S6— S 14步骤, BX-008-1也会入网, 获得其接入网地 址为 0x0002; STB-1也会入网, 获得其接入网地址为 0x0012。
3.1.5接入网设备入网过程中的数据格式定义:
PDU 是用户终端和服务器的信息交互方式, 两者使用原始套接字 ?de )传递 PDU, 其数据格式如下:
Figure imgf000054_0001
系统消息 (PDU ) 定义
一. 端口查询
1、 端口查询指令, 由服务器发出, 32W短信令:
Figure imgf000054_0002
2、 端口查询应答指令, 32W短信令: 机顶盒收到 8A01发给服务器的应答
Figure imgf000055_0001
交换机收到 8A01发给服务器的应答
Figure imgf000055_0002
编码板收到 8A01发给服务器的应答
字段号 长度 代码 说明 0 1W 8A06 编码板端口查询应答指令
1 1W 设备类型 ( 0x5131 )
2-4 3W 设备标识 (每个都不同)
5-8 4W 网络地址 (从 8A01获取 )
9-31 23W 0000 填充 设备标识: 暂时可以取值 0x5131 0201 000X ( X = 0-f )
3、 入网指令, 由服务器发出, 32W短信令:
Figure imgf000056_0001
4、 入网确认指令, 32W短信令:
字段
长度 代码 说明
0 1W 8A12 入网确认指令
1 1W 设备类型
2-4 3W 设备标识
5-10 6W 设备型号 (辅助信息, ASCII码)
11-13 3W 用户号码 (从 8al l获取) 14 1W 用户分机号 (从 8al l获取)
15-18 4W 网络地址 (从 8a01获取)
19-31 13W 0000 填充
5、 状态查询指令, 由服务器发出, 32W短信令:
给交换机的状态查询指令
Figure imgf000057_0001
给机顶盒的状态查询指令 字段
长度 代码 说明 号
0 1W 8A31 机顶盒状态查询指令
1-4 4W 网络地址
5-7 3W 实时时钟 (由服务器维护) 8-10 3W 机顶盒号码
11-31 21W 0000 填充
6、 状态查询应答指令, 32W短信令:
交换机的状态查询应答
Figure imgf000058_0001
机顶盒的状态查询应答 字段号 长度 代码 说明
0 1W 8A23 用户终端状态查询应答指令
1 1W 设备类型
2-4 3W 设备标识
5-7 3W 用户号码
8 1W 用户终端分机号码
9 1W 终端设备状态 (正常 =0 ) 10 1W 用户终端环境温度
11 1W 终端所接设备的类型
终端所接设备的开关状态 0=
12 1W
下电 1=上电
13-30 18W 同 11-12
31 lw CRC 编码板的状态查询应答
Figure imgf000059_0001
3.2节点服务器的入网过程:
3. 2. 1 城域网实现
为了简化设计, 定义在城域网中的数据包的类型为 4种, 分别是: 參城域查询标签包 (由城域服务器发往节点交换机、 节点服务器 的带标签的协议包);
參城域应答标签包 (由节点交换机、 节点服务器回应给城域服务 器的带标签的协议包);
*单播标签数据包 (由节点服务器在单组播数据包上添加标签构 成); *组播标签数据包 (由节点服务器在单组播数据包上添加标签构 成);
城域网的地址总共是 40bit, 本文把它分成 3层, 分别是 8bit 、 16bit、 16bit, 并依次定义为国家网、 广域网、城域网。 同一城域网、 同一接入网的终端之间的数据传送由这接入网的节点服器控制。
假设终端 STB_0位于接入网 A, 终端 STB_1位于接入网 B, 接入网 A、 接入网 B属于同一城域网 C。 实现过程如下:
51、 STB_0发出请求和 STB_1进行可视电话;
52、 接入网 A的节点服务器根据 STB_1的号码查到不属于接入网
A, 然后它向城域网 C的城域服务器发出查询;
53、 城域网 C的城域服务器根据 STB_1的号码查到它属于接入网
B, 于是它向接入网 B的节点服务器发出查询;
54、 接入网 B的节点服务器根据 STB_1的号码查到 STB_1在接入 网 B, 接入网 B的节点服务器发送呼叫菜单给 STB_1;
55、 STB_1 可以选择接受或者拒接, 应答发送给接入网 B的节点 服务器, 这里假设选择接受;
56、 接入网 B的节点服务器向城域网 C的城域服务器发出应答;
57、 城域网 C的城域服务器向接入网 A的节点服务器发出应答;
58、 接入网 A的节点服务器向 STB_0发出应答。
从上述描述可以得知, 对于本发明实施例而言, 终端只和本接入 网的节点服务器交互, 节点服务器和本城域网的城域服务器交互, 以 此类推, 城域服务器和本广域网的广域服务器交互。
假设城域查询包的数据报类型为" 1001 0000" (二进制), 也就是 0x90 (十六进制 ), 城域应答包的数据报类型为 "00001001" (二进制 ), 也就是 0x09(十六进制),单播标签数据包的数据报类型为" 0001 0000" (二进制), 也就是 0x10 (十六进制), 组播标签数据包的数据报类型 为" 0111 1000" (二进制), 也就是 0x78 (十六进制), 需要 4张查找表, 例如
*城域查询标签包的标签查找表, 定义为 4号表, 大小为 64K; *城域应答标签包的标签查找表, 定义为 5号表, 大小为 64K;
*单播标签数据包的标签查找表, 定义为 6号表, 大小为 64K;
*组播标签数据包的标签查找表, 定义为 7号表, 大小为 64K; 城域查询标签包、 城域应答标签包、 单播标签数据包、 组播标签 数据包的标签查找表的输出除了表示数据包导向的端口外, 另外还有 16bit的出标签。 例如, 其中的一种节点交换机 MX-4, 它有 4个千兆 光口, 1个 CPU模块接口。 如果 4个千兆光口依次定义为 0号端口到 3 号端口, CPU模块接口定义为 4号端口,则总共需要 64k X 21bit ( 5bit + 16bit )城域查询标签包地址查找表, 64k X 21bit ( 5bit + 16bit ) 城域应答标签包地址查找表, 64K X 21bit ( 5bit + 16bit ) 单播标 签数据包, 64K X 21bit ( 5bit + 16bit ) 组播标签数据包。 例如, 入标签为 0x0001的城域查询标签包查找表的输出为 "1 0000 0000 0000 00000000", 表示包导向 4号端口 (CPU端口), 出标签为 0x0000; 入 标签为 0x0001 的组播标签数据包查找表的输出为 "0 0011 0000 0011 0000 0000", 表示数据包导向 0号、 1号端口, 出标签为 0x0300, 以 此类推。
一种单组播标签数据包实例如下:
假设 0 号端口进来一个数据包它的头部数据是 0x1056 0x1500 0x0000 0x55aa 0x0056 0x1500 0001 0xaa55 0x0000 0x0000 0x00001, 其中 DA是 0x1056 0x1500 0x0000 0x55aa, SA是 0x0056 0x1500 0001 0xaa55保留字节是 0x0000, 标签是 0x0001, 那么它的数据包类型为 0x10, 才艮据查表规则这时查 6号表, 即地址为" 0000 0000 0000 0001", 此地址对应的查找表的输出为" 0 1100 1000 0000 0000 0001", 表示 数据包导向的 2号、 3号端口, 更换标签为 0x8001, 所以当数据包从 2号、 3号端口输出时, 它的头部数据是 0x1056 0x1500 0x0000 0x55aa 0x0056 0x1500 0001 0xaa55 0x0000 0x0000 0x8001。 下面通过实施例详细说明城域网中的通信, 具体包括城域服务器 与节点交换机、 城域服务器与节点服务器的入网流程和服务流程。 3. 2. 2城域网的入网流程
3. 2. 2. 1 城域服务器与节点交换机、 节点服务器的入网流程 首先每台允许入网的交换机都必须在服务器里注册, 交换机的注 册信息有交换机的设备类型、 设备标识, 没有注册的交换机无法入网。 如图 7所示, 所述节点交换机入网的过程涉及以下步骤:
51、 城域服务器向每个端口发送查询包, 节点交换机收到查询包 后发送应答包 (应答中包含交换机的设备类型、 设备标识, 这是每台 交换机的固有信息);
52、 城域服务器收到节点交换机发出的应答后就知道本端口下接 了一台节点交换机, 然后在城域服务器内部的注册信息表里找到该节 点交换机信息, 并向该节点交换机发送入网命令(告诉交换机的城域 网地址和标签), 该节点交换机收到入网命令后就入网了, 同时向所述 城域服务器发送入网命令应答;
5 3、 城域服务器收到交换机发出的入网命令应答后, 就知道该节 点交换机已经入网了, 以后每秒钟向这个端口发送状态查询包, 检查 该节点交换机是否正常工作, 同时还要向该节点交换机的其它端口发 送端口查询包, 检查是否有其他设备接在该节点交换机下面。 如果该 节点交换机正常工作, 收到状态查询包后会发送状态查询应答给所述 城域服务器。 当城域服务器一定时间内 (如 6秒之内) 没有收到状态 查询应答, 就认为该节点交换机已经被移出网络, 不再发送状态查询 包, 继续向本端口发送查询包。
对于连接在节点交换机下的节点服务器入网流程, 与前述过程相 似, 在此不再详述。
3. 2. 2. 2 城域服务器与节点交换机、 节点服务器的入网交互示 例:
所有城域网网中的设备都由设备信息表来描述, 2个 by t e的设备 类型加 6个 by t e的设备标识就可以唯一的标示一个设备,一般根据不 同的设备类型有不同的设备信息表来描述,例如有节点交换机信息表、 节点服务器信息表。 设备信息表的表项由如下构成:
1 ) 设备标识: 6个 byte, 设备注册的时候, 写入城域服务器的 硬盘或 flash, 城域服务器上电后导入 CPU的内存;
2 )设备状态: 2个 byte , 0x0000表示此设备未入网, 0x0001 表示此设备待入网 (城域服务器发出了入网命令包, 但未收到入网命 令应答), 0x0002 表示此设备已入网 (城域服务器收到入网应答包后 设置 );
3 )设备地址: 2个 byte, 该设备分配的城域网地址。
城域网的地址是 16bit, 所有城域网的设备都有唯一的城域网地 址 (包括城域服务器、 节点交换机、 节点服务器), 在城域服务器的 CPU模块维护着一张 2的 16次方的表, 也即 64K的表, 称为城域地址 信息表, 每个表的表项由如下构成:
1 )地址占用描述符: 2个 byte, 0x0000表示此地址未用, 0x0001 表示此地址待用 (城域服务器用此地址发出了入网命令包, 但未收到 入网命令应答), 0x0002 表示此地址已用 (城域服务器收到入网应答 包后设置);
2 )设备类型: 2个 byte, 例如 0x0000表示城域服务器 MS-1000, 0x0001表示其中一种节点交换机 MX-4, 0x0002 表示其中一种节点服 务器 MSS-400;
3 )设备资源描述信息: 若干 byte, 例如该设备是节点交换机, 则包括它的网络端口连接的设备的城域网地址, 它的各个网络端口的 上下行流量计数; 如果该设备是节点服务器, 则包括它的网络端口连 接的设备的接入网地址, 它的网络端口的上下行流量计数, 等等; 所 有这些信息是为了服务流程提供决策依据, 而且每次的服务流程中都 会修改这些信息。
同样, 城域查询标签描述的是城域服务器到节点交换机或节点服 务器的连接, 而城域应答标签描述的是节点交换机或节点服务器到城 域服务器到的连接。 为了简化设计, 假设两者是——对应的, 例如城 域服务器到一节点交换机的城域查询标签是 0x0008, 则该节点交换机 到城域服务器的城域应答标签也是 0x0008, 并且出标签等于入标签。 这样在城域服务器的 CPU模块维护另一张 2的 16次方的表, 也即 64K 的表, 称为城域协议标签信息表, 每个表的表项由如下构成:
1 )标签占用描述符: 2个 byte, 0x0000表示此标签未用, 0x0001 表示此标签待用 (城域服务器用此标签发出了端口查询包, 但未收到 入网应答包), 0x0002 表示此标签已用 (城域服务器收到入网应答包 后设置);
2 ) 标签描述符: 2个 byte, 该标签对应的设备的城域网地址;
3 )标签路由描述信息: 4个 byte, 用来描述该城域查询标签包的 上一跳交换机的城域网地址及端口号,前 2个 byte表示上一跳交换机 的城域网地址, 后 2个 byte表示上一跳交换机的端口号。
如图 8 所示, 举例说明, 假设此城域服务器有 4 个千兆光口, 1 个 CPU模块接口。如果 4个千兆光口依次定义为 0号端口到 3号端口, CPU模块接口定义为 4号端口,此城域服务器型号为 MS-1000,MS-1000 的 0号端口、 1号端口分别接了 1 台 MX-4-0的 2号端口、 3号端口, MX-4-0的 0号端口接了 1台 MSS-400-0, 1号端口接了 1台 MSS-400-l。
入网交互过程如下:
Sl、 MS-1000 服务器上电后初始化硬件, 从硬盘导入配置文件到 CPU内存(例如节点交换机的注册信息、节点服务器的注册信息等等), MS-1000 服务器初始化城域地址信息表、 城域协议标签信息表, 全部 清零(表示所有地址、 标签未用), MS-1000服务器配置自己的城域网 地址为 0x0000, 也即城域地址信息表的第 0x0000项被配置成如下:
•地址占用描述符: 0x0002 表示此地址已用;
•设备描述符: 0x0000表示城域服务器;
•设备资源描述信息: 此城域服务器有 4个千兆光口依次定义为 0 号端口到 3号端口, CPU模块接口定义为 4号端口, 此节点服务器型 号为 MS-1000, 它的网络端口连接的设备的城域网地址未分配, 它的 各个网络端口的上下行流量计数为 0; 下一个可用地址为 0x0001 , 下一个城 i或协议标签为 0x0000;
52、 MS-1000服务器初始化 4、 5、 6、 7号表
•配置 4号表为 "0 0000 0000 0000 0000 0000"到" 0 0000 1111 1111 1111 1111", 即所有城域查询标签包传送关闭;
•配置 5号表为 "1 0000 0000 0000 0000 0000" 到" 1 0000 1111 1111 1111 1111", 即所有的城域应答标签包导向 CPU;
•配置 6号、 7号表为 "0 0000 0000 0000 0000 0000", 即所有单 组播数据包传送关闭;
53、 MS-1000服务器配置知道自己有 4 个千兆光口, 下一个城域 协议标签为 0x0000 , 所以它配置 4个 4号表的表项分别为
• "100 0000 0000 0000 0000" => "0 0001 0000 0000 0000 0000" , 即城域协议标签 0x0000的查询包导向 0号端口;
• "100 0000 0000 0000 0001" => "0 0010 0000 0000 0000 0001" , 即城域协议标签 0x0001的查询包导向 1号端口;
• "100 0000 0000 0000 0010" => "0 0100 0000 0000 0000 0010", 即城域协议标签 0x0002的查询包导向 2号端口;
• "100 0000 0000 0000 0011" => "0 1000 0000 0000 0000 0011", 即城域协议标签 0x0003的查询包导向 3号端口;
下一个城域协议标签为 0x0004 ;
54、 MS-100G服务器发头部信息为 0x9000 0x0000 0x0000 0x0000 0x0000 0x0000 0x0000 0x0000 0x0000 0x0000 0x0000 , 0x9000 0x0000 0x0000 0x0000 0x0000 0x0000 0x0000 0x0000 0x0000 0x0000 0x0001、 0x9000 0x0000 0x0000 0x0000 0x0000 0x0000 0x0000 0x0000 0x0000 0x0000 0x0002 , 0x9000 0x0000 0x0000 0x0000 0x0000 0x0000 0x0000 0x0000 0x0000 0x0000 0x0003的端口查询包, 由于包才艮据标签交换, 所以即使 DA相同也无所谓。 根据 4号表配置端口查询包会依次导向 0 到 3号端口;
标签信息表的第 0x0000项被配置成如下:
•标签占用描述符: 0x0001表示此标签待用; •标签描述符: 不做修改;
•标签路由描述信息: 0x0000 (上一跳交换机的城域网地址即 MS-1000的城域网地址), 0x0000 (MS-1000的 0号端口)。
标签信息表的第 0x0001项被配置成如下:
•标签占用描述符: 0x0001表示此标签待用;
•标签描述符: 不做修改;
•标签路由描述信息: 0x0000 (上一跳交换机的城域网地址即 MS-1000的城域网地址), 0x0001 (MS-1000的 1号端口)。
标签信息表的第 0x0002项被配置成如下:
•标签占用描述符: 0x0001表示此标签待用;
•标签描述符: 不做修改;
•标签路由描述信息: 0x0000 (上一跳交换机的城域网地址即 MS-1000的城域网地址), 0x0002 ( MS-1000的 2号端口)。
标签信息表的第 0x0003项被配置成如下:
•标签占用描述符: 0x0001表示此标签待用;
•标签描述符: 不做修改;
•标签路由描述信息: 0x0000 (上一跳交换机的城域网地址即 MS-1000的城域网地址), 0x0003 (MS-1000的 3号端口)。
下一个可用标签为 0x0004;
55、 MX-4-0交换机上电后初始化硬件,
•配置 4号表为 "1 00000000000000000000"到" 1 00001111 1111 1111 1111", 即所有城域查询标签包导向 CPU;
*配置 5号表为 "0 0000 0000 0000 0000 0000" 到" 0 0000 1111 1111 1111 1111", 即所有的城域应答标签包传送关闭;
•配置 6号、 7号表为 "0 0000 0000 0000 0000 0000", 即所有单 组播数据包传送关闭;
56、根据拓朴图 MX-4-0交换机 2号端口收到城域协议标签 0x0000 的查询包, 则 •配置 5号表" 101 0000 0000 0000 0000"=> "0 0100 0000 0000 0000 0000", 即城域协议标签 0x0000的应答包导向 2号端口;
根据拓朴图 MX-4-G交换机 3号端口收到城域协议标签 0x0001的 查询包, 则
*配置 5号表" 101 0000 0000 0000 0001" => "0 1000 0000 0000 0000 0000", 即城域协议标签 0x0001的应答包导向 3号端口;
发送 2个应答包 (应答中包含本交换机的设备类型、 设备标识以 及收到该查询包的端口号), 一个包的头部为 0x0900 0x0000 0x0000 0x0000 0x0000 0x0000 0x0000 0x0000 0x0000 0x0000 0x0000 , 同 时在包中标明收到该查询包的端口号为 2号;
另一个包的头部 0x0900 0x0000 0x0000 0x0000 0x0000 0x0000 0x0000 0x0000 0x0000 0x0000 0x0001 , 同时在包中标明收到该查询 包的端口号为 3号;
S7、 MS-1000服务器 0号端口收到城域协议标签 0x0000的查询应 答包后,
1 )根据应答包中的设备类型知道是节点交换机, 把应答包中的设 备标识与节点交换机的设备信息表中的设备标识项逐一比对, 直到完 全相同表明此设备已注册, 检查设备状态项为 0x0000 , 知道此设备未 入网;
2 )根据应答包中的城域协议标签 0x0000查城域协议标签信息表 的第 0x0000 项, 知道上一跳交换机是 MS-1000 服务器 (地址为 0x0000 ) 、 端口号是 0号;
3 )根据应答包 PDU中第 5字段号(标明收到查询包的交换机端口 号) MS-1000服务器知道 0号端口下接了一台交换机的 2号端口。
发送入网命令 (告诉交换机的城域网地址为 0x0001 ), 包的头部 为 0x9000 0x0000 0x0001 0x0000 0x0000 0x0000 0x0000 0x0000 0x0000 0x0000 0x0000;
地址信息表的第 0x0001被配置成如下: •地址占用描述符: 0x0001 表示此地址待用 (城域服务器用此地 址发出了入网命令包, 但未收到入网命令应答);
•设备描述符: 不做修改;
•设备资源描述信息: 不做修改;
对应的设备信息表的表项被配置成如下:
•设备标识: 不做修改;
•设备状态: 0x0001 表示此设备待入网 (城域服务器发出了入网 命令包, 但未收到入网命令应答);
*设备地址: 0x0001;
S8、 MS-1000服务器 1号端口收到城域协议标签 0x0001的查询应 答包后,
1 )根据应答包中的设备类型知道是节点交换机, 把应答包中的设 备标识与节点交换机的设备信息表中的设备标识项逐一比对, 直到完 全相同表明此设备已注册, 检查设备状态项为 0x0001, 知道此设备待 入网;
2 )根据应答包中的城域协议标签 0x0001查城域协议标签信息表 的第 0x0001 项, 知道上一跳交换机是 MS-1000 服务器 (地址为 0x0000 ) 、 端口号是 1号;
3 )根据应答包 PDU中第 5字段号(标明收到查询包的交换机端口 号) MS-1000服务器知道 1号端口下接了一台交换机的 3号端口。
发送入网命令 (告诉交换机的城域网地址为 0x0001 ), 包的头部 为 0x90000x0000 0x0001 0x00000x0000 0x0000 0x00000x0000 0x0000 0x0000 0x0001;
地址信息表的第 0x0001被配置成如下:
•地址占用描述符: 0x0001 表示此地址待用 (城域服务器用此地 址发出了入网命令包, 但未收到入网命令应答);
•设备描述符: 不做修改;
•设备资源描述信息: 不做修改;
对应的设备信息表的表项被配置成如下: •设备标识: 不做修改;
•设备状态: 0x0001 表示此设备待入网 (城域服务器发出了入网 命令包, 但未收到入网命令应答);
*设备地址: 0x0001 ;
S9、 MX-4-0交换机 2号端口收到城域协议标签 0x0000的入网命 令包(包的头部为 0x9000 0x0000 0x0001 0x0000 0x0000 0x0000 0x0000 0x0000 0x0000 0x0000 0x0000 )后, 比对其中的设备类型、 设备标识, 知道自己的城域网地址是 0x0001就入网了,同时向服务器发送入网命 令应答,包的头部分别为 0x0900 0x0000 0x0000 0x0000 0x0000 0x0000 0x0001 0x0000 0x0000 0x0000 0x0000 ;
S 10、 MX-4-0交换机 3号端口收到城域协议标签 0x0001的入网命 令包(包的头部为 0x9000 0x0000 0x0001 0x0000 0x0000 0x0000 0x0000 0x0000 0x0000 0x0000 0x0001 )后, 比对其中的设备类型、 设备标识, 知道自己的城域网地址是 0x0001就入网了,同时向服务器发送入网命 令应答,包的头部分别为 0x0900 0x0000 0x0000 0x0000 0x0000 0x0000 0x0001 0x0000 0x0000 0x0000 0x0001 ;
S 1 MS-1000服务器 0号端口收到城域协议标签 0x0000的入网 命令应答包后,
1 )根据入网命令应答包中的城域协议标签 0x0000 , 检查标签信 息表的第 0x0000 项的标签路由描述信息知道该交换机的知道上一跳 交换机是 MS-1 GGG服务器 (地址为 GxGGGG ) 、 端口号是 0号。
2 )根据入网命令应答包 PDU中第 9字段号(标明收到查询包的交 换机端口号) MS-1000服务器知道 0号端口下接了一台交换机的 2号 端口;
3 )根据入网命令应答包中的城域地址 0x0001 , 知道该交换机的 城域地址是 0x0001。
综合 1 )、 2 )、 3 ), 可知 MX-4-0交换机已经入网了。
地址信息表的第 0x0001被配置成如下:
•地址占用描述符: 0x0002表示此地址已用; •设备描述符: 0x0001表示其中一种节点交换机 MX-4-0;
•设备资源描述信息: 4个千兆光口依次定义为 0号端口到 3号端 口, CPU模块接口定义为 4号端口; 它的 2号端口接了一台城域地址 0x0000的 MS-1000的 0号端口,其它的网络端口连接的设备的城域网 地址未知, 它的各个网络端口的上下行流量计数为 0
标签信息表的第 0x0000项被配置成如下:
•标签占用描述符: 0x0002表示此标签已用;
•标签描述符: 0x0000;
•标签路由描述信息: 0x0000 (上一跳交换机的城域网地址即 MS-1000的城域网地址), 0x0000 ( MS-1000的 0号端口)。
对应的设备信息表的表项被配置成如下:
•设备标识: 不做修改;
•设备状态: 0x0002 表示此设备已入网 (城域服务器发出了入网 命令包, 并收到入网命令应答);
*设备地址: 0x0001。
城域地址信息表的第 0x0000项被配置成如下:
•地址占用描述符: 不做修改;
•设备描述符: 不做修改;
•设备资源描述信息: 此城域服务器有 4个千兆光口依次定义为 0 号端口到 3号端口, CPU模块接口定义为 4号端口, 此节点服务器型 号为 MS-1000 ,它的 0号端口接了一台城域地址 0x0001的 MX-4-0的 2 号端口, 其它的网络端口连接的设备的城域网地址未分配, 它的各个 网络端口的上下行流量计数为 0;
以后会定时(如每秒)向端口 0发设备状态查询指令,如果 MS-1000 服务器在一定时间内 (如 6秒) 没有收到状态查询应答, 就不再发送 设备状态查询指令, 继续向端口 0发送查询包。
S 12、 MS-1000服务器 1号端口收到城域协议标签 0x0001 的入网 命令应答包后, 1 )根据入网命令应答包中的城域协议标签 0x0001 , 检查标签信 息表的第 0x0001 项的标签路由描述信息知道该交换机的知道上一跳 交换机是 MS-1000服务器 (地址为 0x0000 ) 、 端口号是 1号;
2 )根据入网命令应答包 PDU中第 9字段号(标明收到查询包的交 换机端口号) MS-1000服务器知道 1号端口下接了一台交换机的 3号 端口。
3 )根据入网命令应答包中的城域地址 0x0001 , 知道该交换机的 城域地址是 0x0001。
综合 1 )、 2 )、 3 ), 可知 MX-4-0交换机已经入网了。
地址信息表的第 0x0001被配置成如下:
•地址占用描述符: 0x0002表示此地址已用;
•设备描述符: 0x0001表示其中一种节点交换机 MX-4-0;
•设备资源描述信息: 4个千兆光口依次定义为 0号端口到 3号端 口, CPU模块接口定义为 4号端口; 它的 2号端口接了一台城域地址 0x0000 的 MS-1000 的 0 号端口, 它的 3 号端口接了一台城域地址 0x0000的 MS-1000的 1号端口,其它的网络端口连接的设备的城域网 地址未知, 它的各个网络端口的上下行流量计数为 0。
标签信息表的第 0x0001项被配置成如下:
•标签占用描述符: 0x0002表示此标签已用;
•标签描述符: 0x0001 ;
•标签路由描述信息: 0x0000 (上一跳交换机的城域网地址即 MS-1000的城域网地址), 0x0001 ( MS-1000的 0号端口)。
对应的设备信息表的表项被配置成如下:
•设备标识: 不做修改;
•设备状态: 0x0002 表示此设备已入网 (城域服务器发出了入网 命令包, 并收到入网命令应答);
*设备地址: 0x0001。
城域地址信息表的第 0x0000项被配置成如下:
•地址占用描述符: 不做修改; •设备描述符: 不做修改;
•设备资源描述信息: 此城域服务器有 4个千兆光口依次定义为 0 号端口到 3号端口, CPU模块接口定义为 4号端口, 此节点服务器型 号为 MS-1000 ,它的 0号端口接了一台城域地址 0x0001的 MX-4-0的 2 号端口, 1号端口接了一台城域地址 0x0001的 MX-4-0的 3号端口, 其它的网络端口连接的设备的城域网地址未分配, 它的各个网络端口 的上下行流量计数为 0;
以后会定时(如每秒)向端口 1发设备状态查询指令,如果 MS-1000 服务器在一定时间内 (如 6秒) 没有收到状态查询应答, 就不再发送 设备状态查询指令, 继续向端口 1发送查询包。
S 13、 MS-1000服务器知道它的 0号端口接了城域地址 0x0001 的 MX-4-0的 2号端口, 1号端口接了 MX-4-0的 3号端口, MX-4-0的 0 号、 1号端口未知, 下一个城域协议标签为 0x0004。 所以它配置 4个 4号表的表项分别为
•"100 0000 0000 0000 0100"=> "0 0001 0000 0000 0000 0100", 即城域协议标签 0x0004的查询包导向 0号端口;
•"100 0000 0000 0000 0101"=> "0 0001 0000 0000 0000 0101", 即城域协议标签 0x0005的查询包导向 0号端口;
•"100 0000 0000 0000 0110"=> "0 0010 0000 0000 0000 0110", 即城域协议标签 0x0006的查询包导向 1号端口;
•"100 0000 0000 0000 0111"=> "0 0010 0000 0000 0000 0111", 即城域协议标签 0x0007的查询包导向 1号端口;
下一个城域协议标签为 0x0008。
MS-1000 通过使用标签 0x0000或 0x0001发包通知 MX-4-0 , 配置 MX-4-0 4号表的表项:
• "100 0000 0000 0000 0100"=> "0 0001 0000 0000 0000 0100", , 即城域协议标签 0x0004查询包导向 0号端口;
• "100 0000 0000 0000 0101"=> "0 0010 0000 0000 0000 0101", , 即城域协议标签 0x0005查询包导向 1号端口; • "100 0000 0000 0000 0110"=> "0 0001 0000 0000 0000 0110' 即城域协议标签 0x0006查询包导向 0号端口
• "100 0000 0000 0000 0111"=> "0 0010 oooo oooo oooo our 即城域协议标签 0x0007查询包导向 1号端口
配置 MX-4-0 5号表的表项:
•"101 0000 0000 0000 0100"=> "0 0100 0000 0000 0000 0100' 即城域协议标签 0x0004应答包导向 2号端口
•"101 0000 0000 0000 0101"=> "0 0100 oooo oooo oooo oior 即城域协议标签 0x0005应答包导向 2号端口
• "101 0000 0000 0000 0110"=> "0 1000 0000 0000 0000 0110' 即城域协议标签 0x0006应答包导向 3号端口
• "101 0000 0000 0000 0111"=> "0 1000 oooo oooo oooo our 即城域协议标签 0x0007应答包导向 3号端口;
S 14、 MS-1000服务器发头部信息为 0x9000 0x0000 0x0000 0x0000 0x0000 0x0000 0x0000 0x0000 0x0000 0x0000 0x0004 , 0x9000 0x0000
0x9000 0x0000 0x0000 0x0000 0x0000 0x0000 0x0000 0x0000 0x0000 0x0000 0x0006 , 0x9000 0x0000 0x0000 0x0000 0x0000 0x0000 0x0000 0x0000 0x0000 0x0000 0x0007的端口查询包, 由于包才艮据标签交换, 所以即使 DA相同也无所谓。 根据 0号表配置, 标签 0x0004、 0x0005 查询包会依次导向 0号端口, 标签 0x0006、 0x0007查询包会依次导 向 1号端口;
标签信息表的第 0x0004项被配置成如下:
•标签占用描述符: 0x0001表示此标签待用;
•标签描述符: 不做修改;
•标签路由描述信息: 0x0001 (上一跳交换机的城域网地址即 MX-4-0的城域网地址), 0x0000 ( MX-4-0的 0号端口)。
标签信息表的第 0x0005项被配置成如下:
•标签占用描述符: 0x0001表示此标签待用; •标签描述符: 不做修改;
•标签路由描述信息: 0x0001 (上一跳交换机的城域网地址即 MX-4-0的城域网地址), 0x0001 ( MX-4-0的 1号端口)。
标签信息表的第 0x0006项被配置成如下:
•标签占用描述符: 0x0001表示此标签待用;
•标签描述符: 不做修改;
•标签路由描述信息: 0x0001 (上一跳交换机的城域网地址即 MX-4-0的城域网地址), 0x0000 ( MX-4-0的 0号端口)。
标签信息表的第 0x0007项被配置成如下:
•标签占用描述符: 0x0001表示此标签待用;
•标签描述符: 不做修改;
•标签路由描述信息: 0x0001 (上一跳交换机的城域网地址即 MX-4-0的城域网地址), 0x0001 ( MS-1000的 1号端口)。
下一个可用标签为 0x0008 ;
515、 MSS-400-0 , MSS-400-1 交换机上电后初始化硬件, 由于节 点服务器是标签的发起端或终结端, 因此无需替换标签;
•配置 4号表为 "001 0000 0000",即所有城域查询标签包导向 CPU;
•配置 5号表为 "100 0000 0000", 即所有的城域应答标签包导向 10号端口 (即上行千兆光口);
•配置 6号、 7号表为 "000 0000 0000 ", 即所有单组播数据包传 送关闭;
516、 根据拓朴图 MSS-400-0交换机 10号端口收到城域协议标签 0x0004、 0x0006的查询包, 则:
发送 2个应答包 (应答中包含本交换机的设备类型、 设备标识以 及收到该查询包的端口号), 一个包的头部为 0x0900 0x0000 0x0000 0x0000 0x0000 0x0000 0x0000 0x0000 0x0000 0x0000 0x0004 , 同 时在包中标明收到该查询包的端口号为 10号; 另一个包的头部 0x0900 0x0000 0x0000 0x0000 0x0000 0x0000 0x0000 0x0000 0x0000 0x0000 0x0006 , 同时在包中标明收到该查询 包的端口号为 10号;
517、 根据拓朴图 MSS-400-1交换机 10号端口收到城域协议标签 0x0005、 0x0007的查询包, 则:
发送 2个应答包 (应答中包含本交换机的设备类型、 设备标识以 及收到该查询包的端口号), 一个包的头部为 0x0900 0x0000 0x0000 0x0000 0x0000 0x0000 0x0000 0x0000 0x0000 0x0000 0x0005 , 同 时在包中标明收到该查询包的端口号为 10号;
另一个包的头部 0x0900 0x0000 0x0000 0x0000 0x0000 0x0000 0x0000 0x0000 0x0000 0x0000 0x0007 , 同时在包中标明收到该查询 包的端口号为 10号;
518、 同样重复 S7、 S8、 S9、 S10、 Sl l、 S 12 , 这 2台 MSS-400也 入网了。
3. 2. 3城域网的服务流程
如图 9所示, 假设此城域服务器有 4个千兆光口, 1个 CPU模块 接口。 如果 4个千兆光口依次定义为 0号端口到 3号端口, CPU模块 接口定义为 4 号端口, 此城域服务器型号为 MS-1000 , MS-1000 的 0 号端口、 1号端口分别接了 1台 MX-4-0的 2号端口、 3号端口, MX-4-0 的 0号端口接了 1台 MSS-400-0 , 1号端口接了 1台 MSS-400-l。
终端 STB- G 接在 MSS-400-G 的 0 号端口, 终端 STB_ 1 接在 MSS-400-1的 1号端口, 它们入网后的地址是 STB_ 0 ( 0x0000 0x0000 0x0002 x0009 ), STB—1 ( 0x0000 0x0000 0x0003 0x0012 ), MX-4-0的 城域网地址是 0x0001 ; MX-4-0和 MSS-1000的城域协议标签 0x0000、 0x0001 ; MSS-400-0和 MSS-100G 的城域协议标签 0x0005、 0x0007 , MSS-400-1和 MSS-100G的城域协议标签 0x0006、 0x0008。
3. 2. 3. 1 服务建立流程:
STB_ 0向 MSS-400-0发出申请和 STB_ 1进行可视通信, 步骤如下: 51、 STB-0发出服务申请指令包, 包的 DA 0x0800 0x0000 0x0002 0x0000、 SA 0x0000 0x0000 0x0002 0x0009, reserved 0x0000, PDU 部分见附录, 服务参数是 SERVICE_TYPE_TELEPHONE_REQUEST 或 SERVICE_TYPE_TELEPH0NE_DIRECTo
52、服务申请指令包根据 1号表的配置导向 MSS-400-0,MSS-400-0 根据包的内容判断收到可视通信的申请; 根据 SA知道是 STB_0, (假 设可视电话的带宽是上下行 6Mbit/s, STB_0和 MSS-400-0的上下行 带宽还剩 80Mbit/s ), 检查 STB_0和 MSS-400-0的上下行带宽发现满 足服务要求, 则继续跳转至 S4, 否则跳转至 S3。
53、 MSS-400-0发送菜单给主叫 STB_0, 表示服务拒绝; 发向 STB_0: 包的 DA 0x80000x00000x0002 0x0009、 SA 0x0000 0x0000 0x0002 0x0000, reserved 0x0000, PDU 部分见附录菜单数 据格式。
54、 MSS-400-0根据被叫号码查 CAM表(内容-地址映射表)知 道被叫不在自己的接入网,所以 MSS-400-0向城域服务器 MSS-1000发 出服务申请指令包, 包的 DA 0x0900 0x0000 0x0000 0x0000, SA 0x00000x00000x00020x0009, reserved 0x0000,切、议标签是 0x0005
( PDU 部分参见城域网入网流程的数据格式定义), 服务参数是 SERVICE_TYPE_TELEPHONE_REQUEST 或 SERVICE_TYPE_TELEPH0NE_DIRECTo
55、 MSS-1000收到 MSS-400-0 的服务申请包根据根据包的内容 判断收到可视通信的申请; 根据 SA知道是 MSS-400-0下的终端, (假 设 MX-4-0和 MSS-400-0的上下行带宽还剩 800Mbit/s ); 根据被叫号 码查 CAM表(内容-地址映射表)知道被叫在 MSS-400-1的接入网下(假
MX-4-0和 MSS-400-1的上下行带宽还剩 800Mbit/s ), 检查 MX-4-0 和 MSS-400-0、 MSS-400-1的上下行带宽发现满足服务要求, 则继续跳 转至 S7, 否则跳转至 S6。
56、 MSS-1000 发送服务拒绝包给 MSS-400-0。 包的 DA 0x9000 0x00000x00020x0009, SA 0x00000x00000x00000x0000, reserved 0x0000, 协议标签是 0x0005, PDU部分忽略, MSS-400-0收到此服务 拒绝包则跳转至 S3。
57、 MSS-1000向 MSS-400-1发出服务申请包, 发出服务申请指 令包, 包的 DA 0x9000 0x0000 0x0003 0x0012、 SA 0x0000 0x0000 0x0000 0x0000, reserved 0x0000, 协议标签是 0x0006, ( PDU部分 参见城域 网 入 网 流程的数据格式定义 ), 服务参数是 SERVICE_TYPE_TELEPHONE_REQUEST 或 SERVICE_TYPE_TELEPH0NE_DIRECTo
58、 MSS-400-1 收到 MSS-1000 的服务申请包, 根据根据包的内 容判断收到可视通信的申请; 根据被叫号码查 CAM表(内容-地址映射 表) 知道被叫是 STB_1 (假设 STB_1和 MSS-400-1 的上下行带宽还剩 80Mbit/s ),检查 STB_1和 MSS-400-1的上下行带宽发现满足服务要求, 则继续跳转至 S10, 否则跳转至 S9。
59、 MSS-1000收到此服务拒绝包则跳转至 S6。
510、 MSS-400-1 分别发送菜单被叫, 等待被叫应答;
发向 STB_1: 包的 DA 0x80000x00000x00030x0012、 SA 0x0000 0x0000 0x0003 0x0000、 reserved 0x0000, PDU 部分参见本说明书 中的菜单数据格式部分。
511、 STB-1收到菜单, 发出申请 SERVICE_TYPE_PERMISSION 接 受通信, 包的 DA 0x0800 0x0000 0x0003 0x0000、 SA 0x0000 0x0000 0x0003 0x0012、 reserved 0x0000, ( PDU 部分参见城 i或网入网流程 的数据格式定义 ), 服务参数是 SERVICE_TYPE_PERMISSI0No
512、 MSS-400-1 收到 STB-l 的应答包, 发送服务允许包给 MSS-1000, 包的 DA 0x90000x00000x00000x0000、 SA 0x00000x0000 0x0003 0x0000, reserved 0x0000, 协议标签是 0x0006, PDU部分忽 略。
513、 MSS-1000 收到服务允许包, 则分配单播标签 (假设 MSS-400-0到 MSS-400-1的入标签、 出标签为 0x0000, MSS-400-1到 MSS-400-0的入标签、 出标签为 0x0001 ); 參 MSS-1000 向 MX-4-0 发标签分配包, 包的 DA 0x9000 0x0000 0x0001 0x0000 , SA 0x0000 0x0000 0x0000 0x0000 , reserved 0x0000 , 协议标签是 0x0000 , PDU部分包含入标签、 出标签及导向端口;
參 MSS-1000向 MSS-400-0发标签分配包, 包的 DA 0x9000 0x0000 0x0002 0x0000 , SA 0x0000 0x0000 0x0000 0x0000 , reserved 0x0000 , 协议标签是 0x0005 , PDU部分包含入标签、 出标签及导向端口, 还包 含 DA、 SA和标签的绑定;
參 MSS-1000向 MSS-400-1发标签分配包, 包的 DA 0x9000 0x0000 0x0003 0x0000 , SA 0x0000 0x0000 0x0000 0x0000 , reserved 0x0000 , 协议标签是 0x0006 , PDU部分包含入标签、 出标签及导向端口, 还包 含 DA、 SA和标签的绑定;
514、 MX-4-0收到标签分配包更新其 6号表, 第 0x0000项为: 出 标签为 0x0000 , 导向端口为 1 号端口; 第 0x0001 项为: 出标签为 0x0001 , 导向端口为 0号端口。
515、 MSS-400-0收到标签分配包更新其 DA、 SA和标签绑定的 CAM 表(地址-标签绑定表), 即此 CAM表的第 0x0000 项为: DA 0x1000 0x0000 0x0003 0x0012 SA 0x0000 0x0000 0x0002 0x0009 ;
更新其 6号表, 第 0x0000项为: 出标签为 0x0000 , 导向端口为 10号端口;
MSS-400-0配置自己的 2号表如下:
• "10 0000 0000 0000 1001" => "000 0000 0001" , 即目的地址 ( DA )是 0x1000 0x0000 0x0002 0x0009的单播数据包导向 0号端口; 向 STB-0发编解码命令包:
包的 DA 0x8000 0x0000 0x0002 0x0009、 SA 0x0000 0x0000 0x0002 0x0000 , reserved 0x0000 , PDU部分见编解码命令。
8704
字段 长 代码 说明 号 度
0 1W 8704 编解码命令(服务器 用户)
1 1W 填充
2-4 3W 填充
5-7 3W 填充
8 1W 填充
9-11 3W 填充
12 1W 填充
13 1W 0x3217
14 1W 0x3217
15-18 4W 0x1000 0x0000 0x0000 0x0012
19-22 4W 0x1000 0x0000 0x0000 0x0009
23 1W Oxffff 维持原状
24 lw 0=关闭响铃
25 1W Oxffff 维持原状
26 1W Oxffff 维持原状
27-31 5w 0
S16、 MSS-400-1收到标签分配包更新其 DA、 SA和标签绑定的 CAM 表, 即此 CAM表的第 0x0001项为: DA 0x10000x00000x00020x0009 SA 0x0000 0x0000 0x0003 0x0012;
更新其 6号表, 第 0x0001项为: 出标签为 0x0001, 导向端口为 10号端口;
MSS-400-1配置自己的 2号表如下:
• "10 0000 0000 0001 0010" => "000 0000 0010", 即目的地址 ( DA)是 0x1000 0x0000 0x0003 0x0012的单播数据包导向 1号端口; 向 STB-1发编解码命令包: 包的 DA 0x8000 0x0000 0x0003 0x0012、 SA 0x0000 0x0000 0x0003 0x0000, reserved 0x0000, PDU部分见编解码命令。
Figure imgf000080_0001
根据 0号表,后面编解码命令包会分别导向 STB-0、 STB-L STB-0、 STB-1根据包的内容开始编解码, 并接收、 发送单播数据。
3.2. 3.2 服务通信流程 1 ) STB-0发向 STB-1的包的 DA是 0x1000 0x0000 0x0003 0x0012 SA是 0x0000 0x0000 0x0002 0x0009 ;
2 )包进入 MSS-400-0 , MSS-400-0的交换引擎检查 DA的城域网地 址发现不是本接入网的, 则查 DA、 SA和标签绑定的 CAM表, 获得单播 标签 0x0000 , 再查 6号表的第 0x0000项获得出标签 0x0000 , 导向端 口是 10号端口, 在 10号端口的发送端添加标签 0x0000 , 即包的头部 为 0x1000 0x0000 0x0003 0x0012 0x0000 0x0000 0x0002 0x0009 0x0000 0x0000 0x0000;
3 )包进入 MX-4-0 , MX-4-0的交换引擎根据组合地址域查表, 查 的是 6号表, 即表的地址是 "110 0000 0000 0000 0000", 才艮据 MX-4-0 的 6号表的配置单播标签为 0x0000的单播标签数据包导向 1号端口, 出标签为 0x0000 , , 即包的头部为 0x1000 0x0000 0x0003 0x0012 0x0000 0x0000 0x0002 0x0009 0x0000 0x0000 0x0000;
4 ) 包进入 MSS-400-1的 10号端口的接收模块则去掉标签进入交 换引擎, MSS-400-1的交换引擎根据组合地址域查表, 查的是 2号表, 即表的地址是 "10 0000 0000 0001 0010", 才艮据 MSS-400-1 的 2号表 的配置, 知道该表项的输出是 "00 0000 0010"表示打开下行 1号端口, 所以包进入了 STB-1 ;
5 ) 同样 STB-1发向 STB-0的包的 DA是 0x1000 0x0000 0x0002 0x0009 SA 0x0000 0x0000 0x0003 0x0012 ;
6 )包进入 MSS-400-1 , MSS-400-1的交换引擎检查 DA的城域网地 址发现不是本接入网的, 则查 DA、 SA和标签绑定的 CAM表, 获得单播 标签 0x0001 , 再查 6号表的第 0x0001项获得出标签 0x0001 , 导向端 口是 10号端口, 在 10号端口的发送端添加标签 0x0001 , 即包的头部 为 0x1000 0x0000 0x0002 0x0009 0x0000 0x0000 0x0003 0x0012 0x0000 0x0000 0x0001 ;
7 )包进入 MX-4-0 , MX-4-0的交换引擎根据组合地址域查表, 查 的是 6号表, 即表的地址是 "110 0000 0000 0000 0001", 根据 MX-4-0 的 6号表的配置单播标签为 0x0001的单播标签数据包导向 0号端口, 出标签为 0x0001 , , 即包的头部为 0x1000 0x0000 0x0002 0x0009 0x0000 0x0000 0x0003 0x0012 0x0000 0x0000 0x0001 ;
8 ) 包进入 MSS-400-0的 10号端口的接收模块则去掉标签进入交 换引擎, MSS-400-0的交换引擎根据组合地址域查表, 查的是 2号表, 即表的地址是 "10 0000 0000 0000 1001", 才艮据 MSS-400-G的 2号表 的配置, 知道该表项的输出是 "00 0000 0001"表示打开下行 0号端口, 所以包进入了 STB-0。
3. 2. 4 城域网入网流程中的数据格式定义:
PDU 是用户终端和服务器的信息交互方式, 两者使用原始套接字 ( Raw Socke t )传递 PDU, 其数据格式如下:
Figure imgf000082_0001
系统消息 (PDU ) 定义:
端口查询指令, 由服务器发出, 32W短信令: 字段 长 代码 说明
号 度
0 1W 9A01 城域服务器端口查询指令
1-4 4W 服务器自己的网络地址
5 1W 服务器类型 (例如是城域 服务器 MX-4 )
6-8 3W 系统时钟
9-31 23W 0000 填充 0 端口查询应答指令, 32W短信令:
节点交换机、 节点服务器收到 9A01发给服务器的应答
Figure imgf000083_0001
设备标识: 暂时可以取值 0x5131 0201 000X ( X = 0- 入网指令, 由服务器发出, 32W短信令:
Figure imgf000083_0002
入网确认指令, 32W短信令: 字段 长
代码 说明
号 度
0 1W 9A12 入网确认指令
1 1W 设备类型
2-4 3W 设备标识
5-8 4W 网络地址
标明收到入网命令包的
9 1W
交换机端口号
10-31 22W 0000 填充 0 状态查询指令, 由服务器发出, 32W短信令:
Figure imgf000084_0001
状态查询应答指令, 32W短信令: 字段 长
代码 说明
号 度
0 1W 8A22 交换机状态查询应答指令 1 1W 设备类型
2-4 3W 设备标识
5-8 4W 网络地址
标明收到状态查询包的交
9 1W
换机端口号
1 0 1W 温度
1 1 1W 交换机工作状态
1 2- 31 2 0W 0000 填充 0 当然, 上述节点服务器接入城域网的方式仅仅是本发明的一种优 选实施例, 在具体实现中, 所述节点服务器采用任一种现有技术入网 均是可行的, 例如, 所述节点服务器可以 IP节点的方式入城域网, 即 所述城域网为 IP网络结构,本领域技术人员结合实际情况任意采用均 可, 本发明对此无需加以限制。
当然, 节点服务器也可以独立控制一个局域网, 则此时的节点服 务器就不需要入网了, 其下的接入交换机和终端接入该节点服务器主 控的局域网即可。 三、 以下通过与 IP互联网对比, 更进一步描述本发明实施例的优 点:
1、 网络地址结构上根治仿冒
IP 互联网的地址由用户设备告诉网络; 本发明的新型网地址由网 络告诉用户设备。
为了防范他人入侵, PC 和互联网设置了繁瑣的口令、 密码障碍。 就算是实名地址, 仍无法避免密码被破译或用户稍不留神而造成的安 全信息泄漏。 连接到 IP 互联网上的 PC 终端, 首先必须自报家门, 告 诉网络自己的 IP 地址。 然而, 谁能保证这个 IP 地址是真是假。 这就 是 IP互联网第一个无法克服的安全漏洞。 本发明的新型网终端的地址是通过入网协议学来的, 用户终端只 能用这个学来的地址进入本发明的新型网, 因此, 无需认证确保不会 错。 。 本发明的新型网创立了有序化结构 "带色彩,,的地址体系 ( D/SCAF )。 本发明的新型网地址不仅具备唯一性, 同时具备可定位 和可定性功能, 如同个人身份证号码一样, 隐含了该用户端口的地理 位置、 设备性质、 服务权限等其他特征。 本发明的新型网交换机根据 这些特征规定了分组包的行为规则, 实现不同性质的数据分流。
2、 每次服务发放独立通行证, 阻断黑客攻击和病毒扩散的途径; IP 互联网可以自由进出, 用户自备防火墙; 本发明的新型网每次 服务必须申请通行证。 而节点服务器能够对每次服务都发放独立通行 证的前提就是节点服务器知悉每个服务终端的地址, 知悉每个服务的 数据包所需经过的路径拓朴。
由于通讯协议在用户终端执行, 可能被篡改。 由于路由信息在网 上广播, 可能被窃听。 网络中的地址欺骗、 匿名攻击、 邮件炸弹、 泪 滴、 隐蔽监听、 端口扫描、 内部入侵、 涂改信息等形形色色固有的缺 陷, 为黑客提供了施展空间。 垃圾邮件等互联网污染难以防范。
由于 IP 互联网用户可以设定任意 IP 地址来冒充别人, 可以向网 上任何设备发出探针窥探别人的信息, 也可以向网络发送任意干扰数 据包 (泼脏水) 。 为此, 许多聪明人发明了各种防火墙, 试图保持独 善其身。 但是, 安装防火墙是自愿的, 防火墙的效果是暂时的和相对 的, IP 互联网本身永远不会干净。 这是 IP 互联网第二项收不了场的 安全败笔。
本发明的新型网用户入网后, 网络交换机仅允许用户向节点服务 器发出有限的服务请求 (因为各个接入网设备的地址都是节点服务器 分配的) , 对其他数据包一律关门。 如果节点服务器批准用户申请, 即向用户所在的交换机发出网络通行证, 用户终端发出的每个数据包 若不符合网络交换机端的审核条件一律丟弃, 彻底杜绝黑客攻击。 每 次服务结束后, 自动撤销通行证。 通行证机制由交换机执行, 不在用 户可控制的范围内: 审核用户数据包的源地址: 防止用户发送任何假冒或匿名数据包 (入网后自动设定)
审核目标地址: 用户只能发送数据包到服务器指定的对象 (服务 申请时确定)
审核数据流量: 用户发送数据流量必须符合服务器规定 (服务申 请时确定)
审核版权标识: 防止用户转发从网上下载的有版权内容(内容供 应商设定)
本发明的新型网不需要防火墙、 杀毒、 加密、 内外网隔离等消极 手段, 本发明的新型网从结构上彻底阻断了黑客攻击和病毒扩散的途 径, 是本质上可以高枕无忧的安全网络。
3、 网络设备与用户数据完全隔离, 切断病毒和木马的生命线;
IP 互联网设备可随意拆解用户数据包; 本发明的新型网设备与用 户数据完全隔离。 即在数据传输中, 新型网设备(例如, 交换机、 网 关等) 对用户数据包并不进行拆解, 而是直接依据数据包的地址, 查 找映射表, 将其从相应端口转发即可。 即本发明的交换机并不存在自 己计算和选择路由的功能。
冯-诺依曼创造的电脑将程序指令和操作数据放在同一个地方, 也 就是说一段程序可以修改机器中的其他程序和数据。 沿用至今的这一 电脑模式, 给特洛伊木马, 蠕虫, 病毒, 后门等留下了可乘之机。 随 着病毒的高速积累, 防毒软件和补丁永远慢一拍, 处于被动状态。
互联网 TCP/IP 协议的技术核心是尽力而为、储存转发、检错重发。 为了实现互联网的使命, 网络服务器和路由器必须具备解析用户数据 包的能力, 这就为黑客病毒留了活路, 网络安全从此成了比谁聪明的 角力, 永无安宁。 这是 IP 互联网第三项遗传性缺陷。
本发明的新型网上所有的服务器和交换机设备中的 CPU 都不可 能接触到任意一个用户数据包。 也就是说, 整个本发明的新型网只是 为业务提供方和接收方的终端设备之间, 建立一条完全隔离和流量行 为规范的透明管道。 用户终端不管收发什么数据, 一概与网络无关。 从结构上切断了病毒和木马的生命线。 因此, 本发明的新型网杜绝网 络上的无关人员窃取用户数据的可能, 同理, 那些想当黑客或制毒的 人根本就没有可供攻击的对象。
4、 用户之间的自由连接完全隔离, 确保有效管理:
IP 互联网是自由市场, 无中间人; 本发明的新型网是百货公司, 有中间人。 对于网络来说, 消费者与内容供应商都属于网络用户范畴, 只是大小不同而已。 IP 互联网是个无管理的自由市场, 任意用户之间 可以直接通讯(P2P )。 也就是说, 要不要管理是用户说了算, 要不要 收费是单方大用户 (供应商)说了算, 要不要遵守法规也是单方大用 户 (吸血鬼网站)说了算。 运营商至多收个入场费, 要想执行法律、 道德、 安全和商业规矩只能是天方夜谭, 现在和将来都不可能。 这是 IP 互联网第四项架构上的残疾。
本发明的新型网创造了服务节点概念, 形成有管理的百货公司商 业模式。 用户之间, 或者消费者与供货商之间, 绝对不可能有任何自 由接触, 一切联系都必须取得节点服务器 (中间人) 的批准。 这是实 现网络业务有效管理的必要条件。要想成为新型网 的用户, 必须先与 网络运营商谈判自己的角色, 从普通消费者、 网上商店、 学校医院、 政府部门、 直到电视台, 都属于运营商的客户, 就好像上述部门都是 电话公司的客户一样。 尽管看起来每个角色无非都是收发视讯内容, 但如何收发必须严格遵守各自商定的行为法则。有了不可逾越的规范, 各类用户之间的关系才能在真正意义上分成 C2C、 B2C、 B2B 等, 或 者统称为有管理的用户与用户间通讯 (MP2P ) 。
节点服务器 (中间人)之所以能够服务, 其前提就是节点服务器 能够知悉每个服务终端的地址, 知悉每个服务的数据包所需经过的路 径拓朴。
5、 商业规则植入通讯协议, 确保盈利模式;
IP 互联网奉行先通讯,后管理模式;本发明的新型网奉行先管理, 后通讯模式。
网上散布非法媒体内容, 只有造成恶劣影响以后, 才能在局部范 围内查封, 而不能防范于未然。 法律与道德不能防范有组织有计划的 "职业攻击"。 而且法律只能对已造成危害的人实施处罚。 IP 互联网将 管理定义成一种额外附加的服务, 建立在应用层。 因此, 管理自然成 为一种可有可无的摆设。 这是 IP 互联网第五项难移的本性。
本发明的新型网用户终端只能在节点服务器许可范围内的指定业 务中, 选择申请其中之一。 服务建立过程中的协议信令, 由节点服务 器执行(不经用户之手) 。 用户终端只是被动地回答服务器的提问, 接受或拒绝服务, 不能参与到协议过程中。 一旦用户接受服务器提供 的服务, 只能按照通行证规定的方式发送数据包, 任何偏离通行证规 定的数据包一律在底层交换机中丟弃。 本发明的新型网协议的基本思 路是实现以服务内容为核心的商业模式, 而不只是完成简单的数据交 流。 在这一模式下, 安全是新型网的固有属性, 而不是附加在网络上 的额外服务项目。 当然, 业务权限审核、 资源确认和计费手续等, 均 可轻易包含管理合同之中。
由于本发明关于一种接入网设备的入网系统、 一种节点服务器及 一种接入交换机的装置实施例基本相应于前述方法实施例, 具体可以 参见前述方法实施例中的相关说明, 在此就不贅述了。
需要说明的是, 在本文中, 诸如第一和第二等之类的关系术语仅 仅用来将一个实体或者操作与另一个实体或操作区分开来, 而不一定 要求或者暗示这些实体或操作之间存在任何这种实际的关系或者顺 序。
以上对本发明所提供的一种接入网设备的入网方法、 一种接入网 设备的入网系统、一种节点服务器及一种接入交换机进行了详细介绍, 实施例的说明只是用于帮助理解本发明的方法及其核心思想; 同时, 对于本领域的一般技术人员, 依据本发明的思想, 在具体实施方式及 应用范围上均会有改变之处, 综上所述, 本说明书内容不应理解为对 本发明的限制。

Claims

权 利 要 求
1、 一种接入网设备的入网方法, 其特征在于, 包括:
接入交换机上电, 在其内部的下行协议包地址表中设置所有下行 协议包导向 CPU模块;
所述接入交换机接收节点服务器发送的下行协议包, 依据所述下 行协议包地址表的设置, 将所述下行协议包导向该接入交换机的 CPU 模块, 所述 CPU模块生成上行协议包, 并发送给节点服务器; 所述下 行协议包中包括一个待分配的接入网地址;
节点服务器向该接入交换机发送入网命令, 所述入网命令中包括 该接入交换机的接入网地址, 所述接入网地址即为该接入交换机所接 收下行协议包中的待分配接入网地址;
所述接入交换机更新其内部的下行协议包地址表为, 仅目的地址 为自己接入网地址的协议包导向 CPU模块。
2、 如权利要求 1所述的方法, 其特征在于, 当已入网的接入交换 机接收到节点服务器发送的端口分配包时, 所述的方法还包括:
已入网的接入交换机将目的地址为自己接入网地址的端口分配包 导向 CPU模块;
依据包中的端口分配信息, 在其内部的下行协议包地址表中, 设 置各个端口下行协议包所导向的下行端口。
3、 如权利要求 2所述的方法, 其特征在于, 当已入网的接入交换 机接收到节点服务器发送的端口下行协议包时, 所述的方法还包括: 所述接入交换机依据其内部下行协议包地址表的设置, 将所述端 口下行协议包导向对应的下行端口; 所述端口下行协议包中包括一个 待分配的接入网地址;
若节点服务器接收到连接在所述接入交换机下行端口的某个下级 接入网设备发送的端口上行协议包, 则向该下级接入网设备发送入网 命令, 所述入网命令中包括该下级接入网设备的接入网地址, 所述接 入网地址即为该下级接入网设备所接收的端口下行协议包中待分配的 接入网地址。
4、 如权利要求 3所述的方法, 其特征在于, 所述下级接入网设备 包括接入交换机或终端。
5、 如权利要求 4所述的方法, 其特征在于, 当所述下级接入网设 备为接入交换机时, 所述的方法还包括:
所述下级接入网设备依据入网命令更新其内部的下行协议包地址 表为, 设置目的地址为自己接入网地址的协议包导向 CPU模块。
6、 如权利要求 1所述的方法, 其特征在于, 当已入网的接入交换 机接收到节点服务器定时发送的状态下行协议包时, 所述的方法还包 括:
已入网的接入交换机依据其内部下行协议包地址表的设置, 将目 的地址为自己接入网地址的状态下行协议包导向 CPU模块;
所述 CPU模块生成状态上行协议包, 并发送给节点服务器。
7、 如权利要求 6所述的方法, 其特征在于, 当已入网的接入交换 机接收到节点服务器定时发送的状态下行协议包时, 还包括:
已入网的接入交换机依据其内部下行协议包地址表的设置, 将目 的地址为下级接入网设备的接入网地址的状态下行协议包导向对应的 端口; 所述状态下行协议包经由所述端口被传送至对应的下级接入网 设备;
所述下级接入网设备针对所接收到的状态下行协议包生成状态上 行协议包, 并发送给节点服务器。
8、 如权利要求 1所述的方法, 其特征在于, 所述接入交换机内部 还设置有上行协议包地址表, 所述的方法还包括:
接入交换机上电时, 在其内部的上行协议包地址表中设置所有上 行协议包导向的上行端口。
9、 如权利要求 1所述的方法, 其特征在于, 所述接入交换机内部 还设置有数据包地址表, 所述的方法还包括: 接入交换机上电时, 在其内部的数据包地址表中设置所有数据包 的导向关闭。
10、 如权利要求 1所述的方法, 其特征在于, 还包括:
节点服务器上电, 导入下级接入网设备的注册信息至 CPU , 获得 城域网地址, 并配置自己的接入网地址。
11、 如权利要求 1所述的方法, 其特征在于, 还包括:
所述节点服务器入城域网, 所述城域网是具有集中控制功能的网 络, 包括城域服务器、 节点交换机和节点服务器, 其中所述节点交换 机连接在城域服务器和节点服务器之间; 所述节点服务器入城域网的 步骤包括:
城域网设备接入城域网, 由城域网中具有集中控制功能的城域服 务器为入网的设备分配协议标签和城域网地址;
其中, 所述城域网设备包括节点交换机和节点服务器, 所述协议 标签用于描述城域网设备与城域服务器之间的连接; 当同一个城域网 设备与城域服务器之间有多个连接时, 城域服务器为每个连接分配不 同的协议标签;
针对每个跨越城域网的服务申请, 城域服务器分配对应服务的数 据标签 ,所述数据标签用于描述服务所涉及的节点服务器之间的连接。
12、 如权利要求 11所述的方法, 其特征在于, 所述标签分为入标 签和出标签, 入标签指数据包进入城域服务器或节点交换机的标签, 出标签指该数据包离开城域服务器或节点交换机的标签;
同一个数据包的入标签和出标签为不同的标签,或者为相同标签。
1 3、 如权利要求 12所述的方法, 其特征在于, 城域网设备接入城 域网, 由城域服务器分配协议标签和城域网地址的步骤包括:
城域服务器向其所有下行端口发送城域查询标签包, 每个城域查 询标签包中包含一个城域服务器分配的待用协议标签;
节点服务器上电后, 收到城域服务器发送的城域查询标签包, 然 后向城域服务器返回城域应答标签包, 所述城域应答标签包中包含城 域网设备的序列号和收到城域查询标签包的端口号; 城域服务器收到城域应答标签包后, 根据包中的序列号验证城域 网设备是否注册, 如果已注册, 则向城域网设备收到城域查询标签包 的端口发送入网命令, 所述入网命令中包含城域服务器为城域网设备 分配的城域网地址和所述待用协议标签;
城域网设备的相应端口收到入网命令后, 返回入网命令应答, 城 域网设备接入城域网。
14、 如权利要求 1所述的方法, 其特征在于, 节点服务器以 IP节 点的方式入城域网, 所述城域网为 IP网络结构。
15、 如权利要求 1所述的方法, 其特征在于, 还包括:
节点服务器在其内部的下行协议包地址表中设置各下行协议包分 别导向的下行端口;
所述节点服务器依据所述下行协议包地址表的设置, 通过相应的 下行端口发送下行协议包。
16、 如权利要求 2所述的方法, 其特征在于, 还包括:
节点服务器生成包括有端口分配信息的端口分配包;
节点服务器依据其内部下行协议包地址表的设置, 通过相应的下 行端口发送端口分配包。
17、 如权利要求 1所述的方法, 其特征在于, 所述节点服务器内 部还设置有地址信息表, 所述地址信息表中记录有地址占用信息、 设 备标识信息和设备资源信息。
18、 如权利要求 17所述的方法, 其特征在于, 还包括: 节点服务器上电时, 初始化所述地址信息表, 所述初始化操作包 括:
在地址信息表的空白表项中写入节点服务器的接入网地址, 并将 地址占用信息标记为该地址已用;
写入设备标识信息为当前节点服务器;
写入设备资源信息为当前节点服务器的端口信息。
19、 如权利要求 18所述的方法, 其特征在于, 还包括: 依据节点服务器在其内部的下行协议包地址表中对各下行协议包 所导向的下行端口的设置, 更新所述地址信息表, 所述更新操作包括: 在地址信息表的空白表项中分别写入各个下行协议包中的待分配 接入网地址, 并将地址占用信息标记为该地址待用。
20、 如权利要求 19所述的方法, 其特征在于, 还包括:
依据对端口下行协议包所导向的其他接入设备的下行端口的设 置, 更新所述地址信息表, 所述更新操作包括:
在地址信息表的空白表项中分别写入各个端口下行协议包中的待 分配接入网地址, 并将地址占用信息标记为该地址待用。
21、 如权利要求 20所述的方法, 其特征在于, 还包括:
当节点服务器接收到接入交换机发送的入网命令应答时, 更新所 述地址信息表, 所述更新操作包括:
将所述接入交换机的接入网地址对应表项中的地址占用信息标记 为该地址已用;
更新所述设备标识信息为当前接入交换机;
更新所述设备资源信息为当前接入交换机的端口信息, 以及, 其 端口所连接的接入网设备的接入网地址信息。
22、 如权利要求 21所述的方法, 其特征在于, 还包括:
当节点服务器接收到终端发送的入网命令应答时, 更新所述地址 信息表, 所述更新操作包括:
将所述终端的接入网地址对应表项中的地址占用信息标记为该地 址已用;
更新所述设备标识信息为当前终端;
更新所述设备资源信息为当前终端的端口信息, 以及, 其端口所 连接的接入网设备的接入网地址信息。
23、 如权利要求 6所述的方法, 其特征在于, 还包括:
若节点服务器在预设时间内未收到所述接入交换机发送的状态上 行协议包, 则停止向所述接入交换机发送状态下行协议包, 并更新所 述地址信息表; 所述更新操作包括: 将所述接入交换机的接入网地址对应表项中的地址占用信息标记 为该地址未用;
清除所述设备标识信息和设备资源信息。
24、 如权利要求 7所述的方法, 其特征在于, 还包括:
若节点服务器在预设时间内未收到所述下级接入网设备发送的状 态上行协议包, 则停止向所述下级接入网设备发送状态下行协议包, 并更新所述地址信息表; 所述更新操作包括:
将所述下级接入网设备的接入网地址对应表项中的地址占用信息 标己为该地址未用;
清除所述设备标识信息和设备资源信息。
25、 如权利要求 1 7中所述的方法, 其特征在于, 所述地址信息表 中还包括当前接入网设备端口的上下行流量信息。
26、 如权利要求 1所述的方法, 其特征在于, 所述节点服务器内 部还设置有上行协议包地址表, 所述的方法还包括:
节点服务器上电时, 在其内部的上行协议包地址表中设置, 所有 上行协议包导向 CPU模块。
27、 如权利要求 1所述的方法, 其特征在于, 所述节点服务器内 部还设置有数据包地址表, 所述的方法还包括:
节点服务器上电时, 在其内部的数据包地址表中设置所有数据包 的导向关闭。
28、 如权利要求 26所述的方法, 其特征在于, 还包括: 节点服务器的 CPU模块依据入网命令应答包, 判断是否存在所述 接入交换机的注册信息, 若是, 则判定该接入交换机合法; 否则, 判 断该接入交换机不合法;
若合法, 则发送入网命令。
29、 如权利要求 26所述的方法, 其特征在于, 还包括: 节点服务器的 CPU模块依据入网命令应答包, 判断是否存在所述 接入网设备的注册信息, 若是, 则判定该接入网设备合法; 否则, 判 断该接入网设备不合法; 若合法, 则发送入网命令。
30、 一种节点服务器, 其特征在于, 包括:
下行协议包发送模块, 用于向接入交换机发送下行协议包, 所述 下行协议包中包括一个待分配的接入网地址;
上行协议包接收模块, 用于接收接入交换机针对所述下行协议包 回复的上行协议包;
第一入网命令发送模块, 用于依据所接收的上行协议包, 向相应 的接入交换机发送入网命令, 所述入网命令中包括分配给该接入交换 机的接入网地址, 所述接入网地址为该接入交换机所接收下行协议包 中的待分配接入网地址。
31、 如权利要求 30所述的节点服务器, 其特征在于, 还包括: 端口分配包发送模块, 用于向已入网的接入交换机发送端口分配 包, 所述端口分配包中包括端口分配信息, 所述端口分配信息为各个 端口下行协议包导向所述接入交换机各个下行端口的信息。
32、 如权利要求 31所述的节点服务器, 其特征在于, 还包括: 端口下行协议包发送模块, 用于向已入网的接入交换机发送端口 下行协议包; 所述端口下行协议包中包括一个待分配的接入网地址; 端口上行协议包接收模块, 用于接收连接在所述接入交换机下行 端口的某个下级接入网设备发送的端口上行协议包;
第二入网命令发送模块,用于向该下级接入网设备发送入网命令 , 所述入网命令中包括该下级接入网设备的接入网地址, 所述接入网地 址即为该下级接入网设备所接收的端口下行协议包中待分配的接入网 地址。
33、 如权利要求 32所述的节点服务器, 其特征在于, 还包括: 入网应答接收模块, 用于接收所述下级接入网设备发送的入网命 令应答。
34、 如权利要求 33所述的节点服务器, 其特征在于, 所述下级接 入网设备包括接入交换机或终端。
35、 如权利要求 30所述的节点服务器, 其特征在于, 还包括: 状态下行协议包发送模块, 用于定时向已入网的接入交换机发送 状态下行协议包;
状态上行协议包第一接收模块, 用于接收所述接入交换机针对所 述状态下行协议包回复的状态上行协议包。
36、 如权利要求 35所述的节点服务器, 其特征在于, 还包括: 状态上行协议包第二接收模块, 用于接收下级接入网设备针对所 述状态下行协议包回复的状态上行协议包。
37、 如权利要求 30所述的节点服务器, 其特征在于, 还包括: 初始化模块, 用于在上电后, 导入下级接入网设备的注册信息至
CPU , 获得城域网地址, 并配置自己的接入网地址。
38、 如权利要求 30所述的节点服务器, 其特征在于, 还包括: 协议标签和地址获取模块, 用于接入城域网后, 从城域服务器获 得分配的协议标签和城域网地址, 所述协议标签用于描述节点服务器 与城域服务器之间的连接; 当同一个节点服务器与上级连接设备有多 个连接时, 获得对应每个连接的不同协议标签, 所述上级连接设备包 括节点交换机和城域服务器;所述城域网是具有集中控制功能的网络, 包括城域服务器、 节点交换机和节点服务器, 其中所述节点交换机连 接在城域服务器和节点服务器之间;
数据标签获取模块, 用于针对每个跨越城域网的服务申请, 从城 域服务器获得分配的对应当次服务的数据标签, 所述数据标签用于描 述服务所涉及的节点服务器之间的连接;
标签添加模块, 用于将节点服务器发给城域网的协议包或数据包 添加相应的协议标签或数据标签;
标签删除模块, 用于将从城域网接收的协议包或数据包去掉相应 的协议标签或数据标签。
39、 如权利要求 38所述的节点服务器, 其特征在于, 所述标签分 为入标签和出标签, 入标签指数据包进入城域服务器或节点交换机的 标签, 出标签指该数据包离开城域服务器或节点交换机的标签; 同一个数据包的入标签和出标签为不同的标签,或者为相同标签; 所述标签包括协议标签和数据标签。
40、 如权利要求 39所述的节点服务器, 其特征在于, 还包括: 地址-标签映射表, 用于针对每个跨越城域网的服务, 记录跨越城 域网的两个终端的接入网地址与出标签的绑定关系; 其中, 接入网地 址为每个节点服务器为其下连接的入网设备分配的地址;
则所述标签添加模块依据地址-标签映射表,查找节点服务器发给 城域网的协议包或数据包对应的出标签,并添加查找到的出标签发送。
41、 如权利要求 40所述的节点服务器, 其特征在于, 还包括: 协议包标签表, 用于节点服务器上电时, 设置所有的城域协议包 导向到 CPU模块, 所述城域协议包包括城域服务器发送的城域查询标 签包和城域服务申请包。
42、 如权利要求 40所述的节点服务器, 其特征在于, 还包括: 应答包标签表,用于将城域应答标签包分别导向到相应上行端口; 应答包标签表初始化模块, 用于节点服务器上电时, 设置所有城 域应答标签包的导向关闭;
应答包标签表配置模块, 用于收到城域服务器发送的城域查询标 签包后, 修改自身的应答包标签表, 将所述协议标签对应的城域应答 标签包导向到接收该城域查询标签包的上行端口。
43、 如权利要求 30所述的节点服务器, 其特征在于, 还包括:
IP入网模块, 用于以 IP节点的方式入城域网, 所述城域网为 IP 网络结构。
44、 如权利要求 37所述的节点服务器, 其特征在于, 还包括:
0 号表初始化配置模块, 用于在上电时, 初始化其内部的下行协 议包地址表为各下行协议包的导向关闭。
45、 如权利要求 30所述的节点服务器, 其特征在于, 还包括:
0 号表第一设置模块, 用于在其内部的下行协议包地址表中设置 各下行协议包分别导向的下行端口; 所述下行协议包发送模块依据所述下行协议包地址表的设置, 通 过相应的下行端口发送下行协议包。
46、 如权利要求 31所述的节点服务器, 其特征在于, 还包括: 端口分配包生成模块, 用于生成包括有所述端口分配信息的端口 分配包;
所述端口分配包发送模块依据其内部下行协议包地址表的设置, 通过相应的下行端口发送所述端口分配包。
47、 如权利要求 30所述的节点服务器, 其特征在于, 所述节点服 务器内部还设置有地址信息表, 所述地址信息表中记录有地址占用信 息、 设备标识信息和设备资源信息。
48、 如权利要求 47所述的节点服务器, 其特征在于, 所述节点服 务器还包括:
地址信息表初始化模块, 用于在上电时, 初始化地址信息表, 所 述初始化包括:
在地址信息表的空白表项中写入节点服务器的接入网地址, 并将 地址占用信息标记为该地址已用;
写入设备标识信息为当前节点服务器;
写入设备资源信息为当前节点服务器的端口信息。
49、 如权利要求 48所述的节点服务器, 其特征在于, 还包括: 地址信息表第一更新模块, 用于依据节点服务器在其内部的下行 协议包地址表中对各下行协议包所导向的下行端口的设置, 更新所述 地址信息表, 所述更新包括:
在地址信息表的空白表项中分别写入各个下行协议包中的待分配 接入网地址, 并将地址占用信息标记为该地址待用。
50、 如权利要求 49所述的节点服务器, 其特征在于, 还包括: 地址信息表第二更新模块, 用于依据对端口下行协议包所导向的 其他接入设备的下行端口的设置, 更新所述地址信息表, 所述更新包 括: 在地址信息表的空白表项中分别写入各个端口下行协议包中的待 分配接入网地址, 并将地址占用信息标记为该地址待用。
51、 如权利要求 50所述的节点服务器, 其特征在于, 还包括: 地址信息表第三更新模块, 用于在接收到接入交换机发送的入网 命令应答时, 更新所述地址信息表, 所述更新包括:
将所述接入交换机的接入网地址对应表项中的地址占用信息标记 为该地址已用;
更新所述设备标识信息为当前接入交换机;
更新所述设备资源信息为当前接入交换机的端口信息, 以及, 其 端口所连接的接入网设备的接入网地址信息。
52、 如权利要求 51所述的节点服务器, 其特征在于, 还包括: 地址信息表第四更新模块, 用于在接收到下级接入网设备发送的 入网命令应答时, 更新所述地址信息表, 所述更新包括:
将所述下级接入网设备的接入网地址对应表项中的地址占用信息 标" ^己为该地址已用;
更新所述设备标识信息为当前下级接入网设备;
更新所述设备资源信息为当前下级接入网设备的端口信息,以及, 其端口所连接的接入网设备的接入网地址信息。
53、 如权利要求 35所述的节点服务器, 其特征在于, 还包括: 状态下行协议包第一停发模块, 用于在预设时间内未收到所述接 入交换机发送的状态上行协议包时, 停止向所述接入交换机发送状态 下行协议包;
地址信息表第四更新模块, 用于在停止向所述接入交换机发送状 态下行协议包时, 更新所述地址信息表; 所述更新包括:
将所述接入交换机的接入网地址对应表项中的地址占用信息标记 为该地址未用;
清除所述设备标识信息和设备资源信息。
54、 如权利要求 36所述的节点服务器, 其特征在于, 还包括: 状态下行协议包第二停发模块, 用于在预设时间内未收到所述下 级接入网设备发送的状态上行协议包, 则停止向所述下级接入网设备 发送状态下行协议包;
地址信息表第五更新模块, 用于在停止向所述下级接入网设备发 送状态下行协议包时, 更新所述地址信息表; 所述更新包括:
将所述下级接入网设备的接入网地址对应表项中的地址占用信息 标己为该地址未用;
清除所述设备标识信息和设备资源信息。
55、 如权利要求 47所述的节点服务器, 所述地址信息表中还包括 当前接入网设备端口的上下行流量信息。
56、 如权利要求 30所述的节点服务器, 其特征在于, 所述节点服 务器内部还设置有上行协议包地址表, 所述节点服务器还包括:
1 号表初始化配置模块, 用于在上电时, 在其内部的上行协议包 地址表中设置, 所有上行协议包导向 CPU模块。
57、 如权利要求 30所述的节点服务器, 其特征在于, 所述节点服 务器内部还设置有数据包地址表, 所述节点服务器还包括:
数据包地址表配置模块, 用于上电时, 在其内部的数据包地址表 中设置所有数据包的导向关闭。
58、 如权利要求 30所述的节点服务器, 其特征在于, 还包括: 第一验证模块, 用于判断是否存在所述接入交换机的注册信息, 若是, 则判定该接入交换机合法; 否则, 判断该接入交换机不合法; 若合法, 则触发第一入网命令发送模块。
59、 如权利要求 31所述的节点服务器, 其特征在于, 还包括: 第二验证模块, 用于判断是否存在所述接入网设备的注册信息, 若是, 则判定该接入网设备合法; 否则, 判断该接入网设备不合法; 若合法, 则触发第二入网命令发送模块。
60、 一种接入交换机, 其特征在于, 包括:
0 号表初始化配置模块, 用于在上电时, 在其内部的下行协议包 地址表中设置所有下行协议包导向 CPU模块; 下行协议包接收模块, 用于接收节点服务器发送的下行协议包, 并依据所述下行协议包地址表的设置, 将所述下行协议包导向该接入 交换机的 CPU模块,所述下行协议包中包括一个待分配的接入网地址; 上行协议包回复模块, 用于由所述 CPU模块生成上行协议包, 并 发送给节点服务器;
第一入网命令接收模块, 用于接收节点服务器发送的入网命令, 所述入网命令中包括该接入交换机的接入网地址, 所述接入网地址即 为该接入交换机所接收下行协议包中的待分配接入网地址;
0 号表第一设置模块, 用于更新其内部的下行协议包地址表为, 仅目的地址为自己接入网地址的协议包导向 CPU模块。
61、 如权利要求 60所述的接入交换机, 其特征在于, 还包括: 端口分配包接收模块, 用于在入网后, 接收节点服务器发送的端 口分配包; 所述端口分配包中包括端口分配信息, 所述端口分配信息 为各个端口下行协议包导向接入交换机各个下行端口的信息;
第一导向模块, 用于将目的地址为自己接入网地址的端口分配包 导向 CPU模块;
0 号表第二设置模块, 用于依据所述端口分配信息, 在其内部的 下行协议包地址表中, 设置各个端口下行协议包所导向的下行端口。
62、 如权利要求 61所述的接入交换机, 其特征在于, 还包括: 端口下行协议包接收模块, 用于在入网后, 接收到节点服务器发 送的端口下行协议包; 所述端口下行协议包中包括一个待分配的接入 网地址;
第二导向模块, 用于依据其内部下行协议包地址表的设置, 将所 述端口下行协议包导向对应的下行端口。
63、 如权利要求 61或 62所述的接入交换机, 其特征在于, 还包 括:
入网命令应答模块, 用于向节点服务器发送入网命令应答。
64、 如权利要求 60所述的接入交换机, 其特征在于, 还包括: 状态下行协议包接收模块, 用于在入网后, 接收节点服务器定时 发送的状态下行协议包;
第三导向模块, 用于依据其内部下行协议包地址表的设置, 将目 的地址为自己接入网地址的状态下行协议包导向 CPU模块;
状态上行协议包回复模块, 用于由所述 CPU模块生成状态上行协 议包, 并发送给节点服务器。
65、 如权利要求 64所述的接入交换机, 其特征在于, 还包括: 第四导向模块, 用于依据其内部下行协议包地址表的设置, 将目 的地址为下级接入网设备的接入网地址的状态下行协议包导向对应的 端口; 所述状态下行协议包经由所述端口被传送至对应的下级接入网 设备。
66、 如权利要求 65所述的接入交换机, 其特征在于, 所述接入交 换机内部还设置有上行协议包地址表, 所述的接入交换机还包括:
1 号表初始化配置模块, 用于在上电时, 在其内部的上行协议包 地址表中设置所有上行协议包导向的上行端口。
67、 如权利要求 60所述的接入交换机, 其特征在于, 所述接入交 换机内部还设置有数据包地址表, 所述的接入交换机还包括:
数据包地址表配置模块, 用于在上电时, 在其内部的数据包地址 表中设置所有数据包的导向关闭。
68、 一种接入网设备的入网系统, 其特征在于, 所述接入网设备 包括节点服务器及连接在其下级的接入交换机, 所述节点服务器包括 向接入交换机发送下行协议包的下行协议包发送模块, 以及, 依据接 入交换机回复的上行协议包发送入网命令的第一入网命令发送模块; 所述接入交换机包括:
0 号表初始化配置模块, 用于在上电时, 在其内部的下行协议包 地址表中设置所有下行协议包导向 CPU模块; 下行协议包接收模块, 用于依据所述下行协议包地址表的设置, 将接收到的下行协议包导向该接入交换机的 CPU模块, 所述下行协议 包中包括一个待分配的接入网地址;
上行协议包回复模块, 用于由所述 CPU模块生成上行协议包, 并 发送给节点服务器;
第一入网命令接收模块, 用于接收节点服务器发送的入网命令, 所述入网命令中包括该接入交换机的接入网地址, 所述接入网地址即 为该接入交换机所接收下行协议包中的待分配接入网地址;
0 号表第一设置模块, 用于更新其内部的下行协议包地址表为, 仅目的地址为自己接入网地址的协议包导向 CPU模块。
69、 如权利要求 68所述的系统, 其特征在于, 所述节点服务器还 包括用于向已入网的接入交换机发送端口分配包的端口分配包发送模 块, 所述端口分配包中包括端口分配信息, 所述端口分配信息为各个 端口下行协议包导向所述接入交换机各个下行端口的信息;
所述接入交换机还包括:
第一导向模块, 用于将目的地址为自己接入网地址的端口分配包 导向 CPU模块;
0 号表第二设置模块, 用于依据所述端口分配信息, 在其内部的 下行协议包地址表中, 设置各个端口下行协议包所导向的下行端口。
70、 如权利要求 69所述的系统, 其特征在于, 所述节点服务器还 包括用于向已入网的接入交换机发送端口下行协议包的端口下行协议 包发送模块, 所述端口下行协议包中包括一个待分配的接入网地址; 所述接入交换机还包括:
第二导向模块, 用于依据其内部下行协议包地址表的设置, 将所 述端口下行协议包导向对应的下行端口。
71、 如权利要求 70所述的系统, 其特征在于, 所述接入网设备还 包括连接在已入网接入交换机下行端口的下级接入网设备, 所述节点 服务器还包括用于向所述下级接入网设备发送入网命令的第二入网命 令发送模块; 所述下级接入网设备包括:
端口上行协议包回复模块, 用于针对接收到的端口下行协议包生 成端口上行协议包, 并发送给节点服务器;
第二入网命令接收模块, 用于接收节点服务器发送的入网命令, 所述入网命令中包括该下级接入交换机的接入网地址, 所述接入网地 址即为该下级接入交换机所接收端口下行协议包中的待分配接入网地 址。
72、 如权利要求 71所述的系统, 其特征在于, 所述下级接入网设 备包括接入交换机或终端。
73、 如权利要求 72所述的系统, 其特征在于, 当所述下级接入网 设备为接入交换机时, 所述接入交换机还包括:
0 号表第三设置模块, 用于依据所述入网命令更新其内部的下行 协议包地址表为, 设置目的地址为自己接入网地址的协议包导向 CPU 模块。
74、 如权利要求 68所述的系统, 其特征在于, 所述节点服务器还 包括用于定时向已入网的接入交换机发送状态下行协议包的状态下行 协议包发送模块;
所述的接入交换机还包括:
第三导向模块, 用于依据其内部下行协议包地址表的设置, 将目 的地址为自己接入网地址的状态下行协议包导向 CPU模块;
状态上行协议包回复模块, 用于由所述 CPU模块生成状态上行协 议包, 并发送给节点服务器。
75、 如权利要求 74所述的系统, 其特征在于, 所述接入交换机还 包括:
第四导向模块, 用于依据其内部下行协议包地址表的设置, 将目 的地址为下级接入网设备的接入网地址的状态下行协议包导向对应的 端口; 所述状态下行协议包经由所述端口被传送至对应的下级接入网 设备。
76、 如权利要求 75所述的系统, 其特征在于, 所述下级接入网设 备为下级接入交换机; 所述下级接入交换机还包括:
状态上行协议包回复模块, 用于针对所接收到的状态下行协议包 生成状态上行协议包, 并发送给节点服务器。
77、 如权利要求 75所述的系统, 其特征在于, 所述下级接入网设 备为终端; 所述终端还包括:
状态上行协议包回复模块, 用于针对所接收到的状态下行协议包 生成状态上行协议包, 并发送给节点服务器。
78、 如权利要求 68所述的系统, 其特征在于, 所述接入交换机内 部还设置有上行协议包地址表, 所述接入交换机还包括:
1 号表初始化配置模块, 用于在上电时, 在其内部的上行协议包 地址表中设置所有上行协议包导向的上行端口。
79、 如权利要求 68所述的系统, 其特征在于, 所述接入交换机内 部还设置有数据包地址表, 所述接入交换机还包括:
数据包地址表配置模块, 用于在上电时, 在其内部的数据包地址 表中设置所有数据包的导向关闭。
80、 如权利要求 68所述的系统, 其特征在于, 所述节点服务器还 包括:
初始化模块, 用于在上电后, 导入下级接入网设备的注册信息至 CPU , 获得城域网地址, 并配置自己的接入网地址。
81、 如权利要求 68所述的系统, 其特征在于, 所述节点服务器还 包括:
协议标签和地址获取模块, 用于接入城域网后, 从城域服务器获 得分配的协议标签和城域网地址, 所述协议标签用于描述节点服务器 与城域服务器之间的连接; 当同一个节点服务器与上级连接设备有多 个连接时, 获得对应每个连接的不同协议标签, 所述上级连接设备包 括节点交换机和城域服务器;所述城域网是具有集中控制功能的网络, 包括城域服务器、 节点交换机和节点服务器, 其中所述节点交换机连 接在城域服务器和节点服务器之间; 数据标签获取模块, 用于针对每个跨越城域网的服务申请, 从城 域服务器获得分配的对应当次服务的数据标签, 所述数据标签用于描 述服务所涉及的节点服务器之间的连接;
标签添加模块, 用于将节点服务器发给城域网的协议包或数据包 添加相应的协议标签或数据标签;
标签删除模块, 用于将从城域网接收的协议包或数据包去掉相应 的协议标签或数据标签。
82、 如权利要求 81所述的系统, 其特征在于, 所述标签分为入标 签和出标签, 入标签指数据包进入城域服务器或节点交换机的标签, 出标签指该数据包离开城域服务器或节点交换机的标签;
同一个数据包的入标签和出标签为不同的标签,或者为相同标签; 所述标签包括协议标签和数据标签。
83、 如权利要求 82所述的系统, 其特征在于, 所述节点服务器还 包括:
地址-标签映射表, 用于针对每个跨越城域网的服务, 记录跨越城 域网的两个终端的接入网地址与出标签的绑定关系; 其中, 接入网地 址为每个节点服务器为其下连接的入网设备分配的地址;
则所述标签添加模块依据地址-标签映射表,查找节点服务器发给 城域网的协议包或数据包对应的出标签,并添加查找到的出标签发送。
84、 如权利要求 81所述的系统, 其特征在于, 所述节点服务器还 包括:
协议包标签表, 用于节点服务器上电时, 设置所有的城域协议包 导向到 CPU模块, 所述城域协议包包括城域服务器发送的城域查询标 签包和城域服务申请包。
85、 如权利要求 81所述的系统, 其特征在于, 所述节点服务器还 包括:
应答包标签表,用于将城域应答标签包分别导向到相应上行端口; 应答包标签表初始化模块, 用于节点服务器上电时, 设置所有城 域应答标签包的导向关闭; 应答包标签表配置模块, 用于收到城域服务器发送的城域查询标 签包后, 修改自身的应答包标签表, 将所述协议标签对应的城域应答 标签包导向到接收该城域查询标签包的上行端口。
86、 如权利要求 68所述的系统, 其特征在于, 所述节点服务器还 包括:
IP入网模块, 用于以 IP节点的方式入城域网, 所述城域网为 IP 网络结构。
87、 如权利要求 68所述的系统, 其特征在于, 所述节点服务器还 包括:
0 号表初始化配置模块, 用于在上电时, 初始化其内部的下行协 议包地址表为各下行协议包的导向关闭。
88、 如权利要求 68所述的系统, 其特征在于, 所述节点服务器还 包括:
0 号表第一设置模块, 用于在其内部的下行协议包地址表中设置 各下行协议包分别导向的下行端口;
所述下行协议包发送模块依据所述下行协议包地址表的设置, 通 过相应的下行端口发送下行协议包。
89、 如权利要求 69所述的系统, 其特征在于, 所述节点服务器还 包括:
端口分配包生成模块, 用于生成包括有所述端口分配信息的端口 分配包;
所述端口分配包发送模块依据其内部下行协议包地址表的设置, 通过相应的下行端口发送所述端口分配包。
90、 如权利要求 68所述的系统, 其特征在于, 所述节点服务器内 部还设置有地址信息表, 所述地址信息表中记录有地址占用信息、 设 备标识信息和设备资源信息。
91、 如权利要求 90所述的系统, 其特征在于, 所述节点服务器还 包括: 地址信息表初始化模块, 用于在上电时, 初始化地址信息表, 所 述初始化包括:
在地址信息表的空白表项中写入节点服务器的接入网地址, 并将 地址占用信息标记为该地址已用;
写入设备标识信息为当前节点服务器;
写入设备资源信息为当前节点服务器的端口信息。
92、 如权利要求 91所述的系统, 其特征在于, 所述节点服务器还 包括:
地址信息表第一更新模块, 用于依据节点服务器在其内部的下行 协议包地址表中对各下行协议包所导向的下行端口的设置, 更新所述 地址信息表, 所述更新包括:
在地址信息表的空白表项中分别写入各个下行协议包中的待分配 接入网地址, 并将地址占用信息标记为该地址待用。
93、 如权利要求 92所述的系统, 其特征在于, 所述节点服务器还 包括:
地址信息表第二更新模块, 用于依据对端口下行协议包所导向的 其他接入设备的下行端口的设置, 更新所述地址信息表, 所述更新包 括:
在地址信息表的空白表项中分别写入各个端口下行协议包中的待 分配接入网地址, 并将地址占用信息标记为该地址待用。
94、 如权利要求 93所述的系统, 其特征在于, 所述节点服务器还 包括:
地址信息表第三更新模块, 用于在接收到接入交换机发送的入网 命令应答时, 更新所述地址信息表, 所述更新包括:
将所述接入交换机的接入网地址对应表项中的地址占用信息标记 为该地址已用;
更新所述设备标识信息为当前接入交换机;
更新所述设备资源信息为当前接入交换机的端口信息, 以及, 其 端口所连接的接入网设备的接入网地址信息。
95、 如权利要求 94所述的系统, 其特征在于, 所述节点服务器还 包括:
地址信息表第四更新模块, 用于在接收到下级接入网设备发送的 入网命令应答时, 更新所述地址信息表, 所述更新包括:
将所述下级接入网设备的接入网地址对应表项中的地址占用信息 标" ^己为该地址已用;
更新所述设备标识信息为当前下级接入网设备;
更新所述设备资源信息为当前下级接入网设备的端口信息,以及, 其端口所连接的接入网设备的接入网地址信息。
96、 如权利要求 74所述的系统, 其特征在于, 所述节点服务器还 包括:
状态下行协议包第一停发模块, 用于在预设时间内未收到所述接 入交换机发送的状态上行协议包时, 停止向所述接入交换机发送状态 下行协议包;
地址信息表第四更新模块, 用于在停止向所述接入交换机发送状 态下行协议包时, 更新所述地址信息表; 所述更新包括:
将所述接入交换机的接入网地址对应表项中的地址占用信息标记 为该地址未用;
清除所述设备标识信息和设备资源信息。
97、 如权利要求 75所述的系统, 其特征在于, 所述节点服务器还 包括:
状态下行协议包第二停发模块, 用于在预设时间内未收到所述下 级接入网设备发送的状态上行协议包, 则停止向所述下级接入网设备 发送状态下行协议包;
地址信息表第五更新模块, 用于在停止向所述下级接入网设备发 送状态下行协议包时, 更新所述地址信息表; 所述更新包括:
将所述下级接入网设备的接入网地址对应表项中的地址占用信息 标己为该地址未用;
清除所述设备标识信息和设备资源信息。
98、 如权利要求 90所述的系统, 其特征在于, 所述节点服务器的 地址信息表中还包括当前接入网设备端口的上下行流量信息。
99、 如权利要求 68所述的系统, 其特征在于, 所述节点服务器内 部还设置有上行协议包地址表, 所述节点服务器还包括:
1 号表初始化配置模块, 用于在上电时, 在其内部的上行协议包 地址表中设置, 所有上行协议包导向 CPU模块。
100、 如权利要求 68所述的系统, 其特征在于, 所述节点服务器 内部还设置有数据包地址表, 所述节点服务器还包括:
数据包地址表配置模块, 用于上电时, 在其内部的数据包地址表 中设置所有数据包的导向关闭。
101、 如权利要求 68所述的系统, 其特征在于, 所述节点服务器 还包括:
第一验证模块, 用于判断是否存在所述接入交换机的注册信息, 若是, 则判定该接入交换机合法; 否则, 判断该接入交换机不合法; 若合法, 则触发第一入网命令发送模块。
102、 如权利要求 69所述的系统, 其特征在于, 所述节点服务器 还包括:
第二验证模块, 用于判断是否存在所述接入网设备的注册信息, 若是, 则判定该接入网设备合法; 否则, 判断该接入网设备不合法; 若合法, 则触发第二入网命令发送模块。
PCT/CN2011/077991 2010-08-06 2011-08-04 一种接入网设备的入网方法及系统 WO2012016531A1 (zh)

Priority Applications (1)

Application Number Priority Date Filing Date Title
US13/814,728 US9154404B2 (en) 2010-08-06 2011-08-04 Method and system of accessing network for access network device

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201010248237.3 2010-08-06
CN201010248237.3A CN102377634B (zh) 2010-08-06 2010-08-06 一种接入网设备的入网方法及系统

Publications (1)

Publication Number Publication Date
WO2012016531A1 true WO2012016531A1 (zh) 2012-02-09

Family

ID=45558945

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2011/077991 WO2012016531A1 (zh) 2010-08-06 2011-08-04 一种接入网设备的入网方法及系统

Country Status (3)

Country Link
US (1) US9154404B2 (zh)
CN (1) CN102377634B (zh)
WO (1) WO2012016531A1 (zh)

Families Citing this family (14)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US9906442B2 (en) * 2015-04-17 2018-02-27 Dell Products Lp Systems and methods for increasing the multiprotocol label switching stack
CN104869461A (zh) * 2015-05-22 2015-08-26 南京创维信息技术研究院有限公司 视频数据处理系统及方法
US10153820B2 (en) * 2015-11-25 2018-12-11 Newracom, Inc. Receiver address field for multi-user transmissions in WLAN systems
CN105681930A (zh) * 2016-01-12 2016-06-15 深圳羚羊极速科技有限公司 与流媒体服务器建立快速的tcp连接方式
CN108259443A (zh) * 2016-12-29 2018-07-06 北京视联动力国际信息技术有限公司 一种视联网虚拟终端入网认证方法和装置
CN107194804B (zh) * 2017-05-27 2020-07-28 国家计算机网络与信息安全管理中心 一种p2p网贷数据自动化核验方法
FR3072236B1 (fr) * 2017-10-10 2020-11-27 Bull Sas Dispositif et procede d'acquisition de valeurs de compteurs associes a une tache de calcul
CN109728982B (zh) * 2017-10-27 2022-01-14 北京乾唐视联网络科技有限公司 一种自治网络中自治云的入网方法和系统
CN108650138A (zh) * 2018-05-18 2018-10-12 王志华 一种火电厂自动控制仪表无线数据通信系统及其通信方法
CN109246203B (zh) * 2018-08-24 2021-07-06 日立楼宇技术(广州)有限公司 数据通讯方法、装置、计算机设备和存储介质
CN110120885B (zh) * 2019-04-04 2022-11-01 视联动力信息技术股份有限公司 一种设备状态信息的处理方法和装置
CN111490910A (zh) * 2020-03-27 2020-08-04 深圳融安网络科技有限公司 设备信息的扫描方法、终端设备及计算机可读存储介质
CN111417115B (zh) * 2020-04-01 2023-05-26 四川爱联科技股份有限公司 基于数据链路的免密认证方法及系统
CN111800308B (zh) * 2020-05-21 2023-06-13 视联动力信息技术股份有限公司 一种数据处理的方法和装置

Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101252587A (zh) * 2008-04-18 2008-08-27 杭州华三通信技术有限公司 用户终端的接入鉴权方法和设备
CN101547194A (zh) * 2008-08-19 2009-09-30 孙宁军 一种嵌入式终端获取互联网上数据的方法及系统

Family Cites Families (15)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JPH07118717B2 (ja) * 1993-01-05 1995-12-18 日本電気株式会社 マルチプロトコル対応のパケット網構成方法
US5909430A (en) * 1996-12-31 1999-06-01 Northern Telecom Limited Address assignment in an ATM switched network
US6501761B1 (en) * 1999-02-25 2002-12-31 Fairchild Semiconductor Corporation Modular network switch with peer-to-peer address mapping communication
US6912592B2 (en) * 2001-01-05 2005-06-28 Extreme Networks, Inc. Method and system of aggregate multiple VLANs in a metropolitan area network
US7092943B2 (en) * 2002-03-01 2006-08-15 Enterasys Networks, Inc. Location based data
US20030195962A1 (en) * 2002-04-10 2003-10-16 Satoshi Kikuchi Load balancing of servers
US7698394B2 (en) * 2003-12-23 2010-04-13 Alcatel Lucent Global network management configuration
JP4392316B2 (ja) * 2004-10-04 2009-12-24 株式会社日立製作所 フレーム中継装置
US7664116B2 (en) * 2005-04-12 2010-02-16 Fujitsu Limited Network based routing scheme
CN1881913A (zh) * 2005-06-15 2006-12-20 上海贝尔阿尔卡特股份有限公司 一种网络接入设备中用户接口组播管理方法及其装置
US7978602B2 (en) * 2005-11-14 2011-07-12 Juniper Networks, Inc. Dynamic construction of label switching protocol interfaces
US7839856B2 (en) * 2007-06-06 2010-11-23 Cisco Technology, Inc. Centrally controlled routing with tagged packet forwarding in a wireless mesh network
JP4734374B2 (ja) * 2008-06-04 2011-07-27 アラクサラネットワークス株式会社 ネットワーク中継装置、および、ネットワーク中継装置方法
GB2462493B (en) * 2008-08-13 2012-05-16 Gnodal Ltd Data processing
JP5443745B2 (ja) * 2008-12-01 2014-03-19 富士通株式会社 スイッチ

Patent Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101252587A (zh) * 2008-04-18 2008-08-27 杭州华三通信技术有限公司 用户终端的接入鉴权方法和设备
CN101547194A (zh) * 2008-08-19 2009-09-30 孙宁军 一种嵌入式终端获取互联网上数据的方法及系统

Also Published As

Publication number Publication date
US9154404B2 (en) 2015-10-06
US20130201990A1 (en) 2013-08-08
CN102377634B (zh) 2014-02-05
CN102377634A (zh) 2012-03-14

Similar Documents

Publication Publication Date Title
WO2012016531A1 (zh) 一种接入网设备的入网方法及系统
US9253106B2 (en) Traffic-control-based data transmission method and communication system
WO2012016536A1 (zh) 一种接入网设备的服务通信方法及系统
US9450861B2 (en) Ethernet-compatible method and system
JP5852116B2 (ja) 新型ネットワークの通信方法およびシステム
US9240943B2 (en) Metropolitan area network communications method and communication system
WO2012016529A1 (zh) 一种接入网设备的通信连接方法及系统
US7639612B2 (en) System and method of providing bandwidth on demand
US8085791B1 (en) Using layer two control protocol (L2CP) for data plane MPLS within an L2 network access node
WO2007147340A1 (fr) Procédé, système et dispositif de la technique ethernet d&#39;échange et de transfert
CN109818960B (zh) 数据处理方法和装置
CN109728922B (zh) 一种在自治网络中配置组播链路的方法和相关设备
CN109862439B (zh) 数据处理方法和装置
CN109728982B (zh) 一种自治网络中自治云的入网方法和系统
CN109729107B (zh) 一种自治网络中自治云的入网方法和相关设备
CN109561080B (zh) 一种动态入网通信的方法和装置
CN109787993B (zh) 一种多用户访问专网的方法和装置
CN111885422A (zh) 一种组播源的处理方法、系统和装置

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 11814114

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

WWE Wipo information: entry into national phase

Ref document number: 13814728

Country of ref document: US

32PN Ep: public notification in the ep bulletin as address of the adressee cannot be established

Free format text: NOTING OF LOSS OF RIGHTS PURSUANT TO RULE 112(1) EPC (EPO FORM 1205N DATED 12/04/2013)

122 Ep: pct application non-entry in european phase

Ref document number: 11814114

Country of ref document: EP

Kind code of ref document: A1