WO2014101525A1 - 完成网络地址转换保活的方法及设备 - Google Patents

完成网络地址转换保活的方法及设备 Download PDF

Info

Publication number
WO2014101525A1
WO2014101525A1 PCT/CN2013/084955 CN2013084955W WO2014101525A1 WO 2014101525 A1 WO2014101525 A1 WO 2014101525A1 CN 2013084955 W CN2013084955 W CN 2013084955W WO 2014101525 A1 WO2014101525 A1 WO 2014101525A1
Authority
WO
WIPO (PCT)
Prior art keywords
pcp
user equipment
port
address
internal address
Prior art date
Application number
PCT/CN2013/084955
Other languages
English (en)
French (fr)
Inventor
徐绍华
李志军
谢宝国
Original Assignee
中兴通讯股份有限公司
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 中兴通讯股份有限公司 filed Critical 中兴通讯股份有限公司
Publication of WO2014101525A1 publication Critical patent/WO2014101525A1/zh

Links

Classifications

    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L61/00—Network arrangements, protocols or services for addressing or naming
    • H04L61/09—Mapping addresses
    • H04L61/25—Mapping addresses of the same type
    • H04L61/2503—Translation of Internet protocol [IP] addresses
    • H04L61/256—NAT traversal
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L61/00—Network arrangements, protocols or services for addressing or naming
    • H04L61/09—Mapping addresses
    • H04L61/25—Mapping addresses of the same type
    • H04L61/2503—Translation of Internet protocol [IP] addresses
    • H04L61/255—Maintenance or indexing of mapping tables
    • H04L61/2553—Binding renewal aspects, e.g. using keep-alive messages

Definitions

  • the present invention relates to the field of mobile communications, and in particular to a third generation partnership program
  • 3GPP 3rd Generation Partnership Project, abbreviated as 3GPP
  • 3GPP 3rd Generation Partnership Project
  • the 3GPP Standards Working Group has proposed a 3G, 4G 3GPP network and proposed the concept of all-IP.
  • the all-IP network is to improve and optimize the previous 3GPP system from the perspective of system concept, including integration with IP technology, support for various access systems, and provide users with high-quality and guaranteed integrated services on this basis to ensure future It is competitive for 10 years or even longer.
  • An all-IP network first needs a set of address mechanisms with sufficient capacity. Due to the need to support multiple access systems, all-IP networks require far more address capacity than any previous 3GPP system. Different terminals, devices, users, and subscription information of different access systems need to be identified by addresses. And this requirement for the capacity of the IP address increases sharply with the increase in the type and number of terminals, the expansion of the device, and the popularity of the user.
  • NAT Network Address Translation
  • WAN wide-area network
  • NAT Network Address Translation
  • NAT is used to resolve the mapping between internal private IP addresses and external public IP addresses.
  • the NAT device When a packet leaves the internal network, the NAT device translates the internal private IP source address into a globally unique address for use on a public network, such as the Internet.
  • the NAT router When a packet enters the internal network from the public network, the NAT router translates the globally unique public IP address into an internal private IP address.
  • the shared node of the internal network sends an outbound communication, the NAT forwards the communication and creates a reverse mapping entry for the shared node in the address mapping and/or port mapping database or table. This reverse entry is used to return to the internal network Should be redirected to the correct shared node.
  • a NAT device uses a timeout timer with a configurable timeout period for mapping states. If a particular entry is not used for inbound or outbound communication for longer than the timeout period, the NAT timer expires and the entry is cleared. Once the entry is cleared, the shared node after the NAT can no longer be reached through the connection and a new connection must be initiated (eg by the shared node).
  • a common mechanism to prevent the NAT timer from timing out (or expiring) is called "keep alive" or "heartbeat” processing. Under the keep-alive mechanism, useless communication is generated on the connection at a time interval shorter than the NAT timeout period to reset the timer and thereby keep the connection valid.
  • the core network of the 3GPP includes a Home Subscriber Server (HSS), a Mobility Management Entity (MME), and a Policy and Charging Rule Function (PCRF). ), Serving Gateway (S-GW for short) and Packet Data Gateway (PDN Gateway, P-GW for short).
  • HSS Home Subscriber Server
  • MME Mobility Management Entity
  • PCRF Policy and Charging Rule Function
  • S-GW Serving Gateway
  • PDN Gateway Packet Data Gateway
  • FIG. 1 illustrates a related art by connecting a user equipment to a public network through a 3GPP EPC network.
  • FIG. 1 is a schematic structural diagram of a user equipment accessing a public network through an evolved core network (Evolved Packet Core, EPC for short).
  • EPC evolved Packet Core
  • the base station is responsible for establishing an air interface connection with the UE, and connecting the transmission control signaling to the MME/Serving GPRS Support Node (SGSN) on the signaling plane; and connecting to the P-GW/Gateway GPRS Support Node (GGSN) on the media plane. , transmitting media information.
  • the base station includes an RNC (Radio Network Controller), a Node B (Node B), or an eNB (E-UTRAN NodeB, node of the evolved universal terrestrial radio access network).
  • RNC Radio Network Controller
  • Node B Node B
  • eNB E-UTRAN NodeB, node of the evolved universal terrestrial radio access network
  • the MME/SGSN is a control network element of the core network, and is responsible for control plane related functions such as mobility management, processing of non-access stratum signaling, and management of user mobility management context.
  • the P-GW is a border gateway between the EPS and the Packet Data Network (PDN), and is responsible for the access of the PDN and the forwarding of data between the EPS and the PDN.
  • the S-GW is not shown in the figure.
  • the network element is an access gateway device connected to the radio access network in an Evolved Packet Core (EPC) network, and the data is forwarded between the base station and the P-GW. , and cache the data.
  • EPC Evolved Packet Core
  • the GGSN can perform protocol conversion on GPRS packet data packets in the GSM network, so that these packet data packets can be transmitted to a remote TCP/IP or X.25 network.
  • Both the GGSN and the P-GW are specific network elements of the public data network gateway in the 3G/4G network.
  • the public data network gateway is used to provide protocol conversion and routing when interworking between networks using different architectures or protocols. Facilities for network-compatible functions such as data exchange.
  • the PCRF is not shown in the figure.
  • the network element is a policy and charging rule function entity. It is connected to the carrier service network through the receiving interface Rx, and is responsible for providing charging control, online credit control, threshold control, and quality of service. Service, referred to as QoS).
  • the user equipment UE user equipment is connected to the base station by wireless communication, initiates a service, establishes a path to the signaling plane of the MME/SGSN, and establishes a medium from the base station to the P-GW/GGSN. path.
  • the P-GW/GGSN allocates an internal private IP address to the UE.
  • the UE sends a data packet to the remote end the data packet passes through the NAT, and the NAT device converts the private IP address/port of the UE into a globally unique public network.
  • the NAT when the data packet sent by the remote end to the UE arrives at the NAT router, the NAT translates the public network IP address/port into the private IP address/port of the UE, so the data packet sent to the UE can reach the UE.
  • the NAT maintains a mapping entry for the private IP address/port and public IP address/port, and sets the timeout period for the entry.
  • the UE needs to send a keep-alive message to the NAT to prevent the NAT from deleting the mapping entries of the private IP address/port and the public IP address/port after the timeout.
  • the embodiment of the invention provides a method and a device for completing network address translation and keeping alive, and solves the problem of heavy burden and large power loss of the NAT keep-alive process in the prior art.
  • the method for completing the network address translation keep-alive includes: a port control protocol (PCP) client acquiring an internal address and a port of the user equipment, and sending a PCP request carrying the internal address and port of the user equipment to a Port Control Protocol (PCP) server, the PCP server transmitting, to a Network Address Translation (NAT) device, an internal address and a port carrying the user equipment Request for keepalive.
  • PCP port control protocol
  • PCP Port Control Protocol
  • NAT Network Address Translation
  • the PCP client is a public data network gateway
  • the step of the PCP client acquiring the internal address and the port of the user equipment includes: receiving, by the public data network gateway, control signaling sent by the user equipment, where the control signaling protocol configuration option (PCO) is carried The internal address and port of the user equipment.
  • PCO control signaling protocol configuration option
  • the PCP client is a public data network gateway
  • the step of the PCP client acquiring the internal address and the port of the user equipment includes: the public data gateway receiving an internal address and a port of the user equipment sent by the base station on the media channel, where the internal part of the user equipment The address and port are sent by the user equipment to the base station.
  • the PCP client is a public data network gateway
  • the step of the PCP client acquiring the internal address and the port of the user equipment includes: the public data network gateway knowing the internal address of the user equipment according to the identifier of the user equipment sent by the control network element of the core network And the control network element of the core network sends the identifier of the user equipment to the public data network gateway after determining that the user equipment needs to perform NAT keepalive.
  • the PCP client is a public data network gateway
  • the step of the PCP client acquiring the internal address and port of the user equipment and sending the PCP request carrying the internal address and port of the user equipment to the PCP server includes: the public data network gateway uplinking from the user equipment
  • the information of the user equipment is detected in the data, where the information includes an internal address and a port of the user equipment, and the PCP is sent to the PCP server when the user equipment needs to perform NAT keep-alive according to the information. request.
  • the PCP client is the user equipment; the PCP client acquires an internal part of the user equipment Address and port, the step of transmitting a PCP request carrying the internal address and port of the user equipment to the PCP server includes: the user equipment carries a PCP request in a protocol configuration option (PCO) of control signaling and controls the The signaling is sent to a public data network gateway, and the public data network gateway sends the PCP request to the PCP server.
  • PCO protocol configuration option
  • the PCP client is the user equipment
  • the step of the PCP client acquiring the internal address and port of the user equipment, and sending the PCP request carrying the internal address and port of the user equipment to the PCP server includes:
  • the public data network gateway adds a PCP server address to an address notification message sent to the user equipment
  • the user equipment learns the address of the PCP server, and sends a PCP request carrying the internal address and port of the user equipment to the PCP server, and the public data network gateway sends the PCP request to the PCP server.
  • the PCP client is an application server
  • the step of the PCP client acquiring the internal address and port of the user equipment, and sending the PCP request carrying the internal address and port of the user equipment to the PCP server includes:
  • the application server determines that the user equipment needs to perform NAT keep-alive, it sends a PCP request carrying an internal address and a port of the user equipment to a policy charging rule function (PCRF), and the PCRF sends the PCP request to the PCP server.
  • PCRF policy charging rule function
  • the step of the application server acquiring the internal address and the port of the user equipment includes: reporting, by the user equipment, the internal address and the port to the application server when registering to the network where the application server is located, or the application The server queries the public data network gateway via the PCRF to learn the internal address and port of the user equipment.
  • the PCP client is a policy charging rule function (PCRF);
  • PCRF policy charging rule function
  • the step of the PCP client acquiring the internal address and port of the user equipment, and sending the PCP request carrying the internal address and port of the user equipment to the PCP server includes: the PCRF acquiring an internal address and a port of the user equipment, When it is determined that the user equipment needs to perform NAT keep-alive, the PCP request is sent to the PCP server.
  • PCRF policy charging rule function
  • the step of the PCRF acquiring the internal address and the port of the user equipment includes: the application server notifying the PCRF of the internal address and port of the user equipment, or the PCRF querying the public data network according to the identifier of the user equipment
  • the gateway knows the internal address and port of the user equipment.
  • An embodiment of the present invention provides a network element device that performs network address translation and keep alive by using a port control protocol, where the network element device includes a first module and a second module.
  • the first module is configured to: obtain an internal address and a port of the user equipment;
  • the second module is configured to: send a PCP request carrying an internal address and a port of the user equipment to a Port Control Protocol (PCP) server.
  • PCP Port Control Protocol
  • the above network element device may also have the following features:
  • the network element device is a public data network gateway as a PCP client;
  • the first module is configured to: learn the protocol configuration option from the received control signaling
  • the above network element device may also have the following features:
  • the network element device is a public data network gateway as a PCP client;
  • the first module is configured to: receive, by using a media channel, an internal address and a port of the user equipment sent by the user equipment from a base station.
  • the above network element device may also have the following features:
  • the network element device is a public data network gateway as a PCP client;
  • the first module is configured to: receive an identifier of the user equipment from a control network element of the core network, and obtain an internal address and a port of the user equipment according to the identifier of the user equipment.
  • the above network element device may also have the following features:
  • the network element device is a public data network gateway as a PCP client;
  • the first module is configured to: detect information of the user equipment from uplink data of the user equipment, where the information includes an internal address and a port of the user equipment.
  • the above network element device may also have the following features:
  • the network element device is an application server that is a PCP client
  • the first module is configured to: obtain an internal address and a port of the user equipment from the user equipment when the user equipment is registered to a network where the application server is located, or query a public data network gateway by using the PCRF
  • the internal address and port of the user device are known.
  • the above network element device may also have the following features:
  • the network element device is a policy charging rule function (PCRF) as a PCP client; the first module is configured to: receive, from the application server, a PCP request carrying an internal address and a port of the user equipment, or according to the user equipment The identity query queries the public data network gateway to know the internal address and port of the user equipment.
  • PCF policy charging rule function
  • the embodiment of the present invention further provides a user equipment, where the user equipment includes a signaling sending module, and the signaling sending module is configured to: carry the internal part of the user equipment in a protocol configuration option (PCO) of control signaling The address and port and the control signaling is sent to the public data network gateway.
  • PCO protocol configuration option
  • An embodiment of the present invention further provides a user equipment, where the user equipment, as a PCP client, includes a signaling sending module;
  • the signaling sending module is configured to: construct a PCP request carrying an internal address and a port of the user equipment, carry a PCP request in a protocol configuration option (PCO) of control signaling, and send the control signaling to a public Data network gateway.
  • PCO protocol configuration option
  • An embodiment of the present invention further provides a user equipment, where the user equipment acts as a port control protocol.
  • the (PCP) client includes a PCP server address obtaining module and a signaling sending module; the PCP server address obtaining module is configured to: learn an address of the PCP server; and the signaling sending module is configured to: send the PCP server to the PCP server A PCP request carrying the internal address and port of the user equipment.
  • the above network element device may also have the following features:
  • the PCP server address obtaining module is configured to: obtain an address of the PCP server from a PCO of the message received from the public data network gateway, or receive an address notification message from the public data network gateway, according to the address notification message The address carried in the message body knows the address of the PCP server.
  • the port control protocol can reduce the communication between the user equipment and the network on the air interface, thereby reducing the battery consumption of the user equipment consumed in the NAT keep-alive process.
  • FIG. 1 is a schematic diagram of a network architecture and a keep-alive method of a NAT through a UE when communicating with an external network through a 3GPP network in the related art;
  • FIG. 2 is a schematic diagram of a method for completing network address translation and keeping alive by using a port control protocol according to an embodiment of the present invention
  • FIG. 3 is a schematic diagram of the public data network gateway in the first embodiment as a PCP client and a PCP server to complete NAT keepalive;
  • FIG. 4 is a flowchart of the public data network gateway performing the NAT keepalive together with the PCP server under the instruction of the UE in the first implementation manner of the first embodiment;
  • FIG. 5 is a flowchart of the public data network gateway performing the NAT keepalive together with the PCP server under the instruction of the UE in the implementation manner 2 of the first embodiment;
  • FIG. 6 is a flowchart of the public data network gateway performing the NAT keepalive together with the PCP server under the indication of the network in the third implementation manner of the first embodiment
  • FIG. 7 is a flowchart of the public data network gateway in the fourth implementation manner of the first embodiment, based on local packet detection, and performing NAT keepalive together with the PCP server;
  • FIG. 8 is a schematic diagram of a UE performing a NAT keepalive operation together with a PCP server in a second embodiment
  • 9 is a flowchart of performing a NAT keepalive by a UE as a PCP client and a PCP server in Embodiment 1 of Embodiment 2;
  • FIG. 10 is a flowchart of performing a NAT keepalive by a UE as a PCP client and a PCP server in Embodiment 2 of Embodiment 2;
  • FIG. 11 is a schematic diagram of the AF as a PCP client and a PCP server performing NAT protection in the third embodiment
  • FIG. 12 is a flow chart of the AF as a PCP client and a PCP server to complete NAT protection in the third embodiment
  • FIG. 13 is a schematic diagram of the PCRF as a PCP client and a PCP server to complete NAT keepalive in Embodiment 4;
  • FIG. 14 is a flow chart of the PCRF as a PCP client and a PCP server to complete NAT keepalive in Embodiment 4;
  • FIG. 15 is a schematic structural diagram of a network element device according to an embodiment of the present invention.
  • FIG. 16 is a schematic structural diagram of another network element device according to an embodiment of the present invention.
  • the method for completing the network address translation keepalive by using the port control protocol includes: the PCP client acquires the internal address and port of the UE, and sends a PCP request carrying the internal address and port of the UE to the port control protocol. (PCP) server, the PCP server sends a keep-alive request carrying the internal address and port of the UE to the NAT device.
  • PCP PCP
  • the network element device as a PCP client includes a first module and a second module
  • the first module is configured to acquire an internal address and a port of the UE
  • the second module is configured to send a PCP request carrying the UE's internal address and port to the PCP server.
  • the public data network gateway performs the NAT keepalive together with the PCP server as the PCP client.
  • the P-GW/GGSN is used as the specific network element of the public data network gateway
  • the MME/SGSN is used as the specific network element for controlling the network element.
  • the P-GW or the GGSN determines that the UE needs to perform NAT preservation, and the P-GW or the GGSN sends a PCP request to the PCP Server, carrying the private network IP address and port of the UE that needs to keep alive.
  • the PCP server periodically sends a keep-alive message to the NAT instead of the UE.
  • the UE sends the internal address and port of the UE to the P-GW/GGSN via the MME/SGSN of the core network, and the P-GW/GGSN receives the message sent by the UE from the signaling path, where the band needs to protect the NAT. Live logo.
  • the UE sends the internal IP address and port of the UE to the base station through the signaling path, and the base station sends the internal IP address and port of the UE to the MME/SGSN, and then the MME/SGSN sends the signal to the P-GW/GGSN.
  • the UE when the UE finds that the data packet it sends passes through the NAT and needs to perform NAT protection, in order to save air interface resources and save the battery loss of the UE, the UE initiates a request to allow the P-GW/GGSN to perform a keep-alive operation. For example, the UE sends a message to the MME/SGSN, where the PCO of the message carries the internal IP address and port of the UE; the MME/SGSN transparently transmits the PCO to the P-GW/GGSN. The P-GW/GGSN works with the PCP Server to complete NAT keepalive. After the service ends, the UE notifies the P-GW/GGSN through the MME/SGSN, and then the P-GW/GGSN cancels the keepalive process of the PCP Server to the NAT. Including steps 400 through 422:
  • Step 400 The UE is attached to the 3GPP network, and is connected to the control network element MME/SGSN of the core network served by the UE.
  • the MME/SGSN selects the P-GW/GGSN for the UE, and the P-GW/GGSN allocates the internal IP address to the UE.
  • the UE initiates a service through the 3GPP network, and the data packet is sent to the remote end through the NAT.
  • Step 401 The UE sends a request message to the MME/SGSN, and carries its own internal IP address and port number in the Protocol Configuration Options (PCO) of the message.
  • the expected timer can be taken as a timer. 1.
  • Expected timer 1 refers to UE prediction
  • the effective time of the service, or the time when the UE is online, can set the timerl to limitless or unlimited, which means that the UE is directly online.
  • the request message is, for example, a Create Session Request, or a Create Bearer Response, or a Bearer Resource Command, or an Update Bearer Response, or a: Create PDP Context Request message, or an Update PDP Context Request, or Create MBMS Context Request message
  • Step 402 The MME/SGSN sends a request message to the P-GW/GGSN.
  • the MME sends the P-GW
  • the request is sent to the S-GW first, and then the S-GW forwards to the P-GW.
  • the S-GW is not shown in the figure, and the following embodiment is also the same.
  • Step 403 After receiving the request message, the P-GW/GGSN sends a PCP request to the PCP Server. Specifically, the P-GW/GGSN parses the internal IP address and port of the UE in the PCO band, and places the internal IP address and port of the UE in a PCP request message, for example, in a PCP MAP message or a PCP PEER message, and the PCP request Carry a timer, remember to be timer2.
  • the value of timer2 can be set to the timerl value carried by the PCO. You can also set a value less than or greater than the timerl carried by the PCO according to the local policy.
  • Step 404 The PCP Server sends a keep-alive keep-alive request to the NAT, carries the internal IP address and port of the UE, and carries the expected keep-alive duration, which is recorded as timer3.
  • Step 405 The NAT returns a response message to the PCP Server, which carries the internal IP address of the UE and the external IP address and port of the UE, and the effective duration of the keepalive, which is recorded as timer3c.
  • the timer3c is smaller than the value of timer2c.
  • Step 406 The PCP Server sends a PCP response message to the P-GW/GGSN, with the internal IP address and port of the UE and the external IP address and port of the UE, and the effective duration of the keep-alive, which is recorded as timsr2c.
  • Step 407 The P-GW/GGSN sends a response message to the MME/SGSN. Response message
  • P-GW/GGSN is the effective duration of NAT protection for the UE, and is recorded as timerlc.
  • Step 408 The MME/SGSN sends a response message to the UE.
  • steps 407 to 408 may be performed before step 406.
  • timerl c can It is thought that limitless or unlimited, or without timerlc, means that P-GW/GGSN can always do NAT keepalive for UE.
  • Step 409 The P-GW/GGSN sends a PCP request to the PCP Server during the timer2c time period. Specifically, the P-GW/GGSN sends a PCP PEER message to the PCP Server, carrying a timer, which can be set to the value of timer2c.
  • Step 410 The PCP Server receives the PCP request and returns a response message during the imer2c time period, and the new timeout period may be the value of timer2c.
  • Step 411 The PCP Server sends a keep-alive message to the NAT in a certain period in the timer 2c, which is shorter than the duration of the timer3c.
  • Step 412 The NAT receives the keep-alive message of the PCP server when the timer expires, and returns a response message, indicating that the keep-alive is successful, and the new timeout period may be timer3c.
  • the step 409 to the step 410 are repeatedly executed in a certain period, and the execution period is less than the negotiated time timer2c, and the steps 411 to 412 are repeatedly executed in another period, which is smaller than the negotiated time timer3c.
  • the internal IP address of the UE and the mapping relationship between the port and the external IP address and port are continuously updated on the NAT, and the UE-to-NAT keep-alive is completed by the P-GW/GGSN and the PCP Server.
  • Step 413 The UE ends the service to the service provider, and the UE sends a message to the base station, and the internal IP address and port of the UE are carried in the PCO, and the timeout timeurl is set to zero, indicating that the NAT is saved.
  • the sent message is, for example, Delete MBMS Context Request, or Delete PDP Context Request; for example, Delete Session Request, or Delete Bearer Request.
  • Step 414 Same as step 402.
  • Step 415 The P-GW/GGSN sends a PCP request to the PCP Server, and the time interval timer carried is set to zero.
  • Step 416 After receiving the PCP request with zero interval, the PCP Server returns a response message, and no longer keeps a keep-alive message to the UE for the UE.
  • Step 417 to step 418 The P-GW/GGSN sends a response message to the MME/SGSN.
  • the MME/SGSN forwards the response message to the UE.
  • the UE sends the internal address and port of the UE to the P-GW/GGSN on the media channel between the P-GW and the GGSN, and the P-GW/GGSN receives the message sent by the UE from the media path, where With an identifier that needs to be kept alive with NAT.
  • the UE sends the internal IP address and port of the UE to the base station through the media path, and the base station sends the internal IP address and port of the UE to the P-GW/GGSN without passing through the MME/SGSN.
  • the P-GW/GGSN performs the keep-alive operation.
  • the UE sends a message to the P-GW/GGSN, where the message carries the internal IP address and port of the UE.
  • the P-GW/GGSN and the PCP Server work together to complete NAT keepalive.
  • the UE notifies the P-GW/GGSN, and the P-GW/GGSN cancels the keep-alive process from the PCP Server to the NAT.
  • the method for the UE to obtain the P-GW/GGSN address for example, the P-GW/GGSN is added and carried in the header of the message sent to the GTP-U of the UE, for example, adding its own address in the extended part of the message header.
  • the P-GW/GGSN is used as the source address to send a notification message to the UE, carrying its own address, and the P-GW/GGSN address can be placed in the header of the GTP-U message, or in the source address. And identify in the message that this is a message from the P-GW/GGSN.
  • Step 500 The UE attaches to the 3GPP network and connects to the control network element of the core network served by the UE.
  • the MME/SGSN selects a P-GW/GGSN for the UE, and the P-GW/GGSN allocates an internal IP address for the UE.
  • the UE initiates a service through the 3GPP network, and the data packet is sent to the far end through the NAT.
  • Step 501 The UE sends a message to the P-GW/GGSN, carrying the internal IP address and port of the UE.
  • the destination is P-GW/GGSN.
  • the expected timerl refers to the time when the UE predicts the service to be valid, or the time when the UE is online. You can set timerl to limitless or unlimited , or without timer. This means that the UE is directly online.
  • Step 502 to step 505 Steps 403 to 406 in the first implementation manner.
  • Step 506 The P-GW/GGSN returns a response message to the UE.
  • Step 507 to step 510 Steps 409 to 412 in the same manner as in the first embodiment.
  • Step 511 The UE ends the service to the service provider, and the UE sends a message to the P-GW/GGSN, with the internal IP address and port of the UE, and the timeout timer is set to zero, indicating that the NAT is saved.
  • Step 512 to step 513 Step 415 to step 416 in the same manner as in Embodiment 1.
  • Step 514 The P-GW/GGSN returns a response message to the UE.
  • the MME/SGSN determines that the UE needs to perform NAT keepalive and then sends the identifier of the UE to the P-GW/GGSN.
  • the P-GW/GGSN learns the internal address and port of the UE according to the identifier of the UE. Specifically, the MME/SGSN determines that the UE can remain online during the service period by using the local configuration or the registration information of the UE, and notifies the P-GW/GGSN to perform the keep-alive operation for the UE during the service period.
  • the MME/SGSN sends a message to the P-GW/GGSN, where the message carries the identity of the UE.
  • the P-GW/GGSN determines that the UE is in the service progress state, and the P-GW/GGSN and the PCP Server complete the NAT keepalive.
  • the identifier of the UE is, for example, the Mobile International Integrated Service Digital Network (IMSSDN) of the UE, the Public User Identity (PUI), and the private user identifier (Private User). Identity, referred to as PVI), nickname (nickname), etc.
  • IMSSDN Mobile International Integrated Service Digital Network
  • PVI Public User Identity
  • PVI private user identifier
  • nickname nickname
  • the method includes steps 600 through 614:
  • Step 600 The UE attaches to the 3GPP network and connects to the control network element of the core network served by it, such as the MME/SGSN.
  • the MME/SGSN selects a P-GW/GGSN for the UE, and the P-GW/GGSN allocates an internal IP address to the UE.
  • the UE initiates a service through the 3GPP network, and the data packet is sent to the remote end through the NAT.
  • Step 601 The MME/SGSN sends a request message to the P-GW/GGSN, carrying the identifier of the UE.
  • the expected timerNet refers to the time when the MME/SGSN predicts the service validity, or the time that the UE allows online.
  • the timerl can be set to Limitless or unlimited, this case means UE - straight online.
  • Step 602 The P-GW/GGSN returns a response message to the MME/SGSN.
  • Step 603 The P-GW/GGSN can learn the internal IP address of the UE according to the identifier of the UE in the request message from the MME/SGSN.
  • the P-GW/GGSN learns that the UE is using the service, and needs to perform NAT keep-alive.
  • the P-GW/GGSN sends a PCP request to the PCP Server, and carries the internal IP address and port of the UE. For example, the P-GW/GGSN learns the UE through packet detection. Internal IP address and port.
  • Step 604 to step 606 Same as step 605 to step 607.
  • Step 607 to step 610 Same as step 611 to step 614.
  • Step 611 The MME/SGSN determines that the UE's service has ended, or the UE's 3GPP network registration times out, that is, the TAU or RAU is not received within the specified time. In the above case, the MME/SGSN sends a request message for canceling the keep-alive to the P-GW/GGSN, carrying the identifier of the UE. timerNet is set to zero.
  • Step 612 The P-GW/GGSN returns a response message to the MME/SGSN.
  • Step 613 The P-GW/GGSN sends a PCP request to the PCP Server. Carrying the interior of the UE
  • time interval timer is set to zero.
  • Step 614 After receiving the PCP request with zero interval, the PCP server no longer sends a keep-alive message to the UE for the NAT.
  • the P-GW/GGSN detects the information of the UE from the uplink data of the UE, and the information includes the internal address and port of the UE, and determines, according to the information, that the UE needs to send a PCP request to the PCP server when performing NAT keep-alive, P-GW/ The GGSN communicates with the PCP server to complete NAT keepalive.
  • the method includes steps 701 to 713:
  • Step 700 The UE attaches to the 3GPP network and connects to the control network element of the core network serving it, such as the MME/SGSN.
  • the MME/SGSN selects a P-GW/GGSN for the UE, and the P-GW/GGSN allocates an internal IP address to the UE.
  • the UE initiates a service through the 3GPP network, and the data packet is sent through the NAT. Send to the far end.
  • Step 701 The P-GW/GGSN performs packet detection on the uplink and downlink data of the UE.
  • the source IP address, the port (ie, the UE's internal IP address and port) and the destination IP address, port, and protocol type, ie, the quintuple, are detected from the upstream data.
  • Step 702 The P-GW/GGSN determines that it needs to perform for the UE according to the detected quintuple information.
  • Step 703 The P-GW/GGSN sends a PCP request to the PCP Server. Specifically, the P-GW/GGSN places the internal IP address and port of the UE in the quintuple in the internal IP address and port in the PCP request message.
  • the specific message is, for example, a PCP MAP message or a PCP PEER message, and the PCP request carries the expected timeout period, which is recorded as timer2.
  • Step 704 The PCP Server sends a keep-alive keep-alive request to the NAT, and carries the internal IP address and port of the UE, and carries the expected keep-alive duration, which is recorded as timer3.
  • Step 705 The NAT returns a response message to the PCP Server, which carries the internal IP address of the UE, the port and the external IP address and port of the UE, and the effective duration of the keepalive, which is recorded as timer3c.
  • timer3c is smaller than the value of timer2c.
  • Step 706 The PCP Server sends a PCP response message to the P-GW/GGSN, with the internal IP address and port of the UE and the external IP address and port of the UE, and the effective duration of the keep-alive, which is recorded as timsr2c.
  • Step 707 to step 710 Same steps 411 to 414.
  • Step 711 The P-GW/GGSN does not receive the uplink or downlink service data packet of the UE within the set time.
  • Step 712 The P-GW/GGSN sends a PCP request message to the PCP Server, carrying the internal IP address and port of the UE, and the time interval timer is set to zero.
  • Step 713 After receiving the PCP request with zero interval, the PCP server no longer sends a keep-alive message to the UE for the NAT.
  • FIG. 8 is a schematic diagram of the UE as a PCP client and a PCP server performing NAT keepalive.
  • the UE itself acts as a PCP client and sends a PCP request to the PCP server.
  • the UE learns the address of the P-GW/GGSN, sends the PCP request to the P-GW/GGSN, and then the P-GW/GGSN queries and selects a PCP server.
  • the PCP request sent by the UE is routed to the PCP server.
  • the UE carries a PCP request message in the PCO of the control signaling message, and sends a control signaling message carrying the internal address and port of the UE to the P-GW/GGSN, and the P-GW/GGSN parses out the PCP request sent to the PCP server.
  • the method includes steps 900 through 922:
  • Step 900 The UE is attached to the 3GPP network, and is connected to the control network element MME/SGSN of the core network served by the UE.
  • the MME/SGSN selects the P-GW/GGSN for the UE, and the P-GW/GGSN allocates the internal IP address to the UE.
  • the UE initiates a service through the 3GPP network, and the data packet is sent to the remote end through the NAT.
  • Step 901 The UE sends a message to the MME/SGSN, and carries a PCP request message in the PCO of the message.
  • the PCP MAP message or the PCP PEER message is placed in the PCO of the message, where the PCP request message carries the internal IP address and port of the UE.
  • the desired timer can be used as the timerl.
  • the expected timerl refers to the time when the UE predicts the service validity, or the time when the UE is online.
  • the timerl can be set to limitless or unlimited, or without the timer. The situation indicates that the UE is directly online.
  • the sent message is, for example, Create Session Request, or may be Create Bearer Response, or Bearer Resource Command, or Update Bearer Response, or: Create PDP Context Request message, or Update PDP Context Request, Or Create MBMS Context Request message.
  • a PCP request message is carried in the PCO of the above message.
  • Step 903 The P-GW/GGSN parses the PCO in the message, and then sends the PCP request message to the PCP Server.
  • Step 904 The PCP Server sends a keep-alive keep-alive request to the NAT, and carries the internal IP address and port of the UE, and carries the expected keep-alive duration, which is recorded as timer2.
  • Step 905 The NAT returns a response message to the PCP Server, which carries the internal IP address of the UE and the external IP address and port of the UE, and the effective duration of the keepalive, which is recorded as timer2c.
  • the timer2c is smaller than the value of timerl.
  • Step 906 The PCP Server sends a PCP response message to the P-GW/GGSN, with the internal IP address and port of the UE and the external IP address and port of the UE, and the effective duration of the keep-alive, which is recorded as timerl catty
  • Step 907 The P-GW/GGSN forwards the response message to the MME/SGSN.
  • the effective time for the UE to perform NAT preservation for the UE is recorded as timerlc.
  • the PCP response message is sent to the UE in the PCO.
  • Step 908 The MME/SGSN sends a message to the UE, and the UE parses the PCP response message of the PCO band.
  • Step 909 The UE sends a message to the MME/SGSN in the time period of the timerl c, and carries the PCP request message in the PCO of the message, for example, placing the PCP PEER message in the PCO of the message, and the PCP request message carries the internal IP address of the UE and
  • the port, carrying the expected timeout duration timer, can be set to the value of timerlc.
  • Step 910 The MME/SGSN forwards the message to the P-GW/GGSN.
  • Step 911 The P-GW/GGSN parses the PCO in the message, and then sends the PCP request message to the PCP Server.
  • Step 912 to step 914 Same as step 906 to step 908.
  • Step 915 The PCP Server sends a keep-alive message to the NAT in a period of timeout timer2c, which is shorter than the duration of the timerlc.
  • Step 916 The NAT receives the keep-alive message of the PCP server when the timeout period is long, and returns a response message, indicating that the keep-alive is successful, and the new timeout period may be timer3c.
  • the steps 909 to 914 are repeatedly executed in a certain period, and the execution period is less than or equal to the negotiated time timerc.
  • Steps 915 to 916 are repeatedly executed in another cycle, and the cycle is smaller than the negotiated time timer2c.
  • the internal IP address of the UE and the mapping relationship between the port and the external IP address and port are continuously updated on the NAT to implement UE-to-NAT keep-alive.
  • Step 917 The UE ends the service, and the UE sends a message to the MME/SGSN, and carries a PCP request message in the PCO of the message, for example, placing a PCP MAP message or a PCP PEER message in the PCO of the message.
  • the PCP request message carries the internal IP address and port of the UE, and the timeout timeurl is set to zero, indicating that the NAT is saved.
  • Step 918 The MME/SGSN forwards the message to the P-GW/GGSN. .
  • Step 919 After receiving the PCP request with zero interval, the PCP server no longer sends a keep-alive message to the UE for the UE, and returns a response message to the P-GW/GGSN.
  • Step 920 to step 922 Same as step 906 to step 908.
  • the UE learns the address of the PCP server and sends a PCP request carrying the internal address and port of the UE to the PCP server.
  • the P-GW/GGSN selects a PCP server for the UE and transmits the address of the PCP server to the UE.
  • the method includes steps 1000 to 1022:
  • Step 1000 The UE attaches to the 3GPP network and connects to the control network element of the core network serving the same.
  • the MME/SGSN selects the P-GW/GGSN for the UE, and the P-GW/GGSN assigns the internal IP address to the UE.
  • the P-GW/GGSN notifies the UE of the address of its PCP server, and the UE sends a PCP request message to the PCP server.
  • the method for the P-GW/GGSN to notify the UE of the address of the PCP server is that the P-GW/GGSN adds the address of the PCP Server to the bearer control message sent to the UE, and may add the address of the PCP Server to the PCO of the message, or
  • the address of the PCP server is carried in the message body in XML format.
  • a request message with a PCP Server address in a PCO carrying a control message The ratio is the Create Session Response, or the Create Bearer Request, or the Modify Bearer Response, or the Update Bearer Request. It can also be: Create PDP Context Response, or Update PDP Context Response, or Create MBMS Context Response.
  • Step 1001 The UE sends a bearer control message to the MME/SGSN, and establishes a UE with
  • Step 1002 The MME/SGSN sends a bearer control message to the P-GW/GGSN.
  • Step 1003 The P-GW/GGSN returns a response message to the UE, and carries a PCP Server address in the PCO of the response message.
  • Step 1004 The MME/SGSN forwards the response message to the UE.
  • Step 1005 The UE sends a PCP request message to the P-GW/GGSN, where the message carries the internal IP address and port of the UE, and the destination address is a PCP server. For example, send a PCP MAP message or a PCP PEER message.
  • Optional can take the expected timer, remember as timerl.
  • the expected timerl refers to the time when the UE predicts the service to be valid, or the time when the UE is online. You can set timerl to limitless or unlimited, or no timer. This means that the UE is directly online.
  • Step 1006 The P-GW/GGSN forwards the PCP request message to the PCP Server.
  • Step 1007 The PCP Server sends a keep-alive keep-alive request to the NAT, and carries the internal IP address and port of the UE, and carries the expected keep-alive duration, which is recorded as timer2.
  • Step 1008 The NAT returns a response message to the PCP Server, which carries the internal IP address and port of the UE and the external IP address and port of the UE, and the effective duration of the keepalive, which is recorded as timer2c.
  • the timer2c is smaller than the timerl. value.
  • Step 1009 The PCP Server returns a PCP response message, with the internal IP address and port of the UE and the external IP address and port of the UE, and the effective duration of the keep-alive, which is recorded as timerlc.
  • Step 1010 The P-GW/GGSN forwards the PCP response message to the UE, and the effective time length of the NAT keep-alive for the UE by the PCP Server in the response message is recorded as timerlc.
  • Step 1011 The UE sends a PCP request to the PCP server in the time period of the timerlc, for example, sending a PCP PEER message carrying the expected timeout duration timer, which can be set to the value of timerlc.
  • Step 1012 The P-GW/GGSN forwards the PCP request message to the PCP server.
  • Step 1013 to step 1014 Same as step 1009 to step 1010.
  • Step 1015 The PCP Server sends a keep-alive message to the NAT in a certain period within the timer 2c of the timeout period.
  • Step 1016 The NAT receives the keep-alive message of the PCP server when the timer expires, and returns a response message, indicating that the keep-alive is successful.
  • the steps 1011 to 1014 are repeatedly executed in a certain period, and the execution period is less than or equal to the negotiated time timerlc.
  • Steps 1015 to 1016 are repeatedly executed in another period, which is smaller than the negotiated time timer2c.
  • the internal IP address of the UE and the mapping relationship between the port and the external IP address and port are continuously updated on the NAT.
  • Step 1017 The service of the UE ends, and the UE P-GW/GGSN sends a PCP request message, where the message carries the internal IP address and port of the UE, and the destination address is a PCP server.
  • the timeout timeurl is set to zero, indicating that the keepalive to NAT is canceled.
  • Step 1018 After receiving the PCP request with zero interval, the PCP server no longer sends a keep-alive message to the UE for the NAT.
  • Step 1019 to step 1020 The PCP server returns a response message.
  • the application server as the PCP client, determines that the UE needs to perform NAT keepalive, and sends a PCP request carrying the internal address and port of the UE to the policy charging rule function (PCRF), and the PCRF will be the PCP.
  • the request is sent to the PCP server.
  • the PCRF After receiving the PCP request, the PCRF performs the function of the PCP Proxy, and forwards the PCP request to the PCP server, and the PCP server replaces the UE to send a keep-alive message to the NAT.
  • the AF obtains the internal IP address and port of the UE, which may be registered in the UE.
  • the method includes steps 1200 to 1216:
  • Step 1200 The UE attaches to the 3GPP network and connects to the core network control network element MME/SGSN serving it.
  • the MME/SGSN selects a P-GW/GGSN for the UE, and the P-GW/GGSN allocates an internal IP address for the UE.
  • the UE initiates a service through the 3GPP network, and the data packet is sent to the far end through the NAT.
  • Step 1201 The AF sends a PCP request to the PCRF, carrying the IP address and port of the UE, for example, sending a PCP MAP message or a PCP PEER message; optionally, a desired timer can be taken as a timerl, and the expected timerl refers to a service predicted by the UE.
  • the valid time, or the time when the UE is online, can be set to limitless or unlimited, or without timer. This means that the UE is directly online.
  • Step 1202 The PCRF forwards the PCP request to the PCP server.
  • the process of the PCRF to route the PCP request to the PCP server may refer to the protocol related to the PCP in the IETF, and details are not described herein.
  • Step 1203 The PCP Server sends a keep-alive keep-alive request to the NAT, and carries the internal IP address and port of the UE, and carries the expected keep-alive duration, which is recorded as timer2.
  • Step 1204 The NAT returns a response message to the PCP Server, which carries the internal IP address and port of the UE and the external IP address and port of the UE, and the effective duration of the keep-alive, which is recorded as timer2c.
  • the timer2c is smaller than the timerl. value.
  • Step 1205 The PCP Server sends a PCP response message to the PCRF. With the internal IP address and port of the UE and the external IP address and port of the UE, and the effective duration of keep-alive, it is recorded as timerlc.
  • Step 1206 The PCRF forwards the PCP response message to the AF.
  • Step 1207 The AF sends a PCP request to the PCRF in the time period of the timer lc. Specifically, the AF sends a PCP PEER message to the PCRF, and carries a timer, which can be set to the value of the timer 1 c.
  • Step 1208 The PCRF routes the PCP request to the PCP Server.
  • Step 1209 The PCP Server receives the PCP request in the imerlc time period and returns a response message.
  • the new timeout period may be the value of timerlc.
  • Step 1210 The PCRF forwards the response message to the AF.
  • Step 1211 The PCP Server sends a keep-alive message to the NAT in a certain period within a timeout period of the AF timeout period, and the period is less than the keep-alive period of the NAT, timer2c.
  • the steps 1207 to 1210 are repeatedly executed in a certain period, and the execution period is less than the negotiation time timerlc. Steps 1211 to 1212 are also repeatedly executed in another period, which is smaller than the negotiated time timer2c.
  • the internal IP address of the UE and the mapping relationship between the port and the external IP address and port are continuously updated on the NAT.
  • the UE-to-NAT keep-alive is completed by AF, PCRF, and PCP Server.
  • Step 1213 After the service of the UE ends, the AF sends a message to the PCRF, with the internal IP address and port of the UE, and the timeout timer is set to zero, indicating that the NAT keep-alive process is cancelled.
  • Step 1214 The PCRF routes the PCP request to the PCP Server.
  • Step 1215 After receiving the PCP request with zero interval, the PCP Server returns a PCP response message. And the PCP Server no longer sends a keep-alive message to the UE for the NAT.
  • Step 1216 The PCRF forwards the PCP response to the AF.
  • the PCRF is used as the PCP client to obtain the internal IP address and port of the UE.
  • the PCRF sends the PCP request carrying the internal IP address and port of the UE to the PCP server.
  • the PCP server replaces the UE.
  • the NAT sends a keep-alive message.
  • the PCRF obtains the internal IP address and port of the UE, and may be notified by the AF to the PCRF when the UE initiates the service, or may be learned by the PCRF from the P-GW/GGSN according to the identifier of the UE. For example, the PCRF queries the P-GW/GGSN for the internal IP address and port of the UE according to the identity of the UE sent by the AF.
  • the identifier of the UE is, for example, a mobile subscriber's International Integrated Service Digital Network (ISDN) number (MSISDN), a public user identity (PUI), and a private user identity (Private User ID).
  • ISDN International Integrated Service Digital Network
  • PVI Public User Identity
  • PVI nickname
  • nickname nickname
  • Step 1400 The UE attaches to the 3GPP network and connects to the core network control network element MME/SGSN serving it.
  • the MME/SGSN selects a P-GW/GGSN for the UE, and the P-GW/GGSN allocates an internal IP address to the UE.
  • the UE initiates a service through the 3GPP network, and the data packet is sent to the remote end through the NAT.
  • Step 1401 The PCRF carries the IP address and port of the UE to the PCP Server. Such as sending a PCP
  • MAP message or PCP PEER message can take the expected timer, remember as timerl.
  • the expected timerl refers to the time when the UE predicts the service to be valid, or the time when the UE is online. You can set the value of timerl to limitless or unlimited, or without timer. This means that the UE is always online.
  • Step 1402 The PCP Server sends a keep-alive keep-alive request to the NAT, and carries the internal IP address and port of the UE, and carries the expected keep-alive duration, which is recorded as timer2.
  • Step 1403 The NAT returns a response message to the PCP Server, and carries the internal IP address and port of the UE and the external IP address and port of the UE, and the effective duration of the keep-alive, which is recorded as timer2c.
  • the timer2c is smaller than the timerl in the case of '1'. Value.
  • Step 1404 The PCP Server sends a PCP response message to the PCRF, with the internal IP address and port of the UE and the external IP address and port of the UE, and the effective duration of the keep-alive, which is recorded as timerlc.
  • Step 1405 The PCRF sends a PCP request to the PCP Server during the time period of the timer.
  • the PCRF sends a PCP PEER message to the PCP server, which carries a timer, which can be set to the timerlc value.
  • Step 1406 The PCP Server receives the PCP request and returns a response message during the imerlc time period, and the new timeout period may be the value of timerlc.
  • Step 1407 The PCP Server sends a keep-alive message to the NAT in a certain period with the timeout period of the PCRF. The period is less than the keep-alive timer of the NAT.
  • Step 1408 The NAT receives the keep-alive message of the PCP server when the timer expires, and returns a response message, indicating that the keep-alive is successful, and the new timeout period may be timer2c.
  • the steps 1405 to 1406 are repeatedly executed in a certain period, the execution period is less than the negotiated time timerc, and the steps 1407 to 1408 are repeatedly executed in another period, which is smaller than the negotiated time timer2c.
  • Step 1409 The service of the UE ends, and the PCRF sends a message to the PCP Server, with the internal IP address and port of the UE, and the timeout timer is set to zero, indicating that the NAT keep-alive process is cancelled.
  • Step 1410 After receiving the PCP request with zero interval, the PCP Server returns a PCP response message, and the PCP server no longer sends a keep-alive message to the UE for the NAT.
  • the network element device includes: a first module, configured to: obtain an internal address and a port of the user equipment; and a second module, configured to: send a PCP request carrying the internal address and port of the user equipment to the PCP server.
  • the network element device of the PCP client is a public data network gateway
  • the first module of the network element device is configured to receive the internal address and port of the UE sent by the UE through the control network element of the core network, or receive the UE from the base station through the media channel. The internal address and port of the transmitted UE.
  • the first module may receive the control signaling and learn the internal address and port of the UE from the protocol configuration option (PCO) of the control signaling, and may also receive the internal address of the user equipment sent by the user equipment from the base station through the media channel. And port.
  • PCO protocol configuration option
  • the PCP request message may be forwarded, and the first module is configured to receive control signaling and obtain the PCP carried by the protocol configuration option (PCO) of the control signaling.
  • the request message, the PCP request message carries the internal address and port of the UE.
  • the first module of the network element device is configured to receive the identifier of the UE from the control network element of the core network, and obtain the internal address and port of the UE according to the identifier of the UE.
  • the first module of the network element device is configured to detect information of the UE from the uplink data of the UE, where the information includes the internal address and port of the UE.
  • the first module of the network element device is configured to learn the internal address and port of the UE from the UE when the UE registers with the network where the application server is located. Or the internal address and port of the UE are known by querying the public data network gateway via the PCRF.
  • the first module of the network element device is configured to receive a PCP request that carries the internal address and port of the UE from the application server, or query the internal of the UE from the public data network gateway according to the identifier of the UE. Address and port.
  • the method when the network element device of the PCP client is a UE, the method includes a signaling sending module, and the signaling sending module is configured to construct a PCP request carrying the internal address and port of the UE, and a protocol configuration option in the control signaling ( The PCO carries the PCP request and sends this control signaling to the public data network gateway.
  • the UE when the network element device of the PCP client is the UE, the UE includes a PCP server acquiring module and a signaling sending module, the gateway address obtaining module is configured to learn the address of the PCP server, and the signaling sending module is configured to send the carrying to the PCP server.
  • the PCP server address obtaining module is configured to learn the address of the PCP server from the PCO of the message received from the public data network gateway, or receive an address notification message from the public data network gateway, according to the address notification message.
  • the address carried in the message body knows the address of the PCP server.
  • the UE's signaling module is configured to carry the UE's internal address and port in the Protocol Configuration Option (PCO) of the control signaling and send control signaling to the public data network gateway.
  • PCO Protocol Configuration Option
  • the port control protocol can reduce the communication between the user equipment and the network on the air interface, thereby reducing the battery loss consumed by the user equipment during the NAT keep-alive process.

Landscapes

  • Engineering & Computer Science (AREA)
  • Computer Networks & Wireless Communication (AREA)
  • Signal Processing (AREA)
  • Data Exchanges In Wide-Area Networks (AREA)
  • Mobile Radio Communication Systems (AREA)

Abstract

本发明实施例公开了采用端口控制协议完成网络地址转换保活的方法及设备，端口控制协议（PCP）客户端获取用户设备的内部地址和端口，将携带所述用户设备的内部地址和端口的 PCP请求发送至端口控制协议（PCP）服务器，所述 PCP服务器向网络地址转换（NAT）设备发送携带所述用户设备的内部地址和端口的保活请求。

Description

完成网络地址转换保活的方法及设备 技术领域
本发明涉及移动通信领域， 具体涉及一种应用于第三代合作伙伴计划
( 3rd Generation Partnership Project, 简称为 3 GPP )网络的完成网络地址转换 保活的方法及设备。
背景技术
为了保持在通信领域的竟争力， 3GPP标准工作组研究提出来 3G、 4G的 3GPP网络， 并且提出全 IP的概念。 全 IP网络是从系统概念的层面上对以往 3GPP系统进行改进和优化， 包括与 IP技术融合、 支持各种接入系统以及在 此基础上向用户提供高质量有保障的综合业务，确保在未来 10年甚至更长的 时间内具有竟争力。
全 IP网络首先需要一套具备足够容量的地址机制。 由于需要支持多接入 系统， 全 IP网络对于地址容量的要求会远远高于以往任何的 3GPP系统， 不 同接入系统的各种终端、 设备、 用户和签约信息等都需要通过地址来标识， 并且这种对 IP地址容量的要求会随着终端种类和数量的增加、 设备的扩充、 用户的普及等情况而急剧增加。
网络地址转换 (Network Address Translation, 简称为 NAT)属接入广域网 (WAN)技术， 是一种将私有（保留）地址转化为合法 IP地址的转换技术， 它 被广泛应用于各种类型 Internet接入方式和各种类型的网络中。不仅解决了 IP 地址不足的问题， 而且还能够有效地避免来自网络外部的攻击， 隐藏并保护 网络内部的计算机。
NAT用来解决内部私有 IP地址和外部公共 IP地址之间的映射。 当数据 包离开内部网络时， NAT设备将内部私有 IP源地址转换成全球唯一地址以便 用于公共网络， 例如因特网。 当数据包从公共网络进入内部网络时， NAT路 由器将全球唯一的公网 IP地址转换成内部私有 IP地址。 当内部网络的共享 节点发送出站通信时， NAT转发该通信并且在地址映射和 /或端口映射数据库 或表中创建共享节点的反向映射条目。 该反向条目被用于回到内部网络的响 应重定向到正确共享节点。
通常 NAT设备釆用具有用于映射状态的可配置超时时间段的超时定时 器。 如果特定条目在超过超时时间段内未被入站或出站的通信使用， 则 NAT 定时器期满并且该条目被清除。 一旦该条目被清除， 则 NAT之后的共享节点 不能再通过该连接达到， 并且必须发起新连接（例如由该共享节点） 。 防止 NAT定时器超时（或期满）的一种常见机制被称为 "保活"或 "心跳 " 处理。 在保活机制下，以比 NAT超时时间段短的时间间隔在连接上生成无用通信以 复位定时器， 并由此保持该连接有效。
3GPP的核心网中， 包含了归属用户服务器（Home Subscriber Server, 简 称为 HSS )、 移动性管理实体（Mobility Management Entity, 简称为 MME ) 、 策略计费规则功能（Policy and Charging Rule Function, 简称为 PCRF )、 服务 网关（ Serving Gateway, 简称为 S-GW )和分组数据网关（ PDN Gateway, 简 称为 P-GW ) 。
图 1是以用户设备通过 3GPP EPC网络连接到公网为例阐述相关技术。 图 1是用户设备通过演进的核心网（ Evolved Packet Core, 简称为 EPC )接入 到公网的结构示意图。
如图 1所示，基站负责与 UE建立空口连接，在信令面与 MME/服务 GPRS 支持节点 ( SGSN )连接传输控制信令； 在媒体面与 P-GW/网关 GPRS支持节 点 ( GGSN ) 连接， 传输媒体信息。 例如基站包括 RNC ( Radio Network Controller, 无线网络控制器）、 Node B (节点 B )或 eNB ( E-UTRAN NodeB , 演进的通用陆地无线接入网络的节点 Β ) 。
MME/SGSN是核心网的控制网元， 负责移动性管理、 非接入层信令的处 理和用户移动性管理上下文的管理等控制面相关功能。
P-GW是 EPS与分组数据网络（ Packet Data Network, 简称为 PDN ) 的 边界网关， 负责 PDN的接入及其在 EPS与 PDN之间转发数据等功能。 S-GW 在图中未画出， 该网元是演进分组核心网 （Evolved Packet Core, 简称 EPC ) 网络中与无线接入网相连的接入网关设备， 在基站和 P-GW之间转发数据， 并对数据进行緩存。 GGSN可以把 GSM网中的 GPRS分组数据包进行协议转换， 从而可 以把这些分组数据包传送到远端的 TCP/IP或 X.25网络。
GGSN和 P-GW都是公共数据网网关在 3G/4G网络中的具体网元， 公共 数据网网关在釆用不同体系结构或协议的网络之间进行互通时， 用于提供协 议转换、 路由选择、 数据交换等网络兼容功能的设施。
图中未画出 PCRF,该网元是策略和计费规则功能实体，其通过接收接口 Rx和运营商业务网络相连， 负责提供计费控制、 在线信用控制、 门限控制、 以及服务质量（Quality of Service, 简称为 QoS ) 。
如图 1所示， 用户设备 UE用户设备通过无线通信连接到基站， 发起业 务， 建立到 MME/SGSN 的信令面的路径， 并建立从 11£一 >基站一> P-GW/GGSN的媒体路径。 在本例中 P-GW/GGSN为 UE分配了内部私有 IP 地址， UE发送数据包到远端的时候， 数据包经过 NAT , NAT设备将 UE的 私有 IP地址 /端口转换成全球唯一的公网 IP地址 /端口， 当远端发送到 UE的 数据包到达 NAT路由器时， NAT将公网 IP地址 /端口转换成 UE的私有 IP 地址 /端口， 所以发往 UE的数据包可以到达 UE。 在此过程中 NAT维护私有 IP地址 /端口和公网 IP地址 /端口的映射条目， 并设定该条目的超时时间段。 在 UE的业务有效期间， UE需要向 NAT发送保活消息， 以防止 NAT在超时 后将私有 IP地址 /端口和公网 IP地址 /端口的映射条目删除。
然而，相关技术存在如下问题： 在 NAT保活过程中频繁的保活通信消耗 用户设备的电池寿命； 导致大量无线通信， 增加空口负担。
发明内容
本发明实施例提供一种完成网络地址转换保活的方法及设备， 解决现有 技术的 NAT保活流程中空口负担重以及电力损耗多的问题。
本发明实施例提供的完成网络地址转换保活的方法， 包括： 端口控制协 议（ PCP )客户端获取用户设备的内部地址和端口， 将携带所述用户设备的 内部地址和端口的 PCP请求发送至端口控制协议（ PCP )服务器， 所述 PCP 服务器向网络地址转换（NAT )设备发送携带所述用户设备的内部地址和端 口的保活请求。
上述方法还可具有以下特点：
所述 PCP客户端是公共数据网网关；
所述 PCP客户端获取所述用户设备的内部地址和端口的步骤包括： 所述 公共数据网网关接收所述用户设备发送的控制信令， 所述控制信令的协议配 置选项 （PCO ) 中携带所述用户设备的内部地址和端口。
上述方法还可具有以下特点：
所述 PCP客户端是公共数据网网关；
所述 PCP客户端获取所述用户设备的内部地址和端口的步骤包括： 所述 公共数据网关接收基站在媒体通道上发送的所述用户设备的内部地址和端 口， 其中， 所述用户设备的内部地址和端口是所述用户设备发送给基站的。
上述方法还可具有以下特点：
所述 PCP客户端是公共数据网网关；
所述 PCP客户端获取所述用户设备的内部地址和端口的步骤包括： ， 所 述公共数据网网关根据核心网的控制网元发送的所述用户设备的标识获知所 述用户设备的内部地址和端口， 其中， 所述核心网的控制网元是在判断所述 用户设备需进行 NAT保活后向所述公共数据网网关发送所述用户设备的标 识。
上述方法还可具有以下特点：
所述 PCP客户端是公共数据网网关；
所述 PCP客户端获取用户设备的内部地址和端口将携带所述用户设备的 内部地址和端口的 PCP请求发送至所述 PCP服务器的步骤包括：所述公共数 据网网关从所述用户设备的上行数据中检测到所述用户设备的信息， 所述信 息中包括所述用户设备的内部地址和端口， 根据所述信息判断所述用户设备 需进行 NAT保活时向所述 PCP服务器发送所述 PCP请求。
上述方法还可具有以下特点：
所述 PCP客户端是所述用户设备；所述 PCP客户端获取用户设备的内部 地址和端口 , 将携带所述用户设备的内部地址和端口的 PCP请求发送至 PCP 服务器的步骤包括： 所述用户设备在控制信令的协议配置选项 （PCO ) 中携 带 PCP请求并将所述控制信令发送到公共数据网网关， 所述公共数据网网关 向 PCP服务器发送所述 PCP请求。
上述方法还可具有以下特点：
所述 PCP客户端是所述用户设备；
所述 PCP客户端获取用户设备的内部地址和端口， 将携带所述用户设备 的内部地址和端口的 PCP请求发送至 PCP服务器的步骤包括：
所述公共数据网网关在向所述用户设备发送的地址通知消息中添加 PCP 服务器地址；
所述用户设备获知所述 PCP服务器的地址，向所述 PCP服务器发送携带 所述用户设备的内部地址和端口的 PCP请求，所述公共数据网网关向 PCP服 务器发送所述 PCP请求。
上述方法还可具有以下特点：
所述 PCP客户端是应用服务器；
所述 PCP客户端获取用户设备的内部地址和端口， 将携带所述用户设备 的内部地址和端口的 PCP请求发送至 PCP服务器的步骤包括：
应用服务器判断所述用户设备需进行 NAT保活时 ,向策略计费规则功能 ( PCRF )发送携带所述用户设备的内部地址和端口的 PCP请求， 所述 PCRF 将所述 PCP请求发送至所述 PCP服务器。
上述方法还可具有以下特点：
所述应用服务器获取所述用户设备的内部地址和端口的步骤包括： 所述 用户设备在注册到所述应用服务器所在的网络时将其内部地址和端口上报给 所述应用服务器， 或者所述应用服务器经由所述 PCRF查询公共数据网网关 而获知用户设备的内部地址和端口。
上述方法还可具有以下特点：
所述 PCP客户端是策略计费规则功能（PCRF ) ； 所述 PCP客户端获取用户设备的内部地址和端口， 将携带所述用户设备 的内部地址和端口的 PCP请求发送至 PCP服务器的步骤包括： 所述 PCRF获 取所述用户设备的内部地址和端口， 判断所述用户设备需进行 NAT保活时 , 向所述 PCP服务器发送所述 PCP请求。
上述方法还可具有以下特点：
所述 PCRF获取所述用户设备的内部地址和端口的步骤包括： 所述应用 服务器将所述用户设备的内部地址和端口通知至所述 PCRF, 或者所述 PCRF 根据用户设备的标识查询公共数据网网关获知用户设备的内部地址和端口。 本发明实施例提供一种釆用端口控制协议完成网络地址转换保活的网元 设备， 所述网元设备包括第一模块和第二模块；
所述第一模块， 设置为： 获取用户设备的内部地址和端口；
所述第二模块， 设置为： 向端口控制协议（PCP )服务器发送携带用户 设备的内部地址和端口的 PCP请求。
上述网元设备还可具有以下特点：
所述网元设备为作为 PCP客户端的公共数据网网关；
所述第一模块是设置为： 从接收的所述控制信令中获知其协议配置选项
( PCO ) 中携带的所述用户设备的内部地址和端口。
上述网元设备还可具有以下特点：
所述网元设备为作为 PCP客户端的公共数据网网关；
所述第一模块是设置为： 通过媒体通道从基站接收所述用户设备发送的 所述用户设备的内部地址和端口。
上述网元设备还可具有以下特点：
所述网元设备为作为 PCP客户端的公共数据网网关；
所述第一模块是设置为：从核心网的控制网元接收所述用户设备的标识， 根据所述用户设备的标识获知所述用户设备的内部地址和端口。
上述网元设备还可具有以下特点：
所述网元设备为作为 PCP客户端的公共数据网网关； 所述第一模块是设置为： 从所述用户设备的上行数据中检测到所述用户 设备的信息， 所述信息中包括所述用户设备的内部地址和端口。
上述网元设备还可具有以下特点：
所述网元设备为作为 PCP客户端的应用服务器；
所述第一模块是设置为： 在所述用户设备注册到所述应用服务器所在的 网络时从所述用户设备获知所述用户设备的内部地址和端口， 或者经由所述 PCRF查询公共数据网网关而获知用户设备的内部地址和端口。
上述网元设备还可具有以下特点：
所述网元设备为作为 PCP客户端的策略计费规则功能（PCRF ) ； 所述第一模块是设置为： 从应用服务器接收携带所述用户设备的内部地 址和端口的 PCP请求， 或者根据用户设备的标识查询公共数据网网关获知用 户设备的内部地址和端口。
本发明实施例还提供一种用户设备， 所述用户设备包括信令发送模块； 所述信令发送模块， 设置为： 在控制信令的协议配置选项 （PCO ) 中携 带所述用户设备的内部地址和端口并将所述控制信令发送到公共数据网网 关。
本发明实施例还提供一种用户设备， 所述用户设备作为 PCP客户端包括 信令发送模块；
所述信令发送模块， 设置为： 构建携带所述用户设备的内部地址和端口 的 PCP请求， 在控制信令的协议配置选项（PCO )中携带 PCP请求并将所述 控制信令发送到公共数据网网关。
本发明实施例还提供一种用户设备， 所述用户设备作为端口控制协议
( PCP )客户端包括 PCP服务器地址获取模块和信令发送模块； 所述 PCP服务器地址获取模块， 设置为： 获知 PCP服务器的地址； 所述信令发送模块， 设置为： 向所述 PCP服务器发送携带所述用户设备 的内部地址和端口的 PCP请求。
上述网元设备还可具有以下特点： 所述 PCP服务器地址获取模块， 是设置为： 从公共数据网网关接收到的 消息的 PCO中获知 PCP服务器的地址，或者，从所述公共数据网网关接收地 址通知消息， 根据此地址通知消息的消息体中携带的地址获知所述 PCP服务 器的地址。
本方案釆用端口控制协议（PCP )可以减少用户设备与网络在空口上的 通信， 进而降低用户设备消耗在 NAT保活过程中的电池损耗。 附图概述
附图说明用来提供对本发明的进一步理解， 并且构成说明书的一部分， 与本发明的实施例一起用于解释本发明， 并不构成对本发明的限制。 在附图 中：
图 1为相关技术中 UE经过 3GPP网络与外部网络进行通信时经过 NAT 的网络架构和保活方法的示意图；
图 2为本发明实施例中釆用端口控制协议完成网络地址转换保活的方法 示意图；
图 3为实施例一中公共数据网网关作为 PCP客户端与 PCP服务端共同完 成 NAT保活的示意图；
图 4为实施例一的实现方式一中公共数据网网关在 UE的指示下与 PCP 服务器共同完成 NAT保活的流程图；
图 5为实施例一的实现方式二中公共数据网网关在 UE的指示下与 PCP 服务器共同完成 NAT保活的流程图；
图 6为实施例一的实现方式三中公共数据网网关在网络的指示下与 PCP 服务器共同完成 NAT保活的流程图；
图 7为实施例一的实现方式四中公共数据网网关基于本地的包检测从而 与 PCP服务器共同完成 NAT保活的流程图；
图 8为实施例二中 UE作为 PCP客户端与 PCP服务器共同完成 NAT保 活的示意图； 图 9为实施例二的实现方式一中 UE作为 PCP客户端与 PCP服务器共同 完成 NAT保活的流程图；
图 10为实施例二的实现方式二中 UE作为 PCP客户端与 PCP服务器共 同完成 NAT保活的流程图；
图 11为实施例三中 AF作为 PCP客户端与 PCP服务器共同完成 NAT保 活的示意图；
图 12为实施例三中 AF作为 PCP客户端与 PCP服务器共同完成 NAT保 活的流程图；
图 13为实施例四中 PCRF作为 PCP客户端与 PCP服务器共同完成 NAT 保活的示意图；
图 14为实施例四中 PCRF作为 PCP客户端与 PCP服务器共同完成 NAT 保活的流程图；
图 15为本发明实施例中一种网元设备的结构示意图；
图 16为本发明实施例中另一种网元设备的结构示意图。
本发明的较佳实施方式
下文中将结合附图对本发明的实施例进行详细说明。 需要说明的是， 在 不冲突的情况下， 本申请中的实施例及实施例中的特征可以相互任意组合。
如图 2所示，釆用端口控制协议完成网络地址转换保活的方法包括： PCP 客户端获取 UE的内部地址和端口，将携带所述 UE的内部地址和端口的 PCP 请求发送至端口控制协议（PCP )服务器， PCP服务器向 NAT设备发送携 带所述 UE的内部地址和端口的保活请求。
可作为 PCP客户端的网元设备包括第一模块和第二模块；
第一模块设置为获取 UE的内部地址和端口；
第二模块设置为向 PCP服务器发送携带 UE的内部地址和端口的 PCP请 求。 下面通过不同的实施例进行详细说明。
实施例一
实施例一中公共数据网网关作为 PCP客户端与 PCP服务器共同完成 NAT保活。 本实施例以 P-GW/GGSN作为公共数据网网关的具体网元， 以 MME/SGSN作为控制网元的具体网元。 如图 3所示， P-GW或者 GGSN判断 需要代替 UE进行 NAT保活， P-GW或者 GGSN向 PCP Server发送 PCP请求， 携带需要保活的 UE的私网 IP地址和端口。 PCP服务器代替 UE周期性向 NAT发送保活消息。 包括四种实现方式。
实现方式一
实现方式一中 UE经由核心网的 MME/SGSN将 UE的内部地址和端口发 送至 P-GW/GGSN, P-GW/GGSN从信令路径收到 UE发来的消息， 其中带需 要向 NAT保活的标识。 UE通过信令路径将 UE的内部 IP地址和端口发送到 基站 ,基站将 UE的内部 IP地址和端口发送到 MME/SGSN,然后 MME/SGSN 发送到 P-GW/GGSN。
如图 4所示， 当 UE发现其发送的数据包经过 NAT并需要进行 NAT保 活时，为了节省空口资源及节约 UE的电池损耗 , UE发起请求，让 P-GW/GGSN 执行保活操作。 例如， UE向 MME/SGSN发送消息， 其中在消息的 PCO中 携带 UE的内部 IP地址和端口； MME/SGSN将 PCO透传给 P-GW/GGSN。 P-GW/GGSN 与 PCP Server共同完成 NAT 保活。 业务结束后 UE 通过 MME/SGSN通知 P-GW/GGSN, 然后 P-GW/GGSN取消 PCP Server到 NAT 的保活过程。 包括步骤 400至 422:
步骤 400: UE附着到 3GPP网络， 连接到为其服务的核心网的控制网元 MME/SGSN, MME/SGSN为 UE选择 P-GW/GGSN, P-GW/GGSN为 UE分 配了内部 IP地址， UE通过 3GPP网络发起业务， 数据包经过 NAT发送到远 端。
步骤 401 : UE 向 MME/SGSN发送请求消息， 在消息的协议配置选项 ( Protocol Configuration Options , 简称 PCO ) 中携带其自身的内部 IP地址和 端口号， 可选的可以带期望的 timer , 记做 timer 1。 期望的 timer 1指 UE预测 的业务有效的时间，或者称为 UE在线的时间，可以将 timerl的设置为 limitless 或 unlimited , 该情况表示 UE—直在线。
具体实施时， 请求消息比如是 Create Session Request, 也可以是 Create Bearer Response,也可以是 Bearer Resource Command,也可以是 Update Bearer Response,也可以是： Create PDP Context Request message , 或者 Update PDP Context Request, 或者 Create MBMS Context Request message
步骤 402: MME/SGSN将请求消息发给 P-GW/GGSN。
需要说明的是， 如果是 MME发到 P-GW, 那么请求先发给 S-GW, 然后 S-GW转发到 P-GW。 为了描述简单， 图中没有画出 S-GW, 下面的实施例也 是一样。
步骤 403： P-GW/GGSN收到请求消息后， 向 PCP Server发送 PCP请求。 具体的， P-GW/GGSN解析 PCO带的 UE的内部 IP地址和端口， 将 UE的内 部 IP地址和端口放在在 PCP请求消息中 , 比如放在 PCP MAP消息或者 PCP PEER消息中 , PCP请求中携带一个 timer , 记做 timer2。 timer2的值可以设置 为 PCO携带的 timerl值， 也可以根据本地策略设置一个小于或者大于 PCO 携带的 timerl的值。
步骤 404: PCP Server向 NAT发送 keep-alive保活请求，携带 UE的内部 IP地址和端口， 携带期望的保活时长， 记做 timer3。
步骤 405： NAT向 PCP Server返回响应消息 , 携带 UE的内部 IP地址和 端口和 UE的外部 IP地址和端口， 还有保活的有效时长， 记做 timer3c, 通常 情况下 timer3c小于 timer2c的值。
步骤 406: PCP Server向 P-GW/GGSN发送 PCP响应消息， 带 UE的内 部 IP地址和端口和 UE的外部 IP地址和端口， 还有保活的有效时长， 记做 timsr2c。
步骤 407: P-GW/GGSN向 MME/SGSN发送响应消息。 响应消息中带
P-GW/GGSN为 UE做 NAT保活的有效时长， 记做 timerlc。
步骤 408: MME/SGSN将响应消息发送到 UE。
可选的， 步骤 407〜步骤 408可以在步骤 406之前执行。 此时 timerl c可 以为 limitless或 unlimited, 也可以不带 timerlc, 均表示 P-GW/GGSN可以一 直为 UE做 NAT保活。
步骤 409: P-GW/GGSN在 timer2c的时间周期内向 PCP Server发送 PCP 请求。 具体的， P-GW/GGSN向 PCP Server发送 PCP PEER消息， 携带一个 timer, 可以设为 timer2c的值。
步骤 410: PCP Server在 imer2c的时间周期内收到 PCP请求， 返回响应 消息， 新的超时时长可以是 timer2c的值。
步骤 411： PCP Server在超时时长 timer2c内， 以一定的周期向 NAT发送 保活消息， 该周期小于 timer3c的时长。
步骤 412: NAT在超时时长 timer3c时收到 PCP Server的保活消息，返回 响应消息， 显示保活成功， 新的超时时长可以是 timer3c。
在具体实施时， 步骤 409〜步骤 410是以一定的周期重复执行的， 执行周 期小于协商的时间 timer2c,步骤 411〜步骤 412也是另外一个周期重复执行的， 该周期小于协商的时间 timer3c。
通过上述过程 UE的内部 IP地址和端口和外部 IP地址和端口的映射关系 在 NAT上不断更新， UE到 NAT的保活由 P-GW/GGSN和 PCP Server来完 成。
步骤 413: UE到业务提供商的业务结束， UE向基站发送消息， 在 PCO 中带 UE的内部 IP地址和端口 , 超时时间 timerl设置为零， 表示取消到 NAT 的保活。
发送的消息比如是 Delete MBMS Context Request, 或者是 Delete PDP Context Request;比如是 Delete Session Request,或者是 Delete Bearer Request。
步骤 414: 同步骤 402。
步骤 415: P-GW/GGSN向 PCP Server发送 PCP请求， 携带的时间间隔 timer设为零。
步骤 416: PCP Server收到时间间隔为零的 PCP请求后， 返回响应消息， 不再为该 UE发送到 NAT的保活消息。
步骤 417〜步骤 418 : P-GW/GGSN 向 MME/SGSN 发送响应消息， MME/SGSN将响应消息转发给 UE。
实现方式二
实现方式二中 UE在与 P-GW/GGSN之间的媒体通道上将 UE的内部地址 和端口发送至 P-GW/GGSN, P-GW/GGSN从媒体路径收到 UE发来的消息， 其中带需要向 NAT保活的标识。 其中， UE通过媒体路径将 UE的内部 IP地 址和端口发送到基站，基站将 UE的内部 IP地址和端口发送到 P-GW/GGSN, 不经过 MME/SGSN。
如图 5所示， 当 UE发现其发送的数据包经过 NAT并需要进行 NAT保 活时， 让 P-GW/GGSN执行保活操作。 其中， UE向 P-GW/GGSN发送消息， 在消息中携带 UE的内部 IP地址和端口。 P-GW/GGSN与 PCP Server共同完 成 NAT保活。业务结束后 UE通知 P-GW/GGSN, P-GW/GGSN取消 PCP Server 到 NAT的保活过程。
UE获取 P-GW/GGSN地址的方法，比如是 P-GW/GGSN在发送到 UE的 GTP-U的消息的头部中添加并携带， 比如是在消息头部的扩张部分添加其自 身的地址。 或者比如是 P-GW/GGSN作为源地址向 UE发送一个通知消息， 携带其自身的地址， P-GW/GGSN地址可以放在 GTP-U的消息的头部中， 也 可以放在源地址中， 并在消息中标识这是一个来自 P-GW/GGSN的消息。
包括步骤 500至 514:
步骤 500: UE附着到 3GPP网络， 连接到为其服务的核心网的控制网元
MME/SGSN。 MME/SGSN为 UE选择 P-GW/GGSN, P-GW/GGSN为 UE分 配了内部 IP地址。 UE通过 3GPP网络发起业务， 数据包经过 NAT发送到远 端。
步骤 501： UE向 P-GW/GGSN发送消息，携带 UE的内部 IP地址和端口。 目的地是 P-GW/GGSN。 可选的， 可以带期望的 timer, 记做 timerl。 期望的 timerl指 UE预测的业务有效的时间 ,或者称为 UE在线的时间。可以将 timerl 的设置为 limitless或 unlimited , 也可以不带 timer , 该情况表示 UE—直在线。
消息的发送方式比如是在 GTP-U通道上发送到 P-GW/GGSN。 步骤 502〜步骤 505: 同实现方式一中的步骤 403〜步骤 406。
步骤 506： P-GW/GGSN向 UE返回响应消息。
步骤 507〜步骤 510: 同实现方式一中的步骤 409〜步骤 412。
步骤 511 : UE到业务提供商的业务结束， UE向 P-GW/GGSN发送消息， 带 UE的内部 IP地址和端口， 超时时间 timerl设置为零， 表示取消到 NAT 的保活。
步骤 512〜步骤 513: 同实现方式一中的步骤 415〜步骤 416。
步骤 514: P-GW/GGSN向 UE返回响应消息。
实现方式三
MME/SGSN判断 UE需进行 NAT保活后向 P-GW/GGSN发送 UE的标识， P-GW/GGSN根据 UE的标识获知 UE的内部地址和端口。具体的， MME/SGSN 通过本地配置或者 UE的注册信息判断 UE可以在业务期间一直保持在线状 态， 通知 P-GW/GGSN执行在业务期间为 UE进行保活操作。 MME/SGSN向 P-GW/GGSN发送消息， 其中在消息中携带 UE的标识。 P-GW/GGSN判断 UE处于业务进行状态， P-GW/GGSN与 PCP Server共同完成 NAT保活。
其中 UE的标识比如是 UE的移动用户国际综合业务数字网 （ ISDN )号 码 ( Mobile International Integrated Service Digital Network, 简称 MSISDN, ) , 公共用户标识（ Public User Identity, 简称 PUI ) , 私有用户标识（ Private User Identity, 简称 PVI ) , 别称（nickname )等。
如图 6所示， 本方法包括步骤 600至 614:
步骤 600: UE附着到 3GPP网络，连接到为其服务的核心网的控制网元， 比如是 MME/SGSN。 MME/SGSN为 UE选择 P-GW/GGSN, P-GW/GGSN为 UE分配了内部 IP地址 , UE通过 3GPP网络发起业务， 数据包经过 NAT发 送到远端。
步骤 601 : MME/SGSN向 P-GW/GGSN发送请求消息，携带 UE的标识。 可以携带期望的 timer, 记做 timerNet。 期望的 timerNet指 MME/SGSN预测 的业务有效的时间， 或者称为 UE允许在线的时间， 可以将 timerl的设置为 limitless或 unlimited , 该情况表示 UE—直在线。
步骤 602： P-GW/GGSN向 MME/SGSN返回响应消息。
步骤 603: P-GW/GGSN根据来自 MME/SGSN的请求消息中 UE的标识 可以获知 UE的内部 IP地址。
P-GW/GGSN获知 UE正在使用业务，需要进行 NAT保活， P-GW/GGSN 向 PCP Server发送 PCP 请求， 携带 UE 的内部 IP 地址和端口， 比如 P-GW/GGSN通过包检测获知该 UE的内部 IP地址和端口。
步骤 604〜步骤 606: 同步骤 605〜步骤 607。
步骤 607〜步骤 610: 同步骤 611〜步骤 614。
步骤 611 : MME/SGSN判断 UE的业务已经结束，或者 UE的 3GPP网络 注册超时， 即在规定的时间内没有收到 TAU 或者 RAU。 在上述情况下 MME/SGSN向 P-GW/GGSN发送取消保活的请求消息， 携带 UE的标识。 timerNet设置为零。
步骤 612: P-GW/GGSN向 MME/SGSN返回响应消息。
步骤 613: P-GW/GGSN向 PCP Server发送 PCP请求。 携带 UE的内部
IP地址和端口 , 时间间隔 timer设为零。
步骤 614: PCP Server收到时间间隔为零的 PCP请求后， 不再为该 UE 发送到 NAT的保活消息。
实现方式四
P-GW/GGSN从 UE的上行数据中检测到 UE的信息， 此信息中包括 UE 的内部地址和端口，根据此信息判断 UE需进行 NAT保活时向 PCP服务器发 送 PCP请求， P-GW/GGSN与 PCP服务器通信， 共同完成 NAT保活。
如图 7所示， 本方法包括步骤 701至 713:
步骤 700: UE附着到 3GPP网络，连接到为其服务的核心网的控制网元， 比如是 MME/SGSN。 MME/SGSN为 UE选择 P-GW/GGSN, P-GW/GGSN为 UE分配了内部 IP地址， UE通过 3GPP网络发起业务， 数据包经过 NAT发 送到远端。
步骤 701 : P-GW/GGSN对 UE的上行和下行数据进行包检测。 从上行数 据中检测到源 IP地址、 端口 （即 UE的内部 IP地址和端口）和目标 IP地址、 端口， 以及协议类型， 即五元组。
步骤 702: P-GW/GGSN根据检测到的五元组信息判断需要为 UE进行
NAT保活操作。
步骤 703 : P-GW/GGSN 向 PCP Server 发送 PCP 请求。 具体的， P-GW/GGSN将五元组中的 UE的内部 IP地址和端口 , 放在在 PCP请求消息 中的内部 IP地址和端口中。 具体消息比如是 PCP MAP消息或者 PCP PEER 消息， PCP请求中携带期望的超时时间， 记做 timer2。
步骤 704: PCP Server向 NAT发送 keep-alive保活请求，携带 UE的内部 IP地址和端口， 携带期望的保活时长， 记做 timer3。
步骤 705： NAT向 PCP Server返回响应消息 , 携带 UE的内部 IP地址和 端口和 UE的外部 IP地址和端口， 还有保活的有效时长， 记做 timer3c, 通常 情况下 timer3 c小于 timer2c的值。
步骤 706: PCP Server向 P-GW/GGSN发送 PCP响应消息， 带 UE的内 部 IP地址和端口和 UE的外部 IP地址和端口， 还有保活的有效时长， 记做 timsr2c。
步骤 707〜步骤 710: 同步骤 411〜步骤 414。
步骤 711 : P-GW/GGSN在设定的时间内没有收到该 UE的上行或下行的 业务数据包。
步骤 712: P-GW/GGSN向 PCP Server发送 PCP请求消息， 携带 UE的 内部 IP地址和端口， 时间间隔 timer设为零。
步骤 713: PCP Server收到时间间隔为零的 PCP请求后， 不再为该 UE 发送到 NAT的保活消息。
实施例二 图 8是 UE作为 PCP客户端与 PCP服务器共同完成 NAT保活的示意图 , 如图所示， UE自身作为 PCP客户端， 发送 PCP请求到 PCP服务器。 有两 种方法可以将 UE的 PCP请求路由到 PCP服务器： 实现方式一中 UE在控制 信令消息的 PCO 中携带 PCP 请求消息并将控制信令消息发送至 P-GW/GGSN, P-GW/GGSN将其中的 PCP请求消息发送到 PCP服务器； 实 现方式二中， UE获知 P-GW/GGSN的地址，将 PCP请求发送到 P-GW/GGSN, 然后 P-GW/GGSN查询并选择一个 PCP服务器，将 UE发来的 PCP请求路由 到该 PCP服务器。
实现方式一
UE在控制信令消息的 PCO中携带 PCP请求消息， 向 P-GW/GGSN发送 携带 UE的内部地址和端口的控制信令消息， P-GW/GGSN解析出其中的 PCP 请求发送到 PCP服务器。
本方法包括步骤 900至 922:
步骤 900: UE附着到 3GPP网络， 连接到为其服务的核心网的控制网元 MME/SGSN, MME/SGSN为 UE选择 P-GW/GGSN, P-GW/GGSN为 UE分 配了内部 IP地址， UE通过 3GPP网络发起业务， 数据包经过 NAT发送到远 端。
步骤 901 : UE向 MME/SGSN发送消息， 在消息的 PCO中携带 PCP请 求消息，比如将 PCP MAP消息或者 PCP PEER消息放在消息的 PCO中， PCP 请求消息中携带 UE的内部 IP地址和端口， 可选的可以带期望的 timer, 记做 timerl , 期望的 timerl指 UE预测的业务有效的时间 , 或者称为 UE在线的时 间， 可以将 timerl的设置为 limitless或 unlimited, 也可以不带 timer, 该情况 表示 UE—直在线。
具体实施时，发送的消息比如是 Create Session Request, 也可以是 Create Bearer Response ,也可以是 Bearer Resource Command ,也可以是 Update Bearer Response,也可以是： Create PDP Context Request message , 或者 Update PDP Context Request, 或者 Create MBMS Context Request message。 在上述消息的 PCO中带 PCP请求消息。 步骤 902: MME/SGSN将消息转发给 P-GW/GGSN。
步骤 903: P-GW/GGSN解析消息中 PCO, 然后将其中携带 PCP请求消 息发送到 PCP Server。
步骤 904: PCP Server向 NAT发送 keep-alive保活请求，携带 UE的内部 IP地址和端口， 携带期望的保活时长， 记做 timer2。
步骤 905： NAT向 PCP Server返回响应消息 , 携带 UE的内部 IP地址和 端口和 UE的外部 IP地址和端口， 还有保活的有效时长， 记做 timer2c, 通常 情况下 timer2c小于 timerl的值。
步骤 906: PCP Server向 P-GW/GGSN发送 PCP响应消息， 带 UE的内 部 IP地址和端口和 UE的外部 IP地址和端口， 还有保活的有效时长， 记做 timerl c„
步骤 907： P-GW/GGSN向 MME/SGSN转发响应消息。响应消息中带 PCP Server为 UE做 NAT保活的有效时长，记做 timerlc, 比如将 PCP响应消息放 在 PCO中发送给 UE。
步骤 908: MME/SGSN将消息发送到 UE, UE解析 PCO带的 PCP响应 消息。
步骤 909： UE在 timerl c的时间周期内向 MME/SGSN发送消息， 在消息 的 PCO中携带 PCP请求消息， 比如将 PCP PEER消息放在消息的 PCO中， PCP请求消息中带 UE的内部 IP地址和端口， 携带预期的超时时长 timer, 可 以设为 timerlc的值。
步骤 910: MME/SGSN将消息转发给 P-GW/GGSN。
步骤 911 : P-GW/GGSN解析消息中 PCO, 然后将其中携带 PCP请求消 息发送到 PCP Server。
步骤 912〜步骤 914: 同步骤 906〜步骤 908。
步骤 915: PCP Server在超时时长 timer2c内， 以一定的周期向 NAT发送 保活消息， 该周期小于 timerlc的时长。
步骤 916: NAT在超时时长 timer2c时收到 PCP Server的保活消息，返回 响应消息， 显示保活成功， 新的超时时长可以是 timer3c。 在具体实施时， 步骤 909〜步骤 914是以一定的周期重复执行的， 执行周 期小于等于协商的时间 timerlc,步骤 915〜步骤 916也是另外一个周期重复执 行的， 该周期小于协商的时间 timer2c。
通过上述过程 UE的内部 IP地址和端口和外部 IP地址和端口的映射关系 在 NAT上不断更新 , 实现 UE到 NAT的保活。
步骤 917: UE的业务结束， UE向 MME/SGSN发送消息，在消息的 PCO 中携带 PCP请求消息， 比如将 PCP MAP消息或者 PCP PEER消息放在消息 的 PCO中。 PCP请求消息中带 UE的内部 IP地址和端口， 超时时间 timerl 设置为零， 表示取消到 NAT的保活。
步骤 918: MME/SGSN将消息转发给 P-GW/GGSN。 。
步骤 919: PCP Server收到时间间隔为零的 PCP请求后， 不再为该 UE 发送到 NAT的保活消息， 向 P-GW/GGSN返回响应消息。
步骤 920〜步骤 922: 同步骤 906〜步骤 908。
实现方式二
UE获知 PCP服务器的地址， 向 PCP服务器发送携带 UE的内部地址和 端口的 PCP请求。在本实施例中 P-GW/GGSN为 UE选择 PCP服务器并且将 PCP服务器的地址发送给 UE。
如图 10所示， 本方法包括步骤 1000至 1022:
步骤 1000: UE附着到 3GPP网络，连接到为其服务的核心网的控制网元
MME/SGSN, MME/SGSN为 UE选择 P-GW/GGSN, P-GW/GGSN为 UE分 配了内部 IP地址。
P-GW/GGSN将其 PCP服务器的地址通知给 UE, UE将 PCP请求消息发 送给该 PCP服务器
P-GW/GGSN将 PCP Server的地址通知给 UE的方法是 P-GW/GGSN在 发送到 UE的承载控制消息中添加 PCP Server的地址， 可以在消息的 PCO中 添加 PCP Server的地址， 也可以在消息体中以 XML格式携带 PCP服务器的 地址。 具体实施时， 在承载控制消息的 PCO中带 PCP Server地址的请求消息 比^口是 Create Session Response, 也可以是 Create Bearer Request, 也可以是 Modify Bearer Response , 也可以是 Update Bearer Request , 也可以是： Create PDP Context Response , 或者 Update PDP Context Response, 或者 Create MBMS Context Response。
步骤 1001 : UE 向 MME/SGSN 发送承载控制消息， 建立 UE 与
P-GW/GGSN的承载通道。
步骤 1002: MME/SGSN将承载控制消息发送至 P-GW/GGSN。
步骤 1003: P-GW/GGSN向 UE返回响应消息， 在响应消息的 PCO中带 PCP Server地址。
步骤 1004: MME/SGSN将响应消息转发给 UE。
步骤 1005: UE向 P-GW/GGSN送 PCP请求消息， 在消息中携带 UE的 内部 IP地址和端口， 目的地址是 PCP服务器。 比如发送 PCP MAP消息或者 PCP PEER消息中。 可选的可以带期望的 timer, 记做 timerl。 期望的 timerl 指 UE预测的业务有效的时间，或者称为 UE在线的时间。可以将 timerl的设 置为 limitless或 unlimited, 也可以不带 timer, 该情况表示 UE—直在线。
步骤 1006: P-GW/GGSN将 PCP请求消息转发到 PCP Server。
步骤 1007: PCP Server向 NAT发送 keep-alive保活请求， 携带 UE的内 部 IP地址和端口， 携带期望的保活时长， 记做 timer2。
步骤 1008: NAT向 PCP Server返回响应消息 , 携带 UE的内部 IP地址 和端口和 UE的外部 IP地址和端口， 还有保活的有效时长， 记做 timer2c， 通 常' f青况下 timer2c小于 timerl的值。
步骤 1009: PCP Server返回 PCP响应消息， 带 UE的内部 IP地址和端口 和 UE的外部 IP地址和端口， 还有保活的有效时长， 记做 timerlc。
步骤 1010： P-GW/GGSN向 UE转发 PCP响应消息， 响应消息中带 PCP Server为 UE做 NAT保活的有效时长， 记做 timerlc。
步骤 1011 : UE在 timerlc的时间周期内向 PCP服务器发送 PCP请求， 比如发送 PCP PEER消息， 携带预期的超时时长 timer, 可以设为 timerlc的 值。 步骤 1012: P-GW/GGSN将 PCP请求消息转发 PCP服务器。 步骤 1013〜步骤 1014: 同步骤 1009〜步骤 1010。
步骤 1015: PCP Server在超时时长 timer2c内， 以一定的周期向 NAT发 送保活消息。
步骤 1016: NAT在超时时长 timer2c时收到 PCP Server的保活消息， 返 回响应消息， 显示保活成功。
在具体实施时， 步骤 1011〜步骤 1014是以一定的周期重复执行的， 执行 周期小于等于协商的时间 timerlc,步骤 1015〜步骤 1016也是另外一个周期重 复执行的， 该周期小于协商的时间 timer2c。
通过上述过程 UE的内部 IP地址和端口和外部 IP地址和端口的映射关系 在 NAT上不断更新。
步骤 1017: UE的业务结束， UE P-GW/GGSN送 PCP请求消息， 在消息 中携带 UE的内部 IP地址和端口， 目的地址是 PCP服务器。 超时时间 timerl 设置为零， 表示取消到 NAT的保活。
步骤 1018: PCP服务器收到时间间隔为零的 PCP请求后， 不再为该 UE 发送到 NAT的保活消息。
步骤 1019〜步骤 1020: PCP服务器返回响应消息。
实施例三
如图 11所示， 应用服务器（AF )作为 PCP客户端判断 UE需进行 NAT 保活时， 向策略计费规则功能（PCRF )发送携带所述 UE的内部地址和端口 的 PCP请求， PCRF将 PCP请求发送至 PCP服务器。 PCRF收到 PCP请求后 执行 PCP Proxy的功能， 将 PCP请求转发到 PCP服务器， PCP服务器代替 UE向 NAT发送保活消息。
需要说明的是 AF获取 UE的内部 IP地址和端口， 可以是在 UE注册到
AF所在的网络的时候由 UE上报， 也可以是 AF经过 PCRF从 P-GW/GGSN 查询获得。 如图 12所示， 本方法包括步骤 1200至 1216:
步骤 1200: UE附着到 3GPP网络， 连接到为其服务的核心网控制网元 MME/SGSN。 MME/SGSN为 UE选择 P-GW/GGSN, P-GW/GGSN为 UE分 配了内部 IP地址。 UE通过 3GPP网络发起业务， 数据包经过 NAT发送到远 端。
步骤 1201 : AF向 PCRF发送 PCP请求， 携带 UE的 IP地址和端口， 比 如发送 PCP MAP消息或者 PCP PEER消息中； 可选的可以带期望的 timer, 记做 timerl , 期望的 timerl指 UE预测的业务有效的时间 , 或者称为 UE在线 的时间， 可以将 timerl的设置为 limitless或 unlimited, 也可以不带 timer, 该 情况表示 UE—直在线。
步骤 1202: PCRF将 PCP请求转发给 PCP Server, PCRF将 PCP请求路 由到 PCP Server的过程可以参考 IETF中有关 PCP的协议， 在此不赘述。
步骤 1203： PCP Server向 NAT发送 keep-alive保活请求， 携带 UE的内 部 IP地址和端口， 携带期望的保活时长， 记做 timer2。
步骤 1204: NAT向 PCP Server返回响应消息， 携带 UE的内部 IP地址 和端口和 UE的外部 IP地址和端口， 还有保活的有效时长， 记做 timer2c, 通 常' f青况下 timer2c小于 timerl的值。
步骤 1205: PCP Server向 PCRF发送 PCP响应消息。 带 UE的内部 IP地 址和端口和 UE的外部 IP地址和端口 , 还有保活的有效时长， 记做 timerlc。
步骤 1206: PCRF将 PCP响应消息转发给 AF。
步骤 1207: AF在 timer lc的时间周期内向 PCRF发送 PCP请求，具体的， AF向 PCRF发送 PCP PEER消息， 携带一个 timer , 可以设为 timer 1 c的值。
步骤 1208: PCRF将 PCP请求路由到 PCP Server。
步骤 1209: PCP Server在 imerlc的时间周期内收到 PCP请求，返回响应 消息， 新的超时时长可以是 timerlc的值。
步骤 1210: PCRF将响应消息转发给 AF。
步骤 1211 : PCP Server在与 AF的超时时长 timerlc内， 以一定的周期向 NAT发送保活消息， 该周期小于与 NAT的保活时长 timer2c。 步骤 1212: NAT在超时时长 timer2c时收到 PCP Server的保活消息， 返 回响应消息， 显示保活成功， 新的超时时长可以是 timer2c。
在具体实施时， 步骤 1207〜步骤 1210是以一定的周期重复执行的， 执行 周期小于协商的时间 timerlc,步骤 1211〜步骤 1212也是另外一个周期重复执 行的， 该周期小于协商的时间 timer2c。
通过上述过程 UE的内部 IP地址和端口和外部 IP地址和端口的映射关系 在 NAT上不断更新 , UE到 NAT的保活由 AF、 PCRF和 PCP Server来完成。
步骤 1213: UE的业务结束， AF向 PCRF发送消息， 带 UE的内部 IP地 址和端口， 超时时间 timerl设置为零， 表示取消到 NAT的保活过程。
步骤 1214: PCRF将 PCP请求路由到 PCP Server。
步骤 1215: PCP Server收到时间间隔为零的 PCP请求后，返回 PCP响应 消息。 并且 PCP Server不再为该 UE发送到 NAT的保活消息。
步骤 1216: PCRF将 PCP响应转发给 AF。
实施例四
如图 13所示， PCRF作为 PCP客户端获取 UE的内部 IP地址和端口，判 断 UE需进行 NAT保活时 , 将携带 UE内部 IP地址和端口的 PCP请求发送 到 PCP服务器， PCP服务器代替 UE向 NAT发送保活消息。
需要说明的是 PCRF获取 UE的内部 IP地址和端口， 可以是在 UE发起 业务的时候由 AF通知 PCRF ,也可以是 PCRF根据 UE的标识从 P-GW/GGSN 查询获知。 比如， PCRF根据 AF发来的 UE的标识向 P-GW/GGSN查询该 UE的内部 IP地址和端口。所述 UE的标识比如是 UE的移动用户国际综合业 务数字网（ ISDN )号码（ Mobile International Integrated Service Digital Network, 简称 MSISDN, ) , 公共用户标识（ Public User Identity, 简称 PUI ) , 私有 用户标识（ Private User Identity, 简称 PVI ) , 别称（nickname )等。
PCRF选择 PCP服务器的方法， 可以参考 IETF中有关 PCP Server选择 的 RFC文档， 此处不赘述。
如图 14所示， 本方法包括步骤 1400至 1410: 步骤 1400: UE附着到 3GPP网络， 连接到为其服务的核心网控制网元 MME/SGSN。 MME/SGSN为 UE选择 P-GW/GGSN, P-GW/GGSN为 UE分 配了内部 IP地址。 UE通过 3GPP网络发起业务， 数据包经过 NAT发送到远 端。
步骤 1401： PCRF向 PCP Server携带 UE的 IP地址和端口。比如发送 PCP
MAP消息或者 PCP PEER消息中。 可选的可以带期望的 timer, 记做 timerl。 期望的 timerl指 UE预测的业务有效的时间，或者称为 UE在线的时间。可以 将 timerl的值设置为 limitless或 unlimited,也可以不带 timer,该情况表示 UE 一直在线。
步骤 1402: PCP Server向 NAT发送 keep-alive保活请求， 携带 UE的内 部 IP地址和端口， 携带期望的保活时长， 记做 timer2。
步骤 1403： NAT向 PCP Server返回响应消息， 携带 UE的内部 IP地址 和端口和 UE的外部 IP地址和端口， 还有保活的有效时长， 记做 timer2c, 通 常' 1"青况下 timer2c小于 timerl的值。
步骤 1404： PCP Server向 PCRF发送 PCP响应消息， 带 UE的内部 IP地 址和端口和 UE的外部 IP地址和端口， 还有保活的有效时长， 记做 timerlc。
步骤 1405: PCRF在 timerlc的时间周期内向 PCP Server发送 PCP请求， PCRF向 PCP Server发送 PCP PEER消息， 携带一个 timer, 可以设为 timerlc 的值。
步骤 1406: PCP Server在 imerlc的时间周期内收到 PCP请求， 返回响应 消息， 新的超时时长可以是 timerlc的值。
步骤 1407: PCP Server在与 PCRF的超时时长 timerlc内， 以一定的周期 向 NAT发送保活消息， 该周期小于与 NAT的保活时长 timer2c。
步骤 1408: NAT在超时时长 timer2c时收到 PCP Server的保活消息， 返 回响应消息， 显示保活成功， 新的超时时长可以是 timer2c。
在具体实施时， 步骤 1405〜步骤 1406是以一定的周期重复执行的， 执行 周期小于协商的时间 timerlc,步骤 1407〜步骤 1408也是另外一个周期重复执 行的， 该周期小于协商的时间 timer2c。 通过上述过程 UE的内部 IP地址和端口和外部 IP地址和端口的映射关系 在 NAT上不断更新 , UE到 NAT的保活由 PCRF和 PCP Server来完成。
步骤 1409: UE的业务结束， PCRF向 PCP Server发送消息， 带 UE的内 部 IP地址和端口， 超时时间 timerl设置为零， 表示取消到 NAT的保活过程。
步骤 1410: PCP Server收到时间间隔为零的 PCP请求后，返回 PCP响应 消息， 并且 PCP Server不再为该 UE发送到 NAT的保活消息。
对本方案中的网元设备描述如下：
如图 15所示， 网元设备包括： 第一模块， 设置为： 获取用户设备的内部 地址和端口； 第二模块， 设置为： 向 PCP服务器发送携带用户设备的内部地 址和端口的 PCP请求。 作为 PCP客户端的网元设备为公共数据网网关时，网元设备的第一模块， 设置为通过核心网的控制网元接收 UE发送的 UE的内部地址和端口，或者通 过媒体通道从基站接收 UE发送的 UE的内部地址和端口。第一模块可以接收 控制信令并从控制信令的协议配置选项 （PCO ) 中获知 UE 的内部地址和端 口， 还可以通过媒体通道从基站接收所述用户设备发送的所述用户设备的内 部地址和端口。 公共数据网网关不作 PCP客户端时（例如 UE作为 PCP客户 端）可以转发 PCP请求消息， 第一模块设置为接收控制信令并从此控制信令 的协议配置选项 （PCO ) 中获知其携带的 PCP请求消息， PCP请求消息中携 带 UE的内部地址和端口。
或者， 作为 PCP客户端的网元设备为公共数据网网关时， 网元设备的第 一模块设置为从核心网的控制网元接收 UE的标识， 根据 UE的标识获知 UE 的内部地址和端口。
或者， 作为 PCP客户端的网元设备为公共数据网网关时， 网元设备的第 一模块设置为从 UE的上行数据中检测到 UE的信息，此信息中包括 UE的内 部地址和端口。
作为 PCP客户端的网元设备是应用服务器时， 网元设备的第一模块设置 为在 UE注册到应用服务器所在的网络时从 UE获知 UE的内部地址和端口， 或者经由 PCRF查询公共数据网网关而获知 UE的内部地址和端口。
作为 PCP客户端的网元设备是 PCRF时， 网元设备的第一模块设置为从 应用服务器接收携带 UE的内部地址和端口的 PCP请求， 或者根据 UE的标 识从公共数据网网关查询获知 UE的内部地址和端口。
如图 16所示，作为 PCP客户端的网元设备是 UE时，包括信令发送模块， 信令发送模块设置为构建携带 UE的内部地址和端口的 PCP请求， 在控制信 令的协议配置选项 （PCO ) 中携带 PCP请求并将此控制信令发送到公共数据 网网关。
或者， 作为 PCP客户端的网元设备是 UE时， UE包括 PCP服务器获取 模块和信令发送模块， 网关地址获取模块设置为获知 PCP服务器的地址， 信 令发送模块设置为向所述 PCP服务器发送携带所述用户设备的内部地址和端 口的 PCP请求。 其中， 所述 PCP服务器地址获取模块， 设置为从公共数据网 网关接收到的消息的 PCO中获知 PCP服务器的地址，或者，从所述公共数据 网网关接收地址通知消息 , 根据此地址通知消息的消息体中携带的地址获知 所述 PCP服务器的地址。
UE不作为 PCP客户端时， UE的信令发送模块设置为在控制信令的协议 配置选项 （PCO ) 中携带 UE 的内部地址和端口并将控制信令发送到公共数 据网网关。 当然， 发明 可有其他多 实施例，，在不背离本发，精神及其实质的 但这些相应的改变和变形都应属于本发明所附的权利要求的保护范围。
本领域普通技术人员可以理解上述方法中的全部或部分步骤可通过程序 来指令相关硬件完成， 所述程序可以存储于计算机可读存储介质中， 如只读 存储器、 磁盘或光盘等。 可选地， 上述实施例的全部或部分步骤也可以使用 一个或多个集成电路来实现。 相应地， 上述实施例中的各模块 /单元可以釆用 硬件的形式实现， 也可以釆用软件功能模块的形式实现。 本发明不限制于任 何特定形式的硬件和软件的结合。 工业实用性
本发明实施例釆用端口控制协议（PCP )可以减少用户设备与网络在空 口上的通信， 进而降低用户设备消耗在 NAT保活过程中的电池损耗。

Claims

权 利 要 求 书
1、 完成网络地址转换保活的方法， 所述方法包括：
端口控制协议（PCP )客户端获取用户设备的内部地址和端口， 将携带 所述用户设备的内部地址和端口的 PCP请求发送至端口控制协议（ PCP )服 务器， 所述 PCP服务器向网络地址转换 ( NAT )设备发送携带所述用户设备 的内部地址和端口的保活请求。
2、 如权利要求 1所述的方法， 其中，
所述 PCP客户端是公共数据网网关；
所述 PCP客户端获取所述用户设备的内部地址和端口的步骤包括： 所述 公共数据网网关接收所述用户设备发送的控制信令， 所述控制信令的协议配 置选项 （PCO ) 中携带所述用户设备的内部地址和端口。
3、 如权利要求 1所述的方法， 其中，
所述 PCP客户端是公共数据网网关；
所述 PCP客户端获取所述用户设备的内部地址和端口的步骤包括： 所述 公共数据网关接收基站在媒体通道上发送的所述用户设备的内部地址和端 口， 其中， 所述用户设备的内部地址和端口是所述用户设备发送给基站的。
4、 如权利要求 1所述的方法， 其中，
所述 PCP客户端是公共数据网网关；
所述 PCP客户端获取所述用户设备的内部地址和端口的步骤包括： ， 所 述公共数据网网关根据核心网的控制网元发送的所述用户设备的标识获知所 述用户设备的内部地址和端口， 其中， 所述核心网的控制网元是在判断所述 用户设备需进行 NAT保活后向所述公共数据网网关发送所述用户设备的标 识。
5、 如权利要求 1所述的方法， 其中，
所述 PCP客户端是公共数据网网关；
所述 PCP客户端获取用户设备的内部地址和端口将携带所述用户设备的 内部地址和端口的 PCP请求发送至所述 PCP服务器的步骤包括：所述公共数 据网网关从所述用户设备的上行数据中检测到所述用户设备的信息， 所述信 息中包括所述用户设备的内部地址和端口， 根据所述信息判断所述用户设备 需进行 NAT保活时向所述 PCP服务器发送所述 PCP请求。
6、 如权利要求 1所述的方法， 其中，
所述 PCP客户端是所述用户设备；所述 PCP客户端获取用户设备的内部 地址和端口， 将携带所述用户设备的内部地址和端口的 PCP请求发送至 PCP 服务器的步骤包括： 所述用户设备在控制信令的协议配置选项 （PCO ) 中携 带 PCP请求并将所述控制信令发送到公共数据网网关， 所述公共数据网网关 向 PCP服务器发送所述 PCP请求。
7、 如权利要求 1所述的方法， 其中，
所述 PCP客户端是所述用户设备；
所述 PCP客户端获取用户设备的内部地址和端口， 将携带所述用户设备 的内部地址和端口的 PCP请求发送至 PCP服务器的步骤包括：
所述公共数据网网关在向所述用户设备发送的地址通知消息中添加 PCP 服务器地址；
所述用户设备获知所述 PCP服务器的地址，向所述 PCP服务器发送携带 所述用户设备的内部地址和端口的 PCP请求，所述公共数据网网关向 PCP服 务器发送所述 PCP请求。
8、 如权利要求 1所述的方法， 其中，
所述 PCP客户端是应用服务器；
所述 PCP客户端获取用户设备的内部地址和端口， 将携带所述用户设备 的内部地址和端口的 PCP请求发送至 PCP服务器的步骤包括：
应用服务器判断所述用户设备需进行 NAT保活时 ,向策略计费规则功能 ( PCRF )发送携带所述用户设备的内部地址和端口的 PCP请求， 所述 PCRF 将所述 PCP请求发送至所述 PCP服务器。
9、 如权利要求 8所述的方法， 其中，
所述应用服务器获取所述用户设备的内部地址和端口的步骤包括： 所述 用户设备在注册到所述应用服务器所在的网络时将其内部地址和端口上报给 所述应用服务器， 或者所述应用服务器经由所述 PCRF查询公共数据网网关 而获知用户设备的内部地址和端口。
10、 如权利要求 1所述的方法， 其中，
所述 PCP客户端是策略计费规则功能（ PCRF ) ；
所述 PCP客户端获取用户设备的内部地址和端口， 将携带所述用户设备 的内部地址和端口的 PCP请求发送至 PCP服务器的步骤包括： 所述 PCRF获 取所述用户设备的内部地址和端口， 判断所述用户设备需进行 NAT保活时 , 向所述 PCP服务器发送所述 PCP请求。
11、 如权利要求 10所述的方法， 其中，
所述 PCRF获取所述用户设备的内部地址和端口的步骤包括： 所述应用 服务器将所述用户设备的内部地址和端口通知至所述 PCRF, 或者所述 PCRF 根据用户设备的标识查询公共数据网网关获知用户设备的内部地址和端口。
12、 完成网络地址转换保活的网元设备， 所述网元设备包括第一模块和 第二模块； 其中，
所述第一模块， 设置为： 获取用户设备的内部地址和端口；
所述第二模块， 设置为： 向端口控制协议（PCP )服务器发送携带用户 设备的内部地址和端口的 PCP请求。
13、 如权利要求 12所述的网元设备， 其中，
所述网元设备为作为 PCP客户端的公共数据网网关；
所述第一模块是设置为： 从接收的所述控制信令中获知其协议配置选项 ( PCO ) 中携带的所述用户设备的内部地址和端口。
14、 如权利要求 12所述的网元设备， 其中，
所述网元设备为作为 PCP客户端的公共数据网网关；
所述第一模块是设置为： 通过媒体通道从基站接收所述用户设备发送的 所述用户设备的内部地址和端口。
15、 如权利要求 12所述的网元设备， 其中， 所述网元设备为作为 PCP客户端的公共数据网网关；
所述第一模块是设置为：从核心网的控制网元接收所述用户设备的标识， 根据所述用户设备的标识获知所述用户设备的内部地址和端口。
16、 如权利要求 12所述的网元设备， 其中，
所述网元设备为作为 PCP客户端的公共数据网网关；
所述第一模块是设置为： 从所述用户设备的上行数据中检测到所述用户 设备的信息， 所述信息中包括所述用户设备的内部地址和端口。
17、 如权利要求 12所述的网元设备， 其中，
所述网元设备为作为 PCP客户端的应用服务器；
所述第一模块是设置为： 在所述用户设备注册到所述应用服务器所在的 网络时从所述用户设备获知所述用户设备的内部地址和端口， 或者经由所述
PCRF查询公共数据网网关而获知用户设备的内部地址和端口。
18、 如权利要求 12所述的网元设备， 其中，
所述网元设备为作为 PCP客户端的策略计费规则功能（PCRF ) ； 所述第一模块是设置为： 从应用服务器接收携带所述用户设备的内部地 址和端口的 PCP请求， 或者根据用户设备的标识查询公共数据网网关获知用 户设备的内部地址和端口。
19、 一种用户设备， 所述用户设备包括信令发送模块；
所述信令发送模块， 设置为： 在控制信令的协议配置选项 （PCO ) 中携 带所述用户设备的内部地址和端口并将所述控制信令发送到公共数据网网 关。
20、 一种用户设备， 所述用户设备作为端口控制协议（PCP )客户端包 括信令发送模块；
所述信令发送模块， 设置为： 构建携带所述用户设备的内部地址和端口 的 PCP请求， 在控制信令的协议配置选项（PCO )中携带 PCP请求并将所述 控制信令发送到公共数据网网关。
21、 一种用户设备， 所述用户设备作为端口控制协议（PCP )客户端包 括 PCP服务器地址获取模块和信令发送模块；
所述 PCP服务器地址获取模块， 设置为： 获知 PCP服务器的地址； 所述信令发送模块， 设置为： 向所述 PCP服务器发送携带所述用户设备 的内部地址和端口的 PCP请求。
22、 如权利要求 21所述的用户设备， 其中，
所述 PCP服务器地址获取模块， 是设置为： 从公共数据网网关接收到的 消息的 PCO中获知 PCP服务器的地址，或者，从所述公共数据网网关接收地 址通知消息， 根据此地址通知消息的消息体中携带的地址获知所述 PCP服务 器的地址。
PCT/CN2013/084955 2012-12-25 2013-10-10 完成网络地址转换保活的方法及设备 WO2014101525A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201210572628.X 2012-12-25
CN201210572628.XA CN103906037A (zh) 2012-12-25 2012-12-25 采用端口控制协议完成网络地址转换保活的方法及设备

Publications (1)

Publication Number Publication Date
WO2014101525A1 true WO2014101525A1 (zh) 2014-07-03

Family

ID=50997145

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2013/084955 WO2014101525A1 (zh) 2012-12-25 2013-10-10 完成网络地址转换保活的方法及设备

Country Status (2)

Country Link
CN (1) CN103906037A (zh)
WO (1) WO2014101525A1 (zh)

Families Citing this family (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN106161534B (zh) * 2015-04-09 2021-05-11 中兴通讯股份有限公司 发送、传递和获取能力的方法及装置
CN108989271B (zh) * 2017-06-05 2022-06-10 中兴通讯股份有限公司 一种家庭网关端口防攻击的方法和装置
CN108848144B (zh) * 2018-05-31 2021-03-30 中国联合网络通信集团有限公司 区块链中节点访问方法、装置及区块链节点
CN115499409B (zh) * 2022-09-29 2024-07-26 阿里巴巴（中国）有限公司 Nat网关、服务器和网络系统

Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101026567A (zh) * 2007-01-29 2007-08-29 华为技术有限公司 地址转发表项的保活方法及系统
US20090213792A1 (en) * 2008-01-10 2009-08-27 Samsung Electronics Co., Ltd. Apparatus and method for setting a default gateway address in a mobile communication system
CN102685261A (zh) * 2011-03-15 2012-09-19 中国移动通信集团公司 一种控制设备的地址映射状态的方法、系统及装置

Family Cites Families (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
DE602004008415T2 (de) * 2004-02-25 2007-12-20 Research In Motion Ltd., Waterloo System und Verfahren zum Aufrechterhalten der Netzwerkverbindung
CN101119299A (zh) * 2006-08-02 2008-02-06 华为技术有限公司 导通媒体流的方法、导通检测方法及其系统
CN102447630B (zh) * 2011-12-28 2018-02-27 中兴通讯股份有限公司 协议报文的传输方法、家庭网关及运营商级网络转换设备

Patent Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101026567A (zh) * 2007-01-29 2007-08-29 华为技术有限公司 地址转发表项的保活方法及系统
US20090213792A1 (en) * 2008-01-10 2009-08-27 Samsung Electronics Co., Ltd. Apparatus and method for setting a default gateway address in a mobile communication system
CN102685261A (zh) * 2011-03-15 2012-09-19 中国移动通信集团公司 一种控制设备的地址映射状态的方法、系统及装置

Also Published As

Publication number Publication date
CN103906037A (zh) 2014-07-02

Similar Documents

Publication Publication Date Title
EP3637860B1 (en) User device
US9461919B2 (en) Network system, method, apparatus, and program
US9769852B2 (en) Maintaining current cell location information in a cellular access network
JP6241627B2 (ja) 通信システム
JP6068532B2 (ja) アクセスポイントへのユーザ機器のマルチパケットデータ接続を設定する方法及びノード
US8855045B2 (en) Method and system for controlling establishment of local IP access
WO2011050678A1 (zh) 一种基于控制面与媒体面分离的网络架构实现的通信网络
US20140334418A1 (en) Support of user plane transactions over a mobile network
WO2009043209A1 (fr) Procédé permettant d'établir une porteuse vers un terminal utilisateur en mode repos
WO2012136097A1 (zh) 分组数据网络网关及终端移动性管理的系统
CN102740270B (zh) 一种移动性管理、及为终端创建上下文和建立通道的方法
WO2015052917A1 (ja) 通信システム、通信装置および通信制御方法
WO2011015147A1 (zh) 数据传输方法、装置和通信系统
WO2007147345A1 (fr) Procédé de sélection d'entité plan utilisateur du côté réseau et d'entité plan contrôle
US11576219B2 (en) User equipment, control apparatus, and communication control method
WO2011026392A1 (zh) 一种路由策略的获取方法及系统
WO2017201903A1 (zh) 数据业务控制方法及相关设备
WO2009046666A1 (en) Addressing method of policy decision function entity, network element and network system
WO2011085618A1 (zh) 一种终端切换的方法及相应的通信网络
WO2008154874A1 (fr) Procédé et système permettant d'établir un tunnel dans le réseau en évolution
WO2010133107A1 (zh) 家用基站网关转发消息至家用基站的方法及系统
WO2012130133A1 (zh) 一种接入点及终端接入方法
WO2014101525A1 (zh) 完成网络地址转换保活的方法及设备
CN102625305B (zh) 接入演进分组系统的方法及系统
EP2617259B1 (en) A method for providing a local traffic shortcut in a packet-oriented mobile communication network

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: 13866737

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 13866737

Country of ref document: EP

Kind code of ref document: A1