KR20040066298A - Multiple connection service method on universal plug and play network - Google Patents

Abstract

PURPOSE: A method for a multiconnection service on a UPnP(Universal Plug and Play) network is provided to improve the performance of an IGD(Internet Gateway Device) by providing differential QoS(Quality of Service) with minimal configuration to various traffic in a home network, and efficiently realize QoS. CONSTITUTION: Packets of a specific class lost due to packet transfer rate are dropped to be separated and controlled. The separated packets of the specific class are classified into control packets, real-time packets, and data packets. Service orders of the classified packets are decided. A layer3 forwarding service satisfying scheduling to the scheduled packets is provided.

Description

MULTIPLE CONNECTION SERVICE METHOD ON UNIVERSAL PLUG AND PLAY NETWORK}

The present invention relates to a UPnP network system, and more particularly, to a multi-connection service method on a UPnP network in a wide area network (WAN) connection service.

The communication between two or more devices in a home is called a home network.

UPnP has been proposed as an attempt to integrate networks in the current home.

The UPnP standard establishes devices and services. UPnP devices provide services, which consist of actions, state variables, and events.

The control point (abbreviated as CP) calls a service provided by the device as an action, observes state variables, subscribes to events provided by the device, and reads device changes.

The UPnP Internet Gateway Device (IGD, hereinafter abbreviated as IGD) is located at a boundary between a LAN and a WAN constituting a home network to provide Internet access services to devices in the home network. In other words, UPnP version 1.0's IGD features automatic identification in home networks, convenient control of network address translation (NAT), and support for multiple Internet connections.

Therefore, anyone who wants to connect to the Internet using an IGD 1.0 device invokes the RequestConnection action via IGD-CP.

After that, if the connection is granted, the packet can be sent to the IGD from this time.

If connection termination occurs, receive an event from the IGD 1.0 device and report the cause.

The UPnP IGD 1.0 device may be configured with a plurality of wide area network (WAN) connection services.

However, in the conventional WAN connection service of UPnP IGD version 1.0, since QoS-related state variables are not defined, there is a problem that a client cannot use a specific connection service in consideration of QoS.

Quality of Service (QoS) refers to various technologies and concepts that manage traffic and bandwidth in a limited WAN bandwidth by differentiating service levels according to importance for users or applications.

The QoS solution does not simply increase the network bandwidth by increasing the limited bandwidth, but effectively monitors and analyzes the bandwidth and the traffic generated therein to effectively configure and manage the policy-based network and improve the network management. It is.

Accordingly, an object of the present invention is to add QoS to the basic WAN connection service currently provided by the IGD version 1.0 specification to provide a QoS service effectively in order to improve the conventional problems within the scope of the IGD 1.0 specification. Multi-connection services on UPnP networks created to add vendor dropping, packet classification, packet scheduling, and packet forwarding capabilities to devices in the IGD 1.0 specification. Provide a method.

1 is a block diagram of a multiple connection system for an embodiment of the present invention.

Figure 2 is an exemplary view showing a forwarding table structure in an embodiment of the present invention.

Figure 3 is an operation flowchart showing a forwarding process in an embodiment of the present invention.

4 is a signal flow diagram illustrating a multi-connection service process in an embodiment of the present invention.

5 is an exemplary view showing a packet classification method in FIG.

6 is an exemplary diagram illustrating a packet scheduling method in FIG.

7 is a signal flow diagram illustrating a packet forwarding process in FIG.

Explanation of symbols on the main parts of the drawings

110: QoS-IGD 121,122: QoS-CP

130: UPnP device

The present invention provides a connection service in the QoS-IGD provided with QoS-CP and QoS-IGD on the UPnP network to achieve the above object and the application requests QoS to the QoS-CP. Packet elimination step for isolating packets of a specific class lost due to the transmission rate of the packet, and packet classification for classifying the isolated class of packets into control packets, real-time (RT) packets, data packets, etc. And a packet scheduling step of determining a service order of the classified packets, and a packet forwarding step of providing a connection service capable of satisfying the scheduling of the scheduled packet.

Each of the above steps may be started by an enable action and continue until there is a disable action.

The packet removing step may be performed in a packet transmission process between a wide area network (WAN) and a local area network (LAN), that is, a packet transmission process between an IP engine and a WAN or between a LAN and an IP engine.

In the packet classification step, the packet is classified by class by an action such as class assignment, return and inquiry by tagging the packet with a class number / name.

The packet scheduling step may be configured to selectively or simultaneously use a priority scheduling method and a fair scheduling method. In other words, the packet scheduling is configured to give priority to control signals, video signals, and data, and to distribute bandwidth by weighting the data.

The packet forwarding step includes the steps of inquiring the information of the forwarding entry, checking whether there is QoS content suitable for the QoS requirement of the CP, and if the QoS content suitable for the QoS requirement of the CP is identified, the CP determines It is characterized by consisting of a process of performing a connection request (Request Connection) action to the QoS-based IGD by inputting the Connection Service Instance.

This packet forwarding step is characterized in that performed in addition to QoS-IGD to the Layer3Forwarding service.

Hereinafter, the present invention will be described in detail with reference to the drawings.

1 is a block diagram of a multi-connection service system for an embodiment of the present invention, having a forwarding table including QoS-related information as shown therein, and a vendor specification that is compliant with the UPnP IGD version 1.0 specification. QoS-IGD 110, which provides a connection service for a connection request from any QoS-CP on a UPnP network by referring to a forwarding entry that meets the QoS requirements of the QoS-CP, and UPnP. QoS that has a function capable of controlling the function of the vendor specification of the QoS-IGD 110 in a device 130 and a UPnP CP capable of controlling an IGD 1.0 device, and performs a connection request to the QoS-IGD 110. -CP (121, 122).

The UPnP specification allows UPnP manufacturers to add functional parts of vendor specifics.

Accordingly, in the embodiment of the present invention, the QoS-IGD 110 is configured by adding a vendor-specific function to a standard IGD 1.0 device utilizing multiple wide area network (WAN) connection services. do.

Also, in the embodiment of the present invention, the QoS-IGD 110 provides a QoS connection service to the QoS-CP 121 or 122, and as shown in the signal flow diagram of FIG. 4, is relatively faster than the WAN. A packet dropping step for isolating and controlling packets of a particular class lost due to the transmission speed of the LAN, and controlling packets, real-time (RT) packets, and data packets for the isolated packets. A packet classification step for classifying and managing a packet, a packet scheduling step for determining a service order of the classified packets using various techniques, and a Layer3 Forwarding service, that is, scheduling for the scheduled packet, is performed. It executes a packet forwarding step that provides a connection service that can satisfy.

Detailed operations of each of these steps will be described with reference to FIGS. 5 to 7.

First, as a first step, a packet dropping step may be performed between an IP engine and a WAN, but the embodiment of the present invention will only be described when executed in a process of transitioning from a LAN to an IP engine.

In general, packet dropping techniques include a tail drop that drops incoming packets when the queue is full, a class-based tail drop (CBTD) that applies a tail drop by setting a virtual queue for each class, and an average queue size. If the size is exceeded, there is a RED (Random Early Drop) which deletes the next packet with a probability proportional to the queue size.

In the embodiment of the present invention, any packet dropping technique may be used and implemented by adding actions to the connection service. In other words, the present invention is independent of the packet dropping technique to be used and only the method of using them.

An example of an action to use for packet deletion in an embodiment of the present invention is as follows. At this time, it is assumed that the implementation of the state variable necessary for the implementation of the action is included.

First is the TailDrop Enable / Disable action. In this case, you may also want to maintain a single queue as a whole and determine the queue size.

Second, when applying Tail Drop for each class, CBTD Enable / Disable action is required, and action for specifying queue size for each class is also required. In this case, interworking with the packet classification service that plays a role in class assignment is important.

Third, when RED is applied, actions for manipulating RED parameters such as RED Enable / Disable action, weight parameter, and probability parameter are required. RED can also be used in combination with classes, in which case it must interwork with packet classification services.

As a second step, the packet classification step is a process of configuring what kind of packets are sorted by which criteria the transport packets from the LAN to the WAN of the home network are sorted.

The actual packet classification engine tags the class number / name to the packet so that it can be used for packet dropping, packet scheduling, and packet forwarding.

Actions such as class assignment, return, and inquiry are needed to construct packet classification. That action includes AddClass, DeleteClass, GetGenericClassEntry (lookup by index), and GetSpecificClassEntry (lookup by class number).

Packet tagging refers to an operation of classifying a packet, and a method of writing a packet tag is implementation dependent. The usual way is to copy the packet into memory and add the tag field. Alternatively, the tag may be implicit, in which case the classifier inserts the packet directly into the queue.

The packet classification process will be described below with reference to the exemplary diagram of FIG. 5.

Here, there are three types of classes 1, 2, and 3, class 1 is a control packet, class 2 is a real-time packet, and class 3 is a data packet.

A class is assigned to the class array by the AddClass action. For example, AddClass, Control, and r1 indicate that the class you are assigning is Control and that class number 1 is returned as the return value.

In class 1, control packets are identified by a range of port numbers.

Class 2 is identified by the range of port numbers and packet sizes. This is because a multimedia number or a real-time communication protocol usually has a port number and a packet size depending on the codec characteristics.

Class 3 uses a class name, a source IP, a destination IP and a port number, a type of service (ToS) field, a class description, and the like.

DeleteClass and GetGenericClassEntry use the class number as an index and GetSpecificClassEntry uses the class name as an index.

Although not shown in the example of FIG. 5, packet classification criteria are also passed as an action factor.

In addition, as a third step, the packet scheduling step is to determine the transmission order of packets passing through the deletion and classification steps. When the Enable action is called, scheduling starts and continues until the Disable action is called.

Various scheduling schemes exist, and any scheme can be used like other QoS service components. An embodiment of the present invention proposes a method of implementing an action for configuration.

The scheduling scheme presents priority scheduling and pair scheduling as an example.

Priority scheduling is a method of assigning a priority to a class to send packets existing in a queue of a high priority class first.

Pair scheduling is a method of equally distributing bandwidth among classes.

The pair scheduling includes fair queuing (FQ) and weighted fair queuing (WFQ), and the FQ is one of the methods of equally distributing bandwidth, and the WFQ is a method of distributing bandwidth in proportion to each class.

By the way, although the implementation scheme of the action varies according to each type, the action is basically divided into a scheduling enable / disable action and a configuration action specific to the scheduling technique.

In addition, certain scheduling is divided into globally and internally or by grouping several classes.

6 illustrates an implementation example in which priority scheduling is performed between classes 1, 2, 3, and 5, and WFQ is performed between classes 3, 5, and 5 among the five classes.

Looking at the invocation of the action in Figure 6, PrioritySchedulingSetPriority (4, 3) means giving a priority 3 to class 4. In addition, the priority is equal to 3 between classes 3 to 5, but since WFQ is used, bandwidth is allocated in proportion thereto. WFQSchedulingSetWieght (3,0.1) means giving a weight of 0.1 to class 3.

As a final step, the packet forwarding step is a process for performing an IGD Layer3Forwarding service on scheduled packets.

First, the application communicates the QoS request of the application to the QoS-CP 121 or 122, which can satisfy the QoS request by using actions to retrieve the entries of the forwarding table. Select a connection service Accordingly, the QoS-IGD 110 provides the QoS connection service required by the QoS-CP 121 or 122 with reference to the QoS contents included in the entry of the forwarding table.

In the embodiment of the present invention, the forwarding table mounted in the QoS-IGD 110 may include a destination address, a prefix length, a connection service instance, and a QoS description as illustrated in the example diagram of FIG. and a plurality of entries each containing information such as a description, a general description, and the like.

The destination address is a value used by the QoS-IGD 110 to search for a forwarding table entry such as a destination address of a received packet. This is a value up to the prefix length. The Connection Service Instance represents an Outgoing Interface to which packets will be forwarded. The QoS description information is information representing a qualitative characteristic of the connection service when using the connection service and may be represented by a maximum speed (Mbps) value of the corresponding connection. The general description is a string to be shown to the user through a user interface (UI) of the QoS-CP 121 or 122 and is used to more easily manage the forwarding table. The ForwardingNumberOfEntries state variable is defined to represent the total number of forwarding table entries.

In addition, QoS-CP (121, 122) uses the Action (Add, Delete, GetGenericForwardingEntry, GetSpecificForwardingEntry, etc.) to insert, delete, and query the entry of the forwarding table of Figure 2, which is included in the Layer3Forwarding service.

The Add action inserts a new entry into the forwarding table by inputting a destination address, a prefix length, a connection service instance, and a general description. The Delete action deletes an entry of a corresponding forwarding table by inputting a destination address, a prefix length, and a connection service instance. The GetGenericForwardingEntry action retrieves an entry of a corresponding forwarding table by inputting an index value. The GetSpecificForwardingEntry action retrieves an entry of a corresponding forwarding table by inputting a destination address and a prefix length. The GetForwardingNumberOfEntries action retrieves the number of entries in the forwarding table.

That is, in the embodiment of the present invention, as shown in the operation flowchart of FIG. 3, in order to effectively use a multiple WAN connection service, a connection request from a QoS-based CP 121 or 122 is requested (Request Connetion). When performing the) action, the QoS-IGD 110 refers to the QoS content included in the entry of the forwarding table to provide a connection service that provides the QoS required by the QoS-CP 121 or 122.

3 is a flowchart illustrating a process in which a QoS-based CP 121 or 122 requests a connection to the QoS-IGD 110 with reference to a QoS specification (information).

That is, FIG. 3 shows a process of inquiring (confirming) the number of forwarding entries using the GetForwardingNumverOfEntries Action, and checking the individual forwarding entries using the GetGenericForwardingEntry Action when the number of forwarding entries is completed, and confirming the QoS Confirming whether the QoS specification (information) meets the requirements, and repetitively performing forwarding entries as many as the checked entries if the QoS information does not meet the QoS requirements. If it is confirmed that the QoS information meets the QoS requirements of the CP, the QoS-CP (121 or 122) is a process of performing a connection request action by inputting a connection service instance of the corresponding entry.

A packet forwarding operation for performing this process will be described with reference to the signal flow diagram of FIG. 7.

The signal flow diagram of FIG. 7 illustrates the process from the connection request to the start of the forwarding service when only the bandwidth is assumed as the QoS parameter. Here, it is assumed that the connection service of the IGD specification 1.0 and the layer 3 forwarding service are QoS extended.

The application delivers the QoS requirement bandwidth '10' to the QoS based CP 121 or 122, which retrieves the connection services and QoS characteristics of the forwarding table entry of the IGD 110.

GetGenericForwardingEntry, which exists in QoS extended Layer3Forwarding, is called for each forwarding entry to receive information including the maximum bandwidth and the available bandwidth.

In FIG. 7, the forwarding table entry indicates only QoS related items, and (20, 5) of the forwarding table entries means that the connection service 1 has a maximum bandwidth '20' and an available bandwidth '5'.

The method of selection when multiple connection services can meet QoS requirements is implementation dependent, i.e., arbitrary. At this time, it is assumed that Connection Service 2 and 4 can satisfy the bandwidth requirement of '10', but Connection Service 2 is selected.

Therefore, the QoS extended RequestConnection action is called for Connection Service 2 and requires a bandwidth of '10'. Accordingly, the forwarding table entry and the QoS characteristics of the Connection Service 2 are updated to 500 and 290 based on the execution result.

As described in detail above, the present invention can achieve the following effects.

1. Performance of IGD can be improved by providing differential QoS for various traffic in home network in minimum configuration.

2. In the home network other than UPnP, QoS service can be provided in the same way.

3. By concentrating on the composition of QoS, it is possible to efficiently provide QoS by accommodating various techniques such as packet dropping, classification, and scheduling.

Claims (10)

In the QoS-CPD and the QoS-IGD on the UPnP network, if the application requires QoS in the QoS-CP method for providing a connection service in the QoS-IGD, Packet dropping to isolate packets of a particular class that are lost due to the transmission rate of the packet, A packet classification step of classifying the packets of the specific class isolated in the control packet, a real-time packet, a data packet, and the like; A packet scheduling step of determining a service order of the classified packets; And a packet forwarding step of providing a connection service capable of satisfying the scheduling of the scheduled packet. 2. The method of claim 1, wherein each step is configured to be initiated by an Enable action and terminated by an Disable action. The method of claim 1, wherein the packet removal step A method of multi-connection service on a UPnP network, characterized in that it is performed between an IP engine and a WAN or between a LAN and an IP engine when transmitting packets between a wide area network (WAN) and a local area network (LAN). The method of claim 1, wherein the packet classification step classifies the packets by class by an action such as class assignment, return, and inquiry by tagging a packet with a class number / name. . The method of claim 1, wherein the packet scheduling step A service method for a multiple connectivity service on a UPnP network, characterized in that service order is determined for each class by priority and weighting. The method of claim 1, wherein the packet forwarding step is performed by a Layer3Forwarding service provided in QoS-IGD. The method of claim 1, wherein the packet forwarding step Querying the information of the forwarding table entry; Verifying that there is QoS content that meets the QoS requirements of the CP; If the QoS content suitable for the QoS requirements of the CP is confirmed in the above, the CP performs a request connection action to the QoS-based IGD using the connection service instance of the corresponding entry as an input. A multi-connection service method on a UPnP network. 8. The method of claim 7, further comprising updating a forwarding table entry according to a connection service provision result. 8. The method of claim 7, further comprising inserting or deleting an entry in a forwarding table. 10. The method according to any one of claims 7 to 9, wherein the forwarding table includes a connection service instance representing a destination address of a received packet, a prefix length, an outgoig interface to which the packet is forwarded, and a quality of the connection service. And a plurality of entries each of which includes a QoS description indicating a description, and a general description indicating a character string to be displayed in a user interface.