WO2013064053A1 - Method and system for policy management in stream movement - Google Patents

Abstract

A method for policy management in stream movement comprises: when a terminal initiating stream migrating or switching, the terminal sending a modified stream selecting policy to a network side; and the network side sending the modified stream selecting policy to a policy server; or, when a stream selecting policy based on the terminal exists on the policy server, the network side sending the modified stream selecting policy to the policy server. An embodiment of the present invention further provides a method for policy management in stream movement.

Description

 Strategy management method and system in flow movement

Technical field

 The present invention relates to the field of mobile communications, and in particular to a method and system for managing flow mobility in a multiple access system.

Background technique

 1 shows a location server based communication network architecture. The terminal may be one or more of a mobile terminal, a fixed terminal, and a nomadic terminal, such as a mobile phone, a fixed telephone, a computer, a server, and the like. The access network is used to provide a layer 2 (physical layer and link layer) access means for the terminal, and maintains a physical access link between the terminal and the access gateway. The access gateway is configured to allocate location information to the terminal, maintain the identity and location mapping information of the terminal, perform location registration (also referred to as location registration) to the location server, and query location information of the terminal, and implement routing and forwarding of data packets. . The location information assigned by the access gateway points to the access gateway, that is, the address information of the access gateway, and the location information is used as the destination address of the data packet, and the data packet is routed to the access gateway. The location server is configured to process registration, logout, and query of terminal location information, and save and maintain identity information and location information mapping information of the home user terminal.

 When the terminal communicates with the peer end, the access gateway receives the uplink data packet to encapsulate and forward the location information. Specifically, the location information of the communication peer is locally queried, for example, the location information of the communication peer is used as the destination address, and the location information of the terminal is used as the source address in the data packet including the terminal identifier and the communication peer identifier. The generalized forwarding plane forwards to the access gateway of the communication peer access. If the query is not available, the location information of the communication peer is queried to the home location server of the communication peer and stored locally. In this case, the location information of the terminal may be encapsulated in the packet, and then forwarded to the generalized forwarding plane through the mapping forwarding plane, or after the location information of the communication peer end is queried, the access gateway may query the local query as described above. Location information encapsulation and forwarding processing is performed in the manner of communicating the location information of the opposite end. In the downlink direction, after receiving the data packet sent by the generalized forwarding plane, the access gateway encapsulates the solution location information, strips the location information, and sends the location information to the terminal.

It can be seen that, in order to implement normal forwarding of packets, the access gateway needs to allocate location information to the terminal when the terminal accesses, and register the location information to the location server to update the terminal server. Location information. The access gateway also needs to maintain the identity and location information mapping information of the terminal and its communication peer to implement the encapsulation of the location information of the packet. In an example, the access gateway maintains the communication relationship between the terminal and the communication peer for each terminal, which is referred to herein as the peer information of the terminal, and the peer information includes the correspondence between the terminal identifier and the communication peer identifier. The information may also include the identifier and location information mapping information of the terminal. At the same time, the access gateway uniformly maintains the identity and location information mapping information of the communication peers of all terminals. Of course, it is also possible for the access gateway to separately maintain the identity and location information mapping information of all communication peers of the terminal, including the peer information of the terminal. The peer information is maintained. When the terminal switches, the access gateway on the side of the terminal can determine which communication peers the terminal has, so that the terminal can access the new location information of the terminal to the access gateway accessed by the communication peer. Location update. After the access gateway accessing the communication peer is updated, the data packet sent by the communication peer can be directly routed to the access gateway to which the terminal switches.

 Since the terminal can have multiple connections at the same time (as shown in FIG. 2, FIG. 3 and FIG. 4), there may be multiple access gateways that allocate multiple location information for the terminal. For the same terminal, there may be one in the location server. When the terminal identifier corresponds to multiple location information, the connection considers the characteristics of the application or service flow in the process of establishing, and has related attributes such as service quality and charging, and only allows the corresponding application or service flow to be transmitted. In order to determine the correct route of the data flow (ie, the corresponding access gateway) when the peer end sends data to the terminal, the policy server stores policy information related to the application, or service, or connection, or location. The information may be related to the terminal identifier (ie, the terminal-based policy), or may be independent of the terminal identifier, and provide the location of the terminal according to the application, or service, or connection, or location feature, or terminal identifier when the peer sends data to the terminal. Reference, route with the location server and find the correct location information for the corresponding data stream. The Policy Server can be connected to the Location Server and/or Access Gateway as shown in Figure 2, Figure 3 and Figure 4 in the communication network.

 The data transmission path has been modified due to load sharing, signal strength, user wishes, network planning, operation and maintenance, etc., causing users or networks to migrate or switch certain data flows between access gateways. At this time, the peer access gateway still routes the data to the access gateway before migration or handover, so that the data transmission path cannot be connected or the data transmission quality cannot be guaranteed.

For the PMIP scenario, as shown in Figure 5a is the logical architecture of the PMIP protocol, including the Mobile Node (MN), the Correspondent Node (CN), and the mobile. Access Gateway (MAG) and Local Mobility Anchor (LMA). The CN can be a fixed node or a mobile node, that is, it has a corresponding MAG and LMA. The MAG is the first hop router of the MN, and its main functions include assigning a Care of Address (CoA) to the MN when accessing it, and performing PMIP Binding (MMIP Binding) with the MN's anchor LMA instead of the MN. . As the anchor of the MN, the LMA's main functions include assigning the Home of Address (HoA) and handling the above PMIP binding. The main purpose of the PMIP binding performed between the MAG and the LMA is to let both parties know the address of the other party, the above CoA and HoA, and save it locally. In addition, during the PMIP binding process, a bidirectional tunnel is established between the MAG and the LMA for the MN. It is worth noting that the IP address finally obtained by the MN is the HoA assigned by the LMA. In a typical network deployment, the MAG is generally located at a lower topology, such as at the edge of a metropolitan area network; and the LMA is generally located at a higher topology, such as the core of the provincial backbone. The MAG and the LMA are often connected by a multi-hop router.

 The mobility management of the PMIP protocol is reflected in the fact that the MN's mobility can change the currently connected MAG while keeping the MN's IP address (ie HoA) unchanged. As shown in Figure 5b, changing the MAG of the current connection means changing/switching the connection to the target MAG ( tMAG, target MAG ) from the previously connected source MAG ( sMAG, source MAG ). After changing to tMAG, tMAG assigns a new CoA to the MN, and performs PMIP binding with the MN's anchor LMA for the MN, updating the information saved by the two parties mentioned above, and the t-MAG and LMA. A new two-way tunnel is established for the MN.

 Figure 5b shows the process of sending and receiving IP 4 messages between MN and CN. As shown in Figure 5b, the IP data packet between the MN and the CN must pass through the tunnel between the sMAG and the LMA before the switchover. After the switchover, the tunnel must pass through the tunnel between the tMAG and the LMA.

In order to solve the problem of wasteful transmission path existing in the existing PMIP mechanism, a series of adverse consequences are required, and the existing PMIP mechanism needs to be modified. Figure 5c shows the modified PMIP protocol architecture. Compared with the logical architecture of the existing PMIP protocol (as shown in Figure 5a), the modified PMIP protocol architecture contains the network elements of the mobile node MN, the communication peer CN, the mobile access gateway MAG, and the mapping server. Server, MAPS). The modified PMIP protocol architecture no longer has LMA network elements. The CN can be a fixed node or a mobile node, that is, There are corresponding MAG and MAPS.

 The MAG is the first hop router of the MN. Its main function is to allocate the care-of address CoA for the MN in the existing PMIP architecture, and to perform PMIP binding with the MAP of the MN instead of the MN. The MAPS queries to obtain a bidirectional tunnel between the current CN of the communication peer CN and the MAG of the CN, and forward the IP data packet between the MN and the CN.

 MAPS retains the LMA function to handle MN registration, logout and update functions, assigns HNP (Home Network Prefix) function, and establishes and maintains Binding Cache Entry (BCE) function, but not as MN. The anchor, the IP data packet between the MN and the CN, does not need to go through the MAPS. The MAPS needs to save the current MAG-MN address and/or the MN's CoA, and use the MN as an index, for example, the MN's HoA or the MN's HNP. As an index, the MAG of the communication partner CN or CN of the MN is queried according to the relevant information of the MN.

Figure 5d shows the process of transmitting and receiving IP 4 messages between the UI and the CN when applying the modified PMIP architecture. The IP data packet between the MN and the CN needs to pass through the tunnel between the MAG-MN and the MAG-CN. When the MN sends an uplink IP packet to the CN, it is consistent with the existing PMIP mechanism, and the MN needs to send the IP packet to the MAG-MN. Subsequent to the existing PMIP mechanism, the MAG-MN needs to query the address of the MAG-CN. After querying the address of the MAG-CN (such as the IP address of the MAG-CN), the MAG-MN uses the address of the MAG-CN as the destination to establish a tunnel from the MAG-MN to the MAG-CN (for example, an IP in IP tunnel). At the same time, the above IP packet is placed in the tunnel and sent directly to the MAG-CN. After receiving the IP packet from the tunnel, the MAG-CN sends an IP packet to the CN. It should be noted that the MAG-MN can also query the CoA of the CN, and replace the address of the MAG-CN with the CoA address of the CN, and can achieve the same purpose. At this time, the MAG-MN uses the CoA as the end point and establishes the tunnel of the MAG-MN to the MAG-CN, and the effect is equivalent. It is also worth noting that when the MAG-MN queries the address of the MAG-CN (or the CoA of the CN), it first queries the local cache, and if it does not query, it can query other network elements. For example, the anchor point LMA (LMA-CN) of the CN can be queried according to the HoA of the CN. After querying the desired result, MAG-MN caches the query results locally. The advantage of caching the query results locally is that it can avoid frequent logins to other network elements. By the same token, CN to MN When sending a downlink IP packet, the same method as above is used. The principle is the same and will not be described again. With this method, the path between the MN and the CN to transmit and receive IP 4 变为 变为 becomes the MAG<->CN of the MAG<->CN of the MN<->MN, without the anchor point LMA network of the MN (or MN and CN) yuan.

 Similarly, in the case that the PMIP terminal has multiple connections for stream migration or handover at the same time, the same terminal performs data flow location information change in multiple access gateways, and the peer cannot obtain the correct access gateway. The problem.

Summary of the invention

 The present invention provides a method and a system for managing a policy in a flow movement, so as to solve the problem that the remote terminal cannot learn the correct location information of the terminal and the packet transmission is abnormal when there are multiple access gateways.

 In order to solve the above problem, the present invention provides a policy management method in a flow movement. When a terminal initiates a flow migration or handover, the terminal sends the modified flow selection policy to the network side.

 And the network side sends the modified flow selection policy to the policy server; or, when the flow server selects a flow selection policy based on the terminal, the network side uses the modified flow selection policy Sent to the policy server.

 The invention also provides a policy management system in flow mobility, comprising a network side and a policy server, wherein:

 The network side is configured to: when the terminal initiates a flow migration or handover, receive the modified flow selection policy sent by the terminal; and send the modified flow selection policy to the policy server; or When the flow selection policy based on the terminal exists on the policy server, the modified flow selection policy is sent to the policy server.

 The embodiment of the invention provides a method and a system for managing a policy in a flow movement. When there are multiple access gateways, the peer end learns the correct location information of the terminal to implement message transmission.

BRIEF abstract

The accompanying drawings are intended to provide a further understanding of the invention, Improperly qualified. In the drawing:

 1 is a schematic diagram of a communication network connection in the prior art;

 2 is a schematic diagram 1 of a multi-connection of a communication network in the prior art;

 3 is a schematic diagram 2 of a multi-connection of a communication network in the prior art;

 4 is a schematic diagram 3 of a multi-connection of a communication network in the prior art;

 Figure 5a is a logical architecture of a PMIP protocol in the prior art;

 FIG. 5b is a schematic diagram of a process of transmitting and receiving IP packets between a MN and a CN in the prior art; FIG. 5c is a schematic diagram of a modified PMIP protocol in the prior art;

 FIG. 5d is a schematic diagram of a process of transmitting and receiving IP documents between a MN and a CN when applying the modified PMIP architecture in the prior art;

 6 is a flowchart of Embodiment 1 of a method for registering a terminal of the present invention;

 7 is a flowchart of Embodiment 2 of a method for registering a terminal of the present invention;

 8 is a flowchart of Embodiment 3 of a stream migration method according to the present invention;

 9 is a flow chart of Embodiment 4 of the flow migration method of the present invention;

 10 is a flowchart of Embodiment 5 of the flow migration method of the present invention;

 11 is a flowchart of Embodiment 6 of the stream migration method of the present invention;

 12 is a flowchart of Embodiment 7 of the stream migration method of the present invention;

 FIG. 13 is a flow chart in which the policy server initiates policy modification, and the MAPS-MN notifies the update peer;

 FIG. 14 is a flow chart of the MAPS-MN initiating policy modification, and the MAPS-MN notifying the update peer;

 Figure 15 is a flow chart in which the MAG1-MN initiates policy modification, and the MAPS-MN notifies the update peer;

FIG. 16 is a flowchart of the MAG1-MN notifying the MAPS-MN to initiate policy modification, and the MAPS-MN notifying the update peer. Preferred embodiment of the invention

 In order to more clearly describe the content of the present invention, the following provisions are made in the embodiment of the present invention:

 The CN is a communication peer node of the MN, and the CN may have multiple.

 MAG1-MN refers to the MAG that the MN accesses through interface 1, and the proxy care-of address assigned by the MAG to the MN is Co Al.

 The MAG2-MN refers to the MAG that the MN accesses through the interface 2, and the proxy care-of address assigned by the MAG to the MN is CoA2. MAGI-MN and MAG2-MN may be located in different access networks or in the same access network.

 MAG-CN1 refers to the MAG currently accessed by CN1, and the proxy transfer address assigned by CNG to CN1 is CoA3.

 MAG-CN2 refers to the MAG currently accessed by CN2, and the proxy transfer address assigned by CNG to CN2 is CoA4.

 The MAPS-MN refers to the MAPS accessed by the MN. The home network prefix assigned by the MAPS to the MN is HNP1, and the home address obtained by the MN according to the HNP1 configuration is HoAl.

 The MAPS-CN refers to the MAPS accessed by the CN. The home network prefix assigned by the MAPS to the CN is HNP3, and the home address obtained by the CN according to the HNP3 configuration is HoA3.

 In the embodiment of the present invention, the meaning of multiple access means that the MN may have different connections at the same time, and the connections may be established at different access networks at the same time, or may be established at the same time in the same access network. Different connections.

 In the embodiment of the present invention,

The flow selection policy may be one or more of the following: access type, priority, network name, PLMN (Public Land Mobile Network) ID, QoS parameters, charging information, and communication peer information. The network name may be an APN (Access Point Name), a domain name/identity, a PDN (Packet Data Network) name/identity. The QoS parameter may be a QCI (QoS Class Identifier). The billing information can be a rate or a tariff level. The communication peer information can be the communication peer identifier, port, and address. The flow selection policy may refer to the information of the terminal network or the peer network. If it is the information of the terminal network, it has a corresponding relationship with the peer network. The location server can be MAPS, ILR (Identity Location Register), LMA, HA (Home Agent, Home Agent). The location information may be an access gateway address/identity, or an IP address assigned by the access gateway, or an RID, or a location identifier, or a CoA.

 The identity of the terminal can be AID, IMSI, NAI (Network Access Identifier), HNP, ID, IP address.

 The embodiment of the invention provides a policy management method in flow movement, which includes:

 When the terminal initiates a flow migration or handover, the terminal sends the modified flow selection policy to the network side;

 And the network side sends the modified flow selection policy to the policy server; or, when the flow server selects a flow selection policy based on the terminal, the network side uses the modified flow selection policy Sent to the policy server.

 After the policy server receives the modified flow selection policy, when the flow server selects a flow selection policy based on the terminal, the policy server updates and updates according to the modified flow selection policy. a flow selection policy related to the terminal; when the flow selection policy based on the terminal does not exist on the policy server, the policy server uses the modified flow selection policy as a new flow selection policy of the terminal .

 There are two ways to trigger the policy server:

 method one:

 The terminal sends a flow migration indication message to the source access gateway or the target access gateway of the migrating flow of the terminal, where the flow migration indication message carries the modified flow selection policy;

 Transmitting, by the source access gateway or the target access gateway, the flow migration indication message to a location server of the terminal;

 The location server sends a policy modification request message to the policy server, and carries the modified flow selection policy. A migration stream is a stream in which a stream migration or handover occurs.

 Method 2:

The terminal sends a flow migration indication message to the source access gateway or the target access gateway of the migrating flow of the terminal, where the flow migration indication message carries the modified flow selection policy; The source access gateway or the target access gateway sends a policy modification request message to the policy server, and carries the modified flow selection policy.

 The location server may be further updated. In the second manner, the source access gateway or the target access gateway sends a flow migration indication message to the location server of the terminal, where the flow migration indication message carries the modified Stream selection strategy.

 In the first and second modes, after receiving the modified flow selection policy, the location server updates the local flow selection policy related to the terminal according to the modified flow selection policy.

 When the terminal registers with the location server, the location server obtains the connection information or the flow selection policy from the policy server, and establishes a registration information table of the terminal, where the registration information table of the terminal includes the identity identifier of the terminal, Mapping relationship between interface information and connection attributes currently accessed by the terminal;

 Returning, by the location server, the identity information that is allocated by the location server to the terminal, the access gateway, the identity information of the terminal, or the identity information of the terminal The location information allocated by the access gateway to the terminal is sent to the terminal.

 After receiving the modified flow selection policy, the location server modifies the registration information table of the terminal, and modifies the mapping relationship between the terminal identifier, the source interface information and the connection attribute of the migration flow to the terminal identifier and the migration flow. The mapping relationship between the target interface information and the connection properties.

 An embodiment of the present invention further provides a policy management system in a flow mobility, including a network side and a policy server, where:

 The network side is configured to: when the terminal initiates a flow migration or handover, receive a modified flow selection policy sent by the terminal; and

 And sending the modified flow selection policy to the policy server; or sending the modified flow selection policy to the policy server when the flow selection policy based on the terminal exists on the policy server.

The policy server is further configured to: after receiving the modified flow selection policy, when there is a flow selection policy based on the terminal, the modified flow selection policy is updated according to the modified a terminal-related flow selection policy; when there is no base on the policy server When the flow selection policy of the terminal is used, the modified flow selection policy is used as a new flow selection policy of the terminal.

 The network side includes a source access gateway or a target access gateway, and the location server: the source access gateway or the target access gateway is configured to: receive a flow migration indication message sent by the terminal, and send the flow migration indication Transmitting, by the message, the modified flow selection policy, and forwarding the flow migration indication message to a location server of the terminal;

 The location server is configured to: send a policy modification request message to the policy server, and carry the modified flow selection policy. Further, the location server is further configured to: after receiving the modified flow selection policy, update the local flow selection policy related to the terminal according to the modified flow selection policy.

 Or the network side includes a source access gateway or a target access gateway, where:

 The source access gateway or the target access gateway is configured to: receive a flow migration indication message sent by the terminal, where the flow migration indication message carries the modified flow selection policy; and send a policy modification request message to the policy server Carrying the modified stream selection policy.

 Further, the system further includes a location server, where:

 The source access gateway or the target access gateway is further configured to: send a flow migration indication message to a location server of the terminal, where the flow migration indication message carries the modified flow selection policy; And updating a local flow selection policy related to the terminal according to the modified flow selection policy.

 The location server is further configured to: when the terminal registers by using the access gateway, obtain a connection information or a flow selection policy from the policy server, and establish a registration information table of the terminal, where the registration information table of the terminal includes a mapping relationship between the terminal identity of the terminal, the interface information currently accessed by the terminal, and the connection attribute; and returning, to the access gateway, identity information that is allocated by the location server to the terminal, the access gateway Sending the identity information of the terminal, or sending the identity information of the terminal and the location information allocated by the access gateway to the terminal to the terminal.

The location server is further configured to: after receiving the modified flow selection policy, modify the registration information table of the terminal, and set the terminal identity, the source interface information, and the connection attribute of the migration flow. The mapping relationship is modified to the terminal identity, the target interface information of the migration flow, and the mapping relationship of the connection attributes.

 The following embodiment uses the location server as the MAPS as an example to describe the policy management method in the flow mobility, but all the embodiments are not limited to the MAPS, and are applicable to the scenario where the location server is another network element (ILR, LMA, HA).

 The technical solutions of the present invention are further described in detail below with reference to the accompanying drawings and specific embodiments to enable those skilled in the art to understand the invention. . Embodiment 1

 FIG. 6 is a first embodiment of the terminal registration of the present invention. The present embodiment is applicable to the MAPS-MN interacting with the policy server to obtain the connection attribute when the MN is registered. Specifically, the following steps are included:

 Step 601: After the MN accesses the PMIPv6 domain through the attach process, the MN sends a route request (RS, Router Solicitation) message to the MAG1-MN through the interface 1. This step is optional.

 Step 602: The MAG1-MN is triggered by the step 601, or the mobile network is triggered during the access process, and the MAG1-MN sends a Proxy Binding Update (PBU) message to the MAPS-MN, instead of registering with the MASS-MN. The message carries the MN identity (MN ID), the proxy broadcast address CoAl assigned by the MAG1-MN to the MN, and interface information. The address of the MAG1-MN is also sent to the MAPS-MN in this message.

 The interface information is information that can uniquely identify the interface used by the connection, and includes at least one of the following information: linklocal (local link) address of the terminal interface, MAC address of the MAG, CoA, interface identifier of the terminal, and MAC address of the terminal interface. address. The interface information in the present invention has this meaning. In this embodiment, the interface information is at least one of the following information: a link local address of the terminal interface 1, a MAC address of the MAG1-MN, a CoAl, an interface 1 identifier of the terminal, and a MAC address of the terminal interface 1.

 Step 603: The MAPS-MN receives the PBU message, allocates a home network prefix (HNP1) for the MN, and establishes/updates the BCE. The MAPS-MN sends a connection attribute query request message to the policy server, where the message carries the MN ID and the connection information, and optionally carries HNP1.

The connection information refers to specific information that can identify the connection being established, such as: CoA (this In the embodiment, it is a CoAl), a MAG address (in this embodiment, a MAG1-MN address), QoS information, charging information, priority, access network type, network name, PLMN ID, and communication peer information. The network name can be APN, domain name/identity, PDN name/identity. The QoS parameter can be QCI. The billing information may be a rate or a tariff level. The communication peer information can be the communication peer identifier, port, and address. The connection information in the embodiment of the present invention has this meaning.

 The purpose of the connection attribute query request message is to query the policy server for the attribute of the connection, which can be implemented by adding a new message, or by tampering with the existing message. The name of the message can be other names, as long as the message The above functions can be realized.

 Step 604: After receiving the connection attribute query request message, the policy server determines the connection attribute according to the connection information carried in the message and the policy information stored in the message. If the parameter is carried in step 603, the policy server saves the port and may Query the ΜΝ-based policy with ΗΝΡ as the index.

 The connection attribute refers to the category information that identifies a certain type of connection. Such as: QoS level, high/low rate, QoS parameters, charging information, access network name, priority, access network type, network name, PLMN ID. The network name can be APN, Domain Name/Identity, PDN Name/Identity. The QoS parameter can be QCI. The billing information can be a rate or a tariff level. The connection attributes in the embodiments of the present invention all have this meaning.

 For example, the connection information used in the MAPS-MN query is CoAl, and the policy on the policy server is the connection using the CoA in the range 1 with a QoS level of level 2. The policy server finds that CoAl is in range 1, so it determines that the connection property of the connection is level 2.

 Step 605, the policy server returns the connection attribute to the connection attribute query response message.

MAPS-匪.

 The connection attribute query response message is a connection attribute query request message in response to step 603, which may be implemented by adding a new message, or may be implemented by tampering with an existing message, and the name of the message may be another name, as long as the message is implemented. The above functions can be used.

 Step 606: After receiving the connection attribute query response message, the MAPS-MN saves the connection attribute, and establishes a registration information table, where the content of the table includes: MN ID-HNP-interface information-connection attribute.

Step 607: The MAPS-MN returns a proxy binding confirmation to the MAG1-MN (PBA, Proxy) Binding Ack) message. The message carries the HNP1 assigned by the MAPS-MN to the MN.

 Step 608: The MAG1-MN returns a RA (Router Advertisement Advertisement) message to the MN. After receiving the router advertisement message, the MN performs address configuration to obtain the home address HoAl originating from the home network prefix HNP1. If the interface information is CoA in step 602, the CoAl assigned by the MAG1-MN to the MN needs to be carried in the RA message.

 Embodiment 2

 FIG. 7 is a second embodiment of the terminal registration of the present invention. The present embodiment is applicable to the MAPS-MN interacting with the policy server to obtain a flow selection policy, and determining the connection attribute according to the flow selection policy. Specifically, it includes the following steps:

 Steps 701 and 702 are the same as steps 601 and 602.

 Step 703: The MAPS-MN receives the PBU message, allocates a home network prefix (HNP1) for the MN, and establishes/updates the BCE.

 When there is no user's flow selection policy in the MAPS-MN, the policy request message is sent to the policy server, and the message carries the MN ID, optionally carrying HNP1. When there is a user's flow selection policy in the MAPS-MN, there is no need to send a policy request message to the policy server.

 The purpose of the policy request message is to request the user's flow selection policy from the policy server, which may be implemented by adding a new message, or by modifying an existing message. The name of the message may be another name, as long as the message implements the above function. can.

 Step 704: After receiving the policy request message, the policy server carries the user's flow selection policy in the policy response message, and returns the message to the MAPS-MN. If the HNP is carried in step 703, the policy server saves the HNP, and can query the MN-based policy by using the HNP as an index.

 The policy response message is a response message to the policy request message in step 703, and may be implemented by adding a new message, or may be implemented by modifying an existing message, and the name of the message may be another name, as long as the message implements the above function. .

 Step 705: The MAPS-MN determines the connection attribute according to the obtained flow selection policy and the connection information.

Step 706: The MAPS-MN saves the connection attribute, and establishes a registration information table of the MN. The content of the table includes: MN ID-HNP-interface information-connection attribute. Steps 707, 708 are the same as steps 607 and 608.

 After the first or second embodiment is used, when the MN is registered by the MAG1-MN, the connection already exists on the MAPS-MN. When the peer is actively sending packets, the peer MAG queries the location information of the terminal. The MAPS-MN can directly return the MAG address of the MN's CoA or MN to the peer MAG according to the locally stored registration information table, and no longer needs to query the policy server every time.

 If the first or second embodiment is not used, the content of the MN registration information table in the MAPS-MN is: MN ID-HNP-interface information, and there is no connection attribute. When the MAPS-MN/MAG1-MN informs the communication peer that the MAG-CN is updated and needs to carry the connection attribute, the MAPS-MN/MAG1-MN can obtain the connection attribute from the modified flow selection policy sent by the terminal.

 Embodiment 3

 FIG. 8 is a third embodiment of the flow migration according to the present invention. The embodiment is applicable to the MN indicating that the MAG1-MN performs flow migration, and the MAPS-MN notifies the update peer. Specifically, the following steps are included:

 Step 801: The multi-access terminal MN has accessed the PMIPv6 domain, and uses the interface 1 to complete the registration in the MAPS-MN through the MAG1-MN, and uses the interface 2 to complete the registration in the MAPS-MN through the MAG2-MN.

 Step 802, the MN performs data communication with CN1 on interface 1. The data message of MN and CN1 is MN MAGl-MN MAG-CN1 CN2 as shown in this step.

 Step 803: The MN performs data communication with CN2 on the interface 2. The data message of MN and CN2 is MN MAG2-MN MAG-CN2 CN2 as shown in this step.

 Step 804: The MN considers to migrate the MAG1-MN stream to the MAG2-MN (that is, the stream on the interface 1 is migrated to the interface 2) based on its own situation (such as the signal strength, the rate, and the like).

The MN initiates a flow migration, which means that the MN/user selects an interface/routing strategy for the service flow (hereinafter referred to as a flow selection policy in the present invention), and the MN re-forms the flow selection policy and saves or updates the locally. Stream selection strategy.

Step 805: The MN sends a flow migration indication message to the MAG1-MN to instruct the MAG1-MN to perform flow migration. The message needs to carry the target interface information and the modified flow selection policy. It also optionally carries source interface information and stream information that needs to be migrated. The target interface information refers to related information of the target interface to which the flow needs to be migrated, and includes at least one of the following: a link local address of the terminal interface 2, a MAC address of the MAG2, a CoA2, an interface identifier of the terminal interface 2, and a terminal interface 2 MAC address.

 The modified flow selection policy refers to the flow selection policy formulated by the MN in step 804.

 The source interface information refers to the information about the source interface that needs to be migrated out, including at least one of the following: linklocal address of terminal interface 1, MAC address of MAG1, CoAl, interface identifier of terminal interface 1, and MAC address of terminal interface 1. . When the flow migration indication message is sent to the MAG1-MN in unicast mode, the source interface information may not be carried. Otherwise, the field must be carried in the message.

 The flow information to be migrated refers to the information about the flow that needs to be migrated, and is used to determine a specific flow, which may be a five-tuple or other indication information. You need to carry this information when you need to migrate individual flows on the interface. If all the flows on the interface are migrated, you do not need to carry this field.

 The purpose of the flow migration indication message is to notify the MAG to perform flow migration, and to carry related interface information and flow information, which may be implemented by adding a new message, or by tampering with an existing message. The name of the message may be Other names, as long as the message implements the above functions.

 Step 806: The MAG1-MN forwards the received flow migration indication message to the MAPS-MN. The message needs to carry the MN identity (MN ID), the target interface information, and the modified flow selection policy, and optionally carry the source interface information and the flow information to be migrated. At the same time, it is optional to carry the communication peer table of the MN (that is, the HoA or HNP or ID of all peers communicating with the MN). When the communication peer table of the MN is maintained by the MAPS-MN, the information may not be carried, otherwise the field must be carried.

 Step 807, an optional step. The MN initiates a flow migration, which means that the flow selection policy has changed. When there is a flow selection policy based on the MN (that is, related to the MN) on the MAPS-MN, the policy needs to be updated, and this step is performed. The MAPS-MN updates the local terminal-related flow selection policy according to the modified flow selection policy carried in the message.

Steps 808, 809, the MAPS-MN notifies the policy server to modify the flow selection policy. The MAPS-MN sends a policy modification request message to the policy server, where the message carries the MN ID and the modified flow selection policy. When there is a flow selection policy based on the MN (that is, related to the MN) on the policy server, the policy server updates the flow selection policy related to the MN by using the modified flow selection policy carried in the message, and there is no MN-based on the policy server. Policy server uses messages when streaming a selection policy The modified flow selection policy carried in the new MN adds a flow selection policy. The Policy Server returns a policy tamper response message.

 Alternatively, the MAPS-MN needs to perform this step only when there is a MN-based flow selection policy on the policy server, that is, the policy modification request message is sent.

 The purpose of the policy modification request message is to notify the policy server to update the flow selection policy related to the MN, which may be implemented by adding a new message, or by modifying an existing message, and the name of the message may be another name, as long as the message is implemented. The above functions are all.

 The purpose of the policy modification response message is to respond to the policy modification request message, which can be implemented by adding a new message, or by modifying an existing message. The name of the message can be other names, as long as the message implements the above functions. .

 Step 810, an optional step. When the MN registers with the MAPS-MN, the MAPS-MN interacts with the policy server to obtain a flow selection policy or obtains a connection attribute, and then performs this step. To prevent the CN from actively transmitting packets, select the interface that has been migrated (that is, on interface 1), MAPS-MN needs to update the MN registration information table, and change from HNP1-source interface information-connection attribute 1 to HNP1-target. Interface information - connection attribute 1.

 This step can be performed after step 806, without waiting for step 809 to complete before executing.

 Step 811: The MAPS-MN notifies the MAG of the opposite end to update the address mapping and the connection attribute. The address mapping refers to the mapping between the MN's HNP and the MAG-MN address, or the MN's HNP and CoA.

The MAPS-MN can obtain all the communication peer CNs of the MN in the communication peer table maintained locally or from the communication peer table obtained by the MAGI-MN, and query the corresponding MAG-CN address or the CN of the CN according to the CN, and then Sending a CoA update message to the MAG-CN (MAG-CN1 in this embodiment), where the message carries the HNP of the MN, the source MAG address (ie, the MAGI-MN address) or the source CoA (ie, CoAl), and the target MAG address (ie, The MAG2-MN address) or the target CoA (ie, CoA2) also carries the connection attribute of the migrated stream. The communication peer table obtains all the communication peer CNs of the MN, and obtains the MAPS-CN corresponding to the CN, and then sends a CoA update message to the MAPS-CN, where the message carries the HNP of the MN, and the source MAG The address (that is, the MAG1-MN address) or the source CoA (ie CoAl), the target MAG address (ie, the MAG2-MN address) or the target CoA (ie, CoA2), and also carries the connection attribute of the migrated stream. The CoA update message is forwarded by the MAPS-CN to the MAG of the CN.

 The purpose of the CoA update message is to send the address mapping of HNP1 and MAG2-MN or the HNP1 and CoA2 mapping and connection attributes to the MAG-CN. The name of the message can be another name as long as the message carries the above information.

 If the MAPS-MN does not obtain a flow selection policy or a connection attribute from the policy server when the MN registers with the MAPS-MN, the content of the MN registration information table in the MAPS-MN is: MNID-HNP-interface information, and no connection attribute exists. . Then, the MAPS-MN can obtain the connection attribute from the modified stream selection policy sent by the terminal.

 Step 812: After receiving the message, the MAG-CN1 obtains the MAG2-MN address or the CoA2 and the connection attribute in the message, updates the locally stored MN address mapping table, and sets the HNP1-CoAl (or MAG1-MN address)-connection attribute.爹 Change to the HNPl-CoA2 (or MAG2-MN address) country connection attribute. MAG-CN1 returns a CoA update response message.

 The purpose of the CoA update response message is to respond to the CoA update message of step 812, and the name of the message may be other names as long as the message completes the above functions.

 Step 813: The MAPS-MN returns a flow migration response message to the MAG1-MN.

 The purpose of the flow migration response message is to respond to the flow migration indication message of step 806. The name of the message may be other names as long as the message completes the above functions.

 This step can be performed after completing 810, without waiting for step 812 to complete.

 Step 814: The MAG1-MN returns a flow migration response message to the MN.

 The purpose of the flow migration response message is to respond to the flow migration indication message of step 805. The name of the message may be other names as long as the message completes the above functions.

 Step 815, after the stream migration is completed, the MN performs data communication with CN1 on the interface 2. The data message of MN and CN1 is as shown in this step as MNoMAG2-MNoMAG-CNl oCNl.

 In this embodiment, the policy modification request may also be sent by the MAGI-MN to the policy server. The sending mode is similar to the MAPS-MN. It can be divided into two modes, which are not mentioned here.

This example uses the flow migration initiated by the terminal as an example to describe the flow selection policy caused by the flow migration. The update process of the address mapping relationship. When a multi-access terminal is switched, that is, when the multi-access terminal changes the MAG of an interface due to the movement, or when an interface of the multi-access terminal fails for some reason (such as a failure, etc.) When serving the terminal, all the flows on the interface where the switching/failure occurs need to be switched to other interfaces of the terminal, and the flow of this embodiment can still be used. For the handover scenario, the flow migration indication message in step 805 can be changed to a more handover request message.

 Embodiment 4

 FIG. 9 is a flow migration embodiment 4 of the present invention. The embodiment is applicable to the MN indicating that the MAG1-MN performs flow migration, and the MAG1-MN notifies the update peer. Specifically, the following steps are included:

 Steps 901-905, the same steps 801-805.

 Step 906, the same as step 806, except that the communication pair indication table does not need to carry the communication peer table of the MN.

 Steps 907-910, the same steps as 807-810.

 Step 911: The MAPS-MN returns a flow migration response message to the MAG1-MN. Optional carrying in the message: MAG2-MN is the CoA2 or MAG2-MN address assigned by the MN, and the connection attribute of the migrated stream.

 When the interface information in step 905 is a CoA or MAG address, the address of the CoA2 or MAG2-MN does not need to be carried. Otherwise, the CoA2 or MAG2-MN address needs to be sent to the MAPS-MN.

 If the MAPS-MN does not obtain the flow selection policy or the connection attribute from the policy server when the MN registers with the MAPS-MN, the content of the MN registration information table in the MAPS-MN is: MNID-HNP-interface information, and there is no connection attribute. In this step, the connection attribute may not be carried, or the MAPS-MN may obtain the connection attribute from the modified stream selection policy sent by the terminal in this step, and carry it in the message.

 The purpose of the flow migration response message is to respond to the flow migration indication message of step 906. The name of the message may be other names as long as the message completes the above functions.

 Step 912, the same step 814.

Step 913: The MAG1-MN notifies the MAG of the opposite end to update the address mapping and the connection attribute. The address mapping refers to the HNP and MAG-MN address mapping of the MN or HNP and CoA. Mapping.

 The MAGI-MN can obtain all the communication peer CNs of the MN in the locally maintained communication peer table, and query the corresponding MAG-CN address or the CN's CoA according to the CN, and then to the MAG-CN (the MAG referred to in this embodiment). -CN1) Sends a CoA update message carrying the HNP of the MN, the source MAG address (ie MAG1-MN address) or the source CoA (ie CoAl), the target MAG address (ie MAG2-MN address) or the target CoA (ie CoA2) , also carries the connection properties of the migration stream.

 Alternatively, the MAG1-MN obtains all the communication peer CNs of the MN in the locally maintained communication peer table, and obtains the MAPS-CN corresponding to the CN, and then sends a CoA update message to the MAPS-CN, where the message carries the HNP of the MN. The source MAG address (ie, the MAGI-MN address) or the source CoA (ie, CoAl), the target MAG address (ie, the MAG2-MN address) or the target CoA (ie, CoA2), and also carries the connection attribute of the migrated stream. The CoA update message is forwarded by the MAPS-CN to the MAG of the CN.

 If the connection attribute is not carried in step 911, the MAG1-MN may obtain the connection attribute from the modified stream selection policy sent by the terminal in this step.

 The purpose of the CoA update message is to send the address mapping of HNP1 and MAG2-MN or the HNP1 and CoA2 mapping and connection attributes to the MAG-CN. The name of the message can be another name as long as the message carries the above information.

 Step 914: After receiving the message, the MAG-CN1 obtains the MAG2-MN address or CoA2 and attributes in the message, updates the locally stored MN address mapping table, and sets the HNP1-CoAl (or MAG1-MN address)-connection attribute. Change to the HNPl-CoA2 (or MAG2-MN address) country connection attribute. MAG-CN1 returns a CoA update response message.

 The purpose of the CoA update response message is to respond to the CoA update message of step 913, and the name of the message may be another name as long as the message completes the above functions.

 When the interface information in step 905 is a CoA or MAG address, steps 913 and 914 are executed after step 906, and need not wait until step 911 is completed.

 Step 915, the same step 815.

In this embodiment, the policy modification request may also be sent by the MAGI MN to the policy server. Send The mode is similar to the MAPS-MN class and can be divided into two ways, which are not described here.

 This example uses the flow migration initiated by the terminal as an example to describe the flow of the flow selection policy and the address mapping relationship caused by the flow migration. When a multi-access terminal is switched, that is, when the multi-access terminal changes the MAG of an interface due to the movement, or when an interface of the multi-access terminal fails for some reason (such as a failure, etc.) When serving the terminal, all the flows on the interface where the switching/failure occurs need to be switched to other interfaces of the terminal, and the flow of this embodiment can still be used. For the handover scenario, step 905 may be a handover request message.

 Embodiment 5

 FIG. 10 is a flow migration embodiment 5 of the present invention. The embodiment is applicable to the MN indicating that the MAG2-MN performs flow migration, and the MAPS-MN notifies the update peer. Specifically, the following steps are included:

 Steps 1001-1004, the same steps as 801-804.

 Step 1005: The MN sends a flow migration indication message to the MAG2-MN to instruct the MAG2-MN to perform flow migration. The source interface information and the modified flow selection policy need to be carried in the message. It also optionally carries the target interface information and the flow information that needs to be migrated.

 The target interface information refers to related information of the target interface to which the flow needs to be migrated, and includes at least one of the following: a link local address of the terminal interface 2, a MAC address of the MAG2, a CoA2, an interface identifier of the terminal interface 2, and a terminal interface 2 MAC address. When the flow migration indication message is sent to the MAG2-MN through unicast mode, the target interface information may not be carried, otherwise the field must be carried in the message.

 The modified flow selection policy refers to the flow selection policy formulated by the MN in step 1004.

 The source interface information refers to the information about the source interface that needs to be migrated out, including at least one of the following: linklocal address of terminal interface 1, MAC address of MAG1, CoAl, interface identifier of terminal interface 1, and MAC address of terminal interface 1. .

 The flow information to be migrated refers to the information about the flow that needs to be migrated, and is used to determine a specific flow, which may be a five-tuple or other indication information. You need to carry this information when you need to migrate individual flows on the interface. If all the flows on the interface are migrated, you do not need to carry this field.

The purpose of the flow migration indication message is to notify the MAG to perform flow migration, and carry related interface information and flow information, which may be implemented by adding a new message, or by tampering with the existing message. The name of the message can be other names as long as the message implements the above functions. Step 1006: The MAG2-MN forwards the received flow migration indication message to the MAPS-MN. The message needs to carry the MN identity (MN ID), the target interface information, and the modified flow selection policy. The optional source interface information and the flow information to be migrated are optional.

 Steps 1007-1010, the same steps as 807-810.

 Step 1011: The MAPS-MN returns a flow migration response message to the MAG2-MN.

 The purpose of the flow migration response message is to respond to the flow migration indication message of step 1006, and the name of the message may be another name as long as the message completes the above functions.

 Step 1012: The MAG2-MN returns a flow migration response message to the MN.

 The purpose of the flow migration response message is to respond to the flow migration indication message of step 1005, and the name of the message may be another name as long as the message completes the above functions.

 Steps 1013 and 1014 are optional steps. When the communication peer table is maintained by the MAG, these two steps need to be performed. The MAPS-MN sends a communication peer acquisition request message to the MAGI-MN, where the message carries the MN ID or the HNP1 of the MN. The MAG1-MN includes all the communication peers of the locally maintained MN in the communication peer acquisition response message and returns it to the MAPS-MN.

 Steps 1013 and 1014 can be performed after step 1006.

 The purpose of obtaining the request message by the communication peer is to obtain the communication peer of the MN from the MAG, which may be implemented by adding a new message, or by modifying an existing message, and the name of the message may be another name, as long as the message implements the above The function is OK.

 The purpose of obtaining the response message by the communication peer is to respond to the communication request message, which may be implemented by adding a new message, or by tampering with the existing message. The name of the message may be other names, as long as the name The message implements the above functions.

 Steps 1015 and 1016 are the same as steps 811 and 812. When the communication peer table is maintained by the MAPS, these two steps can be performed after step 1006. When the communication peer table is maintained by the MAG, the two steps need to be executed after steps 1013 and 1014 are completed.

 Step 1017, the same step 815.

In this embodiment, the policy modification request may also be sent by the MAG2 MN to the policy server. Send The mode is similar to the MAPS-MN class and can be divided into two ways, which are not described here.

 This example uses the flow migration initiated by the terminal as an example to describe the flow of the flow selection policy and the address mapping relationship caused by the flow migration. When a multi-access terminal is switched, that is, when the multi-access terminal changes the MAG of an interface due to the movement, or when an interface of the multi-access terminal fails for some reason (such as a failure, etc.) When serving the terminal, all the flows on the interface where the switching/failure occurs need to be switched to other interfaces of the terminal, and the flow of this embodiment can still be used. For the handover scenario, step 1005 may be a handover request message.

 Embodiment 6

 11 is a flow migration embodiment 6 of the present invention. This embodiment is applicable to the MN indicating that the MAG2-MN performs flow migration, and the MAG1-MN notifies the update peer. Specifically, the following steps are included:

 Steps 1101-1112, the same steps as 1001-1012.

 Step 1113: The MAPS-MN sends an update peer indication message to the MAG1-MN, where the message carries the MN ID or the HNP1 of the MN, the target MAG address (ie, the MAG2-MN address) or the target CoA (ie, CoA2), and carries the migration. The connection properties of the stream.

 If the MAPS-MN does not obtain a flow selection policy or a connection attribute from the policy server when the MN registers with the MAPS-MN, the content of the MN registration information table in the MAPS-MN is: MNID-HNP-interface information, and no connection attribute exists. . In this step, the connection attribute may not be carried, or the MAPS-MN may obtain the connection attribute from the modified stream selection policy sent by the terminal in this step, and carry it in the message.

 Step 1114: The MAGI-MN notifies the MAG of the opposite end to update the address mapping and the connection attribute. The address mapping refers to the mapping between the HNP of the MN and the MAG-MN address mapping or HNP and CoA.

The MAGI-MN can obtain all the communication peer CNs of the MN in the locally maintained communication peer table, and query the corresponding MAG-CN address or the CN's CoA according to the CN, and then to the MAG-CN (the MAG referred to in this embodiment). -CN1) Sends a CoA update message carrying the HNP of the MN, the source MAG address (ie MAG1-MN address) or the source CoA (ie CoAl), the target MAG address (ie MAG2-MN address) or the target CoA (ie CoA2) , also carries the connection properties of the migration stream. Alternatively, the MAG1-MN obtains the communication peer CN of the MN in the locally maintained communication peer table, and obtains the MAPS-CN corresponding to the CN, and then sends a CoA update message to the MAPS-CN, where the message carries the HNP of the MN. The source MAG address (that is, the MAG1-MN address) or the source CoA (ie, CoAl), the target MAG address (ie, the MAG2-MN address) or the target CoA (ie, CoA2), and also carries the connection attribute of the migrated stream. The CoA update message is forwarded by the MAPS-CN to the MAG of the CN.

 If the connection attribute is not carried in step 1113, the MAG1-MN may obtain the connection attribute from the modified stream selection policy sent by the terminal in this step.

 The purpose of the CoA update message is to send the address mapping of HNP1 and MAG2-MN or the HNP1 and CoA2 mapping and connection attributes to the MAG-CN. The name of the message can be another name as long as the message carries the above information.

 Step 1115: After receiving the message, the MAG-CN1 obtains the MAG2-MN address or the CoA2 and the connection attribute in the message, updates the locally saved MN address mapping table, and sets the HNP1-CoAl (or MAG1-MN address)-connection attribute.爹 Change to the HNPl-CoA2 (or MAG2-MN address) country connection attribute. MAG-CN1 returns a CoA update response message.

 The purpose of the CoA update response message is to respond to the CoA update message of step 1114. The name of the message may be other names as long as the message completes the above functions.

 Step 1116, the MAG1-MN returns an update peer response message to the MAPS-MN.

 The purpose of updating the peer response message is to respond to the update peer indication message of step 1113. The name of the message may be other names as long as the message completes the above functions.

 Step 1117, the same as step 1017.

 The update of the policy server in this embodiment may also be initiated by the MAG2-MN without the MAPS MN performing (ie, in steps 1108, I, 09, and the MAPS MN is changed to MAG2 MN).

This example uses the flow migration initiated by the terminal as an example to describe the flow of the flow selection policy and the address mapping relationship caused by the flow migration. When a multi-access terminal is switched, that is, when the multi-access terminal changes the MAG of an interface due to the movement, or when an interface of the multi-access terminal fails for some reason (such as a failure, etc.) When serving the terminal, you need to switch all the flows on the interface where the switchover/failure occurs to other interfaces on the terminal. You can still use this implementation. The flow of the example. For the handover scenario, step 1105 may be a handover request message.

 Example 7

 12 is a flow migration embodiment of the present invention. This embodiment is applicable to the MN indicating that the MAG1-MN performs flow migration, and the MAG1-MN updates the policy server and notifies the update peer. Specifically, the following steps are included:

 Steps 1201-1205, the same steps as 801-805.

 In steps 1206-1207, the MAG1-MN notifies the policy server to modify the flow selection policy. The MAGI-MN sends a policy modification request message to the policy server, where the message carries the MN ID and the modified flow selection policy. When there is a flow selection policy based on the MN (that is, related to the MN) on the policy server, the policy server updates the flow selection policy related to the MN by using the modified flow selection policy carried in the message, and there is no MN-based on the policy server. When the flow selection policy is used, the policy server uses the modified flow selection policy carried in the message to add a flow selection policy to the MN. The policy server returns a policy modification response message.

 Alternatively, the MAG1-MN needs to perform this step only when there is a MN-based flow selection policy on the policy server, that is, the policy modification request message is sent.

 The purpose of the policy modification request message is to notify the policy server to update the flow selection policy related to the MN, which may be implemented by adding a new message, or by modifying an existing message, and the name of the message may be another name, as long as the message is implemented. The above functions are all.

 The purpose of the policy modification response message is to respond to the policy modification request message, which can be implemented by adding a new message, or by modifying an existing message. The name of the message can be other names, as long as the message implements the above functions. .

 Steps 1208 and 1209 are the same as steps 913 and 914.

 If the interface information in the step 1205 is a CoA or a MAG address, the steps 1208 and 1209 may be directly performed. Otherwise, the MAG1-MN needs to obtain the CoA2 or MAG2-MN address allocated by the MAG2-MN to the MN to the MAPS-MN. 1208, 1209.

 Steps 1210 and 1211 are the same as steps 814 and 815.

This example uses the flow migration initiated by the terminal as an example to describe the flow of the flow selection policy and the address mapping relationship caused by the flow migration. When a multiple access terminal switches, that is, when the multiple access terminal is When the movement causes the MAG of an interface to change, or when an interface of a multi-access terminal cannot serve the terminal for some reason (such as a failure, etc.), all the interfaces on which the handover/failure occurs will be required. The flow is all switched to other interfaces of the terminal, and the flow of this embodiment can still be used at this time. For the handover scenario, step 1205 may be a handover request message.

 In the foregoing process of the third embodiment to the seventh embodiment, the address mapping and policy update process of the peer MAG may also be initiated by the MAG2-MN or the policy server. In this case, the MAG2-MN or the policy server needs to first obtain the communication peer information of the MN in which the flow migration or handover occurs before initiating the update process.

 In the foregoing embodiments, the method of updating the peer MAG CN, initiating the policy server update, and initiating the MAPS-MN update may be combined with each other without conflict.

 Example eight

 Figure 13 is a flow chart based on the system architecture of Figure 2, where the policy server initiates policy modification and is notified by the MAPS-MN to update the peer. The specific steps are described as follows:

 Step 1301: The multiple access terminal MN has accessed the PMIPv6 domain, and uses the interface 1 to complete the registration in the MAPS-MN through the MAG1-MN, and uses the interface 2 to complete the registration in the MAPS-MN through the MAG2-MN.

 Step 1302, the MN performs data communication with CN1 on interface 1. The data of MN and CN1 is as shown in this step as MN MAGl-MN MAG-CNl ^CNl.

 Step 1303: The MN performs data communication with CN2 on interface 2. The data of MN and CN2 is as shown in this step as MN MAG2-MN MAG-CN2 CN2.

 In steps 1304-1305, the policy server changes the flow selection policy based on load sharing, signal strength, user intention, network planning, or operation and maintenance. The policy server sends a policy modification request message to the MAPS-MN, and carries the modified flow selection policy, optionally carrying the MN ID.

The MAPS-MN returns a policy modification response message.

 The purpose of the policy modification request message is to notify the MAPS to update the related flow selection policy, which may be implemented by adding a new message, or by modifying an existing message. The name of the message may be another name, as long as the message implements the above function. can.

The purpose of the policy modification response message is to respond to the policy modification request message. The message can be implemented by modifying the existing message. The name of the message can be other names, as long as the message implements the above functions.

 Step 1307, because the flow selection policy changes (for example, the gateway corresponding to the high QoS level is changed from MAG1 to MAG2), the MAPS-MN checks whether the updated flow selection policy is related to the connection attribute of the MN (for example, whether the MN passes the MAG1 at the same time) After registration with MAG2, a flow migration indication message is sent to MAG1-MN or MAG2-MN. The message carries the MN identifier (MN ID), the target interface information, and the modified flow selection policy. The optional source interface information and the flow information to be migrated are optional. If the flow migration indication message is sent to the MAG1-MN, the MAGI-MN may delete the associated flow or connection attribute.

 The target interface information refers to related information of the target interface to which the flow needs to be migrated, and includes at least one of the following: a link local address of the terminal interface 2, a MAC address of the MAG2, a CoA2, an interface identifier of the terminal interface 2, and a terminal interface 2 MAC address.

 The source interface information refers to the information about the source interface that needs to be migrated out, including at least one of the following: linklocal address of terminal interface 1, MAC address of MAG1, CoAl, interface identifier of terminal interface 1, and MAC address of terminal interface 1. .

 The flow information to be migrated refers to the flow related information that needs to be migrated, and is used to determine a specific flow, which may be a five-tuple or other indication information. When you need to migrate individual flows on an interface, you need to carry this information. If all the flows on the interface are migrated, you do not need to carry this field.

 The purpose of the flow migration indication message is to notify the MAG to perform flow migration, and to carry related interface information and flow information, which may be implemented by adding a new message, or by tampering with an existing message. The name of the message may be Other names, as long as the message implements the above functions.

 Step 1308: The MAG1-MN or the MAG2-MN forwards the received flow migration indication message to the ΜΝ. The message carries the target interface information and the modified flow selection policy. The optional source interface information and the flow information to be migrated are optional.

Step 1309, an optional step. The network initiates the flow migration, which means that the flow selection policy has changed. When there is a flow selection policy based on ΜΝ (that is, related to ΜΝ) on the MAPS-MN, the policy needs to be updated, and this step is performed. The MAPS-MN updates the local MN-related flow selection policy according to the modified flow selection policy carried in step 1304. This step occurs after 1304 and has no sequence with the other steps.

 Step 1310, an optional step. When the MN registers with the MAPS-MN, the MAPS-MN interacts with the policy server to obtain a flow selection policy or connection attribute, and then performs this step. To prevent the CN from actively transmitting packets, select the interface that has been migrated (that is, on interface 1), MAPS-MN needs to update the MN registration information table, and modify HNP1-source interface information-connection attribute 1 to HNP1-target. Interface information - connection attribute 1.

 This step can be performed after step 1304, and there is no order with other steps.

 Step 1311: The MAPS-MN notifies the MAG of the opposite end to update the address mapping and the connection attribute. This step can be performed after step 1304, and there is no order with other steps.

 The address mapping refers to the mapping between the MN's HNP and the MAG-MN address, or the MN's

HNP and CoA mapping.

 The MAPS-MN can obtain all the communication peer CNs of the MN in the communication peer table maintained locally or from the communication peer table obtained by the MAGI-MN, and query the corresponding MAG-CN address or the CN of the CN according to the CN, and then Sending a CoA update message to the MAG-CN (MAG-CN1 in this embodiment), where the message carries the HNP of the MN, the source MAG address (ie, the MAGI-MN address) or the source CoA (ie, CoAl), and the target MAG address (ie, The MAG2-MN address) or the target CoA (ie, CoA2) also carries the connection attribute of the migrated stream. The communication peer table obtains all the communication peer CNs of the MN, and obtains the MAPS-CN corresponding to the CN, and then sends a CoA update message to the MAPS-CN, where the message carries the HNP of the MN, and the source MAG address (ie, the MAG1-MN address) Or the source CoA (ie, CoAl), the target MAG address (ie, the MAG2-MN address) or the target CoA (ie, CoA2), and also carries the connection attribute of the migrated stream. The CoA update message is forwarded by the MAPS-CN to the MAG of the CN.

 The purpose of the CoA update message is to send the address mapping of HNP1 and MAG2-MN or the HNP1 and CoA2 mapping and connection attributes to the MAG-CN. The name of the message can be another name as long as the message carries the above information.

Step 1312: After receiving the message, the MAG-CN1 obtains the MAG2-MN address or the CoA2 and the connection attribute in the message, and updates the locally saved MN address mapping table to the HNP1-CoAl (or MAGl-MN address) - The connection attribute is changed to the HNPl-CoA2 (or MAG2-MN address) country connection attribute. MAG-CN1 returns a CoA update response message. This step can be performed after step 1311, and there is no order with other steps.

 The purpose of the CoA update response message is to respond to the CoA update message of step 1312, and the name of the message may be other names as long as the message completes the above functions.

 Step 1313: The MN returns a flow migration response message to the MAGI-MN or the MAG2-MN. The purpose of the flow migration response message is to respond to the flow migration indication message of step 1308. The name of the message may be another name as long as the message completes the above functions.

 This step can be executed after completing 1308, and there is no order with other steps.

 In step 1314, MAG1-MN or MAG2-MN returns a stream migration response message to the MAPS-MN.

 The purpose of the flow migration response message is to respond to the flow migration indication message of step 1307, and the name of the message may be another name as long as the message completes the above functions.

 Step 1315: After the stream migration is completed, the MN performs data communication with CN1 on interface 2. The data message of MN and CN1 is MN OMAG2-MN ^MAG-CNl ^CNl as shown in this step.

 This example uses the flow migration initiated by the network as an example to describe the flow of the flow selection policy and the address mapping relationship caused by the flow migration. When a multi-access terminal is switched, that is, when the multi-access terminal changes the MAG of an interface due to the movement, or when an interface of the multi-access terminal fails for some reason (such as a failure, etc.) When serving the terminal, all the flows on the interface where the switching/failure occurs need to be switched to other interfaces of the terminal. The flow of this embodiment can still be used. For handover scenarios, steps 1307 and 1308 may be handover request messages.

 There are two methods for modifying the flow selection policy of the terminal as follows:

 Update terminal method one:

 As mentioned in Figure 13, step 1307, step 1308, step 1313 and step 1314 are performed after the MAPS-MN obtains the modified stream selection policy.

 Update terminal method two:

Can be changed to the MAG1-MN by the policy server after the flow selection policy is changed. The MAG2-MN directly sends a stream migration indication message to replace step 1307 in FIG. The message needs to carry the MN identity (MN ID), the target interface information, and the modified flow selection policy. The optional source interface information and the flow information to be migrated are optional. If sent to MAG1-MN, MAGI-MN can delete the associated stream or connection attribute. At the same time, MAGI-MN or MAG2-MN returns a stream migration response message to the policy server to replace step 1314 in FIG. Steps 1308 and 1313 do not need to be changed.

 There are two ways to modify the MAG-CN stored flow selection policy:

 Update MAG-CN Method 1:

 As mentioned in Figure 13, step 1311 and step 1312 are performed after the MAPS-MN obtains the modified stream selection policy.

 Update MAG-CN Method 2:

 After the modified flow selection policy information is obtained by the MAG1-MN or the MAG2-MN, the policy server notifies the MAG to update the address mapping and the connection attribute of the peer end to replace the above step 1311. At the same time, after receiving the message, MAG-CN1 obtains the MAG2-MN address or CoA2 and connection attribute in the message, updates the locally stored MN address mapping table, and modifies the HNP1-CoAl (or MAG1-MN address)-connection attribute. For the HNPl-CoA2 (or MAG2-MN address) country connection attribute. The MAG-CN1 returns a CoA Update Response message to the MAG1-MN or MAG2-MN or the policy server to replace the above step 1312.

 After the network side decides to change the flow selection policy, it needs to perform policy update on the terminal, MAPS, and MAG-CN respectively. The flow selection policy update for each network element (terminal, MAPS, and MAG-CN) has the above multiple methods. When updating, various methods can be combined in any way to complete the information modification of the relevant network element, and no longer, for example, repeated description.

 Example nine

 Figure 14 is a flow chart based on the system architecture of Figure 2, where the MAPS-MN initiates policy modification and the MAPS-MN notifies the peer. The specific steps are described as follows:

Step 1401: The multiple access terminal MN has accessed the PMIPv6 domain, and uses the interface 1 to complete registration in the MAPS-MN through the MAG1-MN, and uses the interface 2 to complete registration in the MAPS-MN through the MAG2-MN. In step 1402, the MN performs data communication with CN1 on interface 1. The data of MN and CN1 is as shown in this step as MN MAGI-MN MAG-CNI ^CNl.

 Step 1403: The MN performs data communication with CN2 on interface 2. The data of MN and CN2 is as shown in this step as MNoMAG2-MNoMAG-CN2oCN2.

 Step 1404: The MAPS-MN prepares to initiate a flow migration from MAG1 to MAG2 or prepares to modify a flow selection policy based on load sharing, signal strength, user intention, network planning, or operation and maintenance.

 Step 1405 and step 1406, the MAPS-MN sends a policy modification request message to the policy server, where the message carries the MN ID and the modified flow selection policy. The Policy Server updates the MN-related flow selection policy and returns a policy modification response message. This step occurs after 1404, and there are no other steps in the other steps.

 The purpose of the policy modification request message is to notify the MAPS to update the flow selection policy related to the MN, which may be implemented by adding a new message, or by modifying an existing message, and the name of the message may be another name, as long as the message implements the above The function is OK.

 The purpose of the policy modification response message is to respond to the policy modification request message, which can be implemented by adding a new message, or by modifying an existing message. The name of the message can be other names, as long as the message implements the above functions. .

 Step 1407: The MAPS-MN sends a flow migration indication message to the MAG1-MN or the MAG2-MN because the flow selection policy changes (for example, the gateway corresponding to the high QoS level is changed from MAG1 to MAG2). The message needs to carry the MN identity (MN ID), the target interface information, and the modified flow selection policy, optionally carrying the source interface information and the flow information to be migrated. If sent to MAG1-MN, MAG1-MN can delete the associated stream or connection attribute. This step occurs after 1404, in no order with the other steps.

 The target interface information refers to related information of the target interface to which the flow needs to be migrated, and includes at least one of the following: a link local address of the terminal interface 2, a MAC address of the MAG2, a CoA2, an interface identifier of the terminal interface 2, and a terminal interface 2 MAC address.

The source interface information refers to the information about the source interface that needs to be migrated out, including at least one of the following: linklocal address of terminal interface 1, MAC address of MAG1, CoAl, and terminal interface 1 The interface identifier, the MAC address of the terminal interface 1.

 The flow information to be migrated refers to the flow related information that needs to be migrated, and is used to determine a specific flow, which may be a five-tuple or other indication information. When you need to migrate individual flows on an interface, you need to carry this information. If all the flows on the interface are migrated, you do not need to carry this field.

 The purpose of the flow migration indication message is to notify the MAG to perform flow migration, and to carry related interface information and flow information, which may be implemented by adding a new message, or by tampering with an existing message. The name of the message may be Other names, as long as the message implements the above functions.

 Step 1408: The MAG1-MN or MAG2-MN forwards the received flow migration indication message to the UI. The message carries the target interface information and the modified flow selection policy. The optional source interface information and the flow information to be migrated are optional.

 Step 1409, an optional step. The network initiates the flow migration, which means that the flow selection policy has changed. When there is a flow selection policy based on ΜΝ (that is, related to the MA) on the MAPS-MN, the policy needs to be updated and this step is performed. The MAPS-MN updates the local terminal-related flow selection policy according to the modified flow selection policy in step 1404. This step occurs after 1404 and has no prior sequence with the other steps.

 Step 1410, an optional step. When the MN registers with the MAPS-MN, the MAPS-MN interacts with the policy server to obtain a flow selection policy or obtains a connection attribute, and then performs this step. To prevent the CN from actively transmitting packets, select the interface that has been migrated (that is, on interface 1), MAPS-MN needs to update the MN registration information table, and change from HNP1-source interface information-connection attribute 1 to HNP1-target. Interface information - connection attribute 1. This step can be performed after step 1404, and there is no order for other steps.

 Step 1411: The MAPS-MN notifies the MAG of the peer end of the address mapping and connection attributes. This step can be performed after step 1404, and there is no order with other steps.

 The address mapping refers to the mapping between the HNP of the MN and the MAG-MN address mapping or HNP and CoA.

The MAPS-MN can obtain all the communication peer CNs of the MN in the communication peer table maintained locally or from the communication peer table obtained by the MAGI-MN, and query the corresponding MAG-CN address or the CN of the CN according to the CN, and then Send to the MAG-CN (MAG-CN1 referred to in this embodiment) CoA update message, the message carries the HNP of the MN, the source MAG address (ie MAGI-MN address) or the source CoA (ie CoAl), the target MAG address (ie MAG2-MN address) or the target CoA (ie CoA2), and also carries The connection property of the migration stream. The communication peer table obtains all the communication peer CNs of the MN, and obtains the MAPS-CN corresponding to the CN, and then sends a CoA update message to the MAPS-CN, where the message carries the HNP of the MN, and the source MAG address (ie, the MAG1-MN address) Or the source CoA (ie, CoAl), the target MAG address (ie, the MAG2-MN address) or the target CoA (ie, CoA2), and also carries the connection attribute of the migrated stream. The CoA update message is forwarded by the MAPS-CN to the MAG of the CN.

 The purpose of the CoA update message is to send the address mapping of HNP1 and MAG2-MN or the HNP1 and CoA2 mapping and connection attributes to the MAG-CN. The name of the message can be another name as long as the message carries the above information.

 Step 1412: After receiving the message, the MAG-CN1 obtains the MAG2-MN address or the CoA2 and the connection attribute in the message, updates the locally stored MN address mapping table, and sets the HNP1-CoAl (or MAG1-MN address)-connection attribute.爹 Change to the HNPl-CoA2 (or MAG2-MN address) country connection attribute. MAG-CN1 returns a CoA update response message. This step can be performed after step 1411, and there is no order with other steps.

 The purpose of the CoA update response message is to respond to the CoA update message of step 1412, and the name of the message may be other names as long as the message completes the above functions.

 Step 1413: The MN returns a flow migration response message to the MAG1-MN or the MAG2-MN. The purpose of the flow migration response message is to respond to the flow migration indication message of step 1408, and the name of the message may be another name as long as the message completes the above functions.

 This step can be performed after completing 1408, and there is no order with other steps.

 In step 1414, the MAG1-MN or MAG2-MN returns a stream migration response message to the MAPS-MN.

 The purpose of the flow migration response message is to respond to the flow migration indication message of step 1407, and the name of the message may be another name as long as the message completes the above functions.

Step 1415: After the flow migration is completed, the MN performs data communication with CN1 on the interface 2. MN The data message with CN1 is MN oMAG2-MN oMAG-CNl oCNl as shown in this step. This example uses the flow migration initiated by the network as an example to describe the flow of the flow selection policy and the address mapping relationship caused by the flow migration. When a multi-access terminal is switched, that is, when the multi-access terminal changes the MAG of an interface due to the movement, or when an interface of the multi-access terminal fails for some reason (such as a failure, etc.) When serving the terminal, all the flows on the interface where the switching/failure occurs need to be switched to other interfaces of the terminal. The flow of this embodiment can still be used. For handover scenarios, steps 1407 and 1408 may be handover request messages.

 There are two methods for modifying the flow selection policy of the terminal as follows:

 Update terminal method one:

 As mentioned in Figure 14, step 1407, step 1408, step 1413 and step 1414 are performed after the MAPS-MN decides to modify the stream selection policy.

 Update terminal method two:

 After obtaining the new flow selection policy, the policy server may directly send a flow migration indication message to the MAG1-MN or the MAG2-MN to replace the step 1407 in FIG. The message carries the MN identifier (MN ID), the target interface information, and the modified flow selection policy. The optional source interface information and the flow information to be migrated are optional. If sent to MAG1-MN, MAGI-MN can delete the associated stream or connection attribute. At the same time, MAGI-MN or MAG2-MN returns a stream migration response message to the policy server to replace step 1414 in FIG. Steps 1408 and 1413 do not need to be changed.

 There are two ways to modify the MAG-CN stored flow selection policy:

 Update MAG-CN Method 1:

 As mentioned in Figure 14, steps 1411 and 1412 are performed after the MAPS-MN decides to make a stream selection policy modification. Update MAG-CN Method 2:

After obtaining the modified flow selection policy information, the MAG1-MN or the MAG2-MN or the policy server notifies the MAG update address mapping and connection attribute of the opposite end to replace the above step 1411. At the same time, after receiving the message, the MAG-CN obtains the MAG2-MN address or the CoA2 and the connection attribute in the message, and updates the locally stored MN address mapping table, and the HNP1-CoAl (or MAG1-MN) Address) - The connection attribute is modified to HNPl-CoA2 (or MAG2-MN address) - connection attribute.

MAG-CN1 returns a CoA update response message to the MAG1-MN or MAG2-MN or the policy server to replace step 1412 above.

 There are two ways to modify the flow selection policy of the policy server:

 Update Policy Server Method One:

 As mentioned in Figure 14, step 1405 and step 1406 are performed after the MAPS-MN obtains the modified stream selection policy.

 Update Policy Server Method 2:

 After obtaining the modified flow selection policy information, the MAG1-MN or the MAG2-MN sends a policy modification request message to the policy server, where the message carries the MN ID and the modified flow selection policy. The policy server updates the flow selection policy associated with the MN and returns a policy modification response message to replace step 1405 and step 1406 above.

 After the network side decides to change the flow selection policy, it needs to perform policy update on the terminal, MAPS-MN, policy server, and MAG-CN respectively, and perform flow selection on each network element (terminal, MAPS-MN, policy server, and MAG-CN). The policy update has the above multiple manners. In the update, various modes can be combined in any way to complete the information modification of the relevant network element, and no longer, for example, repeated description. In addition, if the policy related to the MN is not stored on the policy server, the operation of the flow selection policy update may not be performed on the policy server.

 Example ten

 Figure 15 is a scenario based on the system architecture of Figure 2, MAG1-MN initiates policy modification, by

The MAPS-MN notifies the flow chart of updating the peer. The specific steps are described as follows:

 Step 1501: The multiple access terminal MN has accessed the PMIPv6 domain, and uses the interface 1 to complete the registration in the MAPS-MN through the MAG1-MN, and uses the interface 2 to complete the registration in the MAPS-MN through the MAG2-MN.

 Step 1502, the MN performs data communication with CN1 on interface 1. The data of MN and CN1 is as shown in this step as MN MAGl-MN MAG-CNl CNl.

Step 1503, the MN performs data communication with CN2 on interface 2. The data of MN and CN2 is as shown in this step as MNoMAG2-MNoMAG-CN2oCN2. Step 1504: MAG1-MN or MAG2-MN prepares to initiate a flow migration from MAG1 to MAG2 or prepares to modify a flow selection policy based on load sharing, signal strength, user intention, network planning, or operation and maintenance. Step 1504a, MAG1-MN or MAG2-MN sends a flow migration indication message to the MAPS-MN. The message needs to carry the MN identifier (MN ID), the target interface information, and the modified flow selection policy. The optional source interface information and the flow information to be migrated are optional. At the same time, it is optional to carry the communication peer table of the MN (that is, all HoA or HNP or ID of the peer that communicates with the MN).

 When the communication peer table of the MN is maintained by the MAPS, the information may not be carried, otherwise the field must be carried.

 Steps 1505 - 1506, the MAPS-MN sends a policy modification request message to the policy server, where the message carries the MN ID and the modified flow selection policy. The Policy Server updates the MN-related flow selection policy and returns a policy modification response message. This step occurs after 1504, in no order with the other steps.

 The purpose of the policy modification request message is to notify the MAPS to update the flow selection policy related to the MN, which may be implemented by adding a new message, or by tampering with the existing message, and the name of the message may be another name, as long as the name The message implements the above functions.

 The purpose of the policy modification response message is to respond to the policy modification request message, which can be implemented by adding a new message, or by modifying an existing message. The name of the message can be other names, as long as the message implements the above functions. .

 Step 1508, an optional step. The network initiates the flow migration, which means that the flow selection policy has changed. When there is a flow selection policy based on the MN (that is, related to the MN) on the MAPS-MN, the policy needs to be updated, and this step is performed. The MAPS-MN updates the local terminal-related flow selection policy according to the modified flow selection policy in step 1504a. This step occurs after 1504a, in no order with the other steps.

Step 1509, an optional step. When the MN registers with the MAPS-MN, the MAPS-MN interacts with the policy server to obtain a flow selection policy or obtains a connection attribute, and then performs this step. To prevent the CN from actively transmitting packets, select the interface that has been migrated (that is, on interface 1), MAPS-MN needs to update the MN registration information table, and change from HNP1-source interface information-connection attribute 1 to HNP1-target. Interface information - connection attribute 1. This step can be performed after step 1504a, instead His steps are in no order.

 Step 1510: The MAPS-MN notifies the MAG of the peer end of the address mapping and connection attributes. This step can be performed after step 1504a, and there is no order with other steps.

 The address mapping refers to the mapping between the HNP of the MN and the MAG-MN address mapping or HNP and CoA.

 The MAPS-MN can obtain all the communication peer CNs of the MN in the communication peer table maintained locally or from the communication peer table obtained by the MAGI-MN, and query the corresponding MAG-CN address or the CN of the CN according to the CN, and then Sending a CoA update message to the MAG-CN (MAG-CN1 in this embodiment), where the message carries the HNP of the MN, the source MAG address (ie, the MAGI-MN address) or the source CoA (ie, CoAl), and the target MAG address (ie, The MAG2-MN address) or the target CoA (ie, CoA2) also carries the connection attribute of the migrated stream. The communication peer table obtains all the communication peer CNs of the MN, and obtains the MAPS-CN corresponding to the CN, and then sends a CoA update message to the MAPS-CN, where the message carries the HNP of the MN, and the source MAG address (ie, the MAG1-MN address) Or the source CoA (ie, CoAl), the target MAG address (ie, the MAG2-MN address) or the target CoA (ie, CoA2), and also carries the connection attribute of the migrated stream. The CoA update message is forwarded by the MAPS-CN to the MAG of the CN.

 The purpose of the CoA update message is to send the address mapping of HNP1 and MAG2-MN or the HNP1 and CoA2 mapping and connection attributes to the MAG-CN. The name of the message can be another name as long as the message carries the above information.

 Step 1511: After receiving the message, the MAG-CN1 obtains the MAG2-MN address or the CoA2 and the connection attribute in the message, updates the locally stored MN address mapping table, and sets the HNP1-CoAl (or MAG1-MN address)-connection attribute.爹 Change to the HNPl-CoA2 (or MAG2-MN address) country connection attribute. MAG-CN1 returns a CoA update response message. This step can be performed after step 1510, and there is no order with other steps.

 The purpose of the CoA update response message is to respond to the CoA update message of step 1511, and the name of the message may be other names as long as the message completes the above functions.

Step 1512: The MAPS-MN returns a flow migration response to the MAGI-MN or MAG2-MN. Interest.

 The purpose of the flow migration response message is to respond to the flow migration indication message of step 1504a, and the name of the message may be another name as long as the message completes the above functions.

 This step can be performed after completing 1504a, and there is no order with other steps.

 Step 1513: The MAG1-MN or the MAG2-MN sends a flow migration indication message to the MN. The message carries the target interface information and the modified flow selection policy. The optional source interface information and the flow information to be migrated are optional.

 This step can be performed after completing 1504, and there is no order with other steps.

 The target interface information refers to related information of the target interface to which the flow needs to be migrated, and includes at least one of the following: a link local address of the terminal interface 2, a MAC address of the MAG2, a CoA2, an interface identifier of the terminal interface 2, and a terminal interface 2 MAC address.

 The source interface information refers to the information about the source interface that needs to be migrated out, including at least one of the following: linklocal address of terminal interface 1, MAC address of MAG1, CoAl, interface identifier of terminal interface 1, and MAC address of terminal interface 1. .

 The flow information to be migrated refers to the flow related information that needs to be migrated, and is used to determine a specific flow, which may be a five-tuple or other indication information. When you need to migrate individual flows on an interface, you need to carry this information. If all the flows on the interface are migrated, you do not need to carry this field.

 The purpose of the flow migration indication message is to notify the MN to perform flow migration, and to carry related interface information and flow information, which may be implemented by adding a new message, or by tampering with an existing message. The name of the message may be Other names, as long as the message implements the above functions.

 In step 1514, the flow migration response message is returned to the MAG1-MN or MAG2-MN. The purpose of the flow migration response message is to respond to the flow migration indication message of step 1513, and the name of the message may be another name as long as the message completes the above functions. This step can be performed after completing 1513, and there is no order with other steps.

 In step 1515, after the stream migration is completed, data communication with CN1 is performed on interface 2.数据 The data message with CN1 is MN MAG2-MN MAG-CNl CNl as shown in this step.

This example uses the flow migration initiated by the network as an example to describe the flow of the flow selection policy and the address mapping relationship caused by the flow migration. When multiple access terminals switch, that is, when multiple access terminals are When the movement causes the MAG of an interface to change, or when an interface of the multiple access terminal cannot serve the terminal for some reason (such as a failure, etc.), it needs to be on the interface where the handover/failure occurs. All the flows are switched to other interfaces of the terminal, and the flow of this embodiment can still be used at this time. For handover scenarios, steps 1504a and 1513 may be handover request messages.

 There are two methods for modifying the flow selection policy stored in the MAPS-MN:

 Update MAPS-MN Method 1:

 As mentioned in Figure 15, step 1504a and step 1512 are performed after MAG1-MN decides to modify the stream selection policy.

 Update MAPS-MN Method 2:

 After obtaining the new flow selection policy, the policy server may directly send a policy modification request message to the MAPS-MN, and carry the modified flow selection policy, optionally with the MN ID. The MAPS-MN returns a policy modification response message to replace steps 1504a and 1512 in FIG. The method for obtaining the new flow selection policy by the policy server may be: MAG1-MN or MAG2-MN sends a policy modification request message to the policy server, where the message carries the MN ID and the modified flow selection policy.

 There are two ways to modify the MAG-CN stored flow selection policy:

 Update MAG-CN Method 1:

 As mentioned in Fig. 15, step 1510 and step 1511 are performed after the MAPS-MN obtains the modified stream selection policy.

 Update MAG-CN Method 2:

 After the MAGI-MN or the MAG2-MN or the policy server determines or obtains the modified flow selection policy information, the MAG of the opposite end is notified to update the address mapping and the connection attribute to replace the above step 1510. At the same time, after receiving the message, the MAG-CN obtains the MAG2-MN address or CoA2 and the connection attribute in the message, updates the locally stored MN address mapping table, and sets the HNP1-CoAl (or MAG1-MN address)-connection attribute. Change to the HNPl-CoA2 (or MAG2-MN address) country connection attribute. The MAG-CN1 returns a CoA Update Response message to the MAG1-MN or MAG2-MN or the Policy Server to replace the above step 1511.

 There are two ways to modify the flow selection policy of the policy server:

Update Policy Server Method One: As mentioned in FIG. 15, step 1505 and step 1506 are performed after the MAPS-MN acquires the modified stream selection policy.

 Update Policy Server Method 2:

 After the MAG1-MN or MAG2-MN decides to modify the flow selection policy, it sends a policy modification request message to the policy server, where the message carries the MN ID and the modified flow selection policy. The policy server updates the flow selection policy associated with the MN and returns a policy modification response message to replace step 1505 and step 1506 above.

 After the network side decides to change the flow selection policy, it needs to perform policy update on the terminal, MAPS-MN, policy server, and MAG-CN respectively, and perform flow selection on each network element (terminal, MAPS-MN, policy server, and MAG-CN). The policy update has the above multiple manners. In the update, various modes can be combined in any way to complete the information modification of the relevant network element, and no longer, for example, repeated description. In addition, if the policy related to the MN is not stored on the policy server, the operation of the flow selection policy update may not be performed on the policy server.

 Embodiment 11

 FIG. 16 is a flowchart of the system architecture of FIG. 2, the MAG1-MN notifying the MAPS-MN to initiate policy modification, and the MAPS-MN notifying the update peer. The specific steps are described as follows:

 Step 1601: The multiple access terminal MN has accessed the PMIPv6 domain, and uses the interface 1 to complete the registration in the MAPS-MN through the MAG1-MN, and uses the interface 2 to complete the registration in the MAPS-MN through the MAG2-MN.

 Step 1602, the MN performs data communication with CN1 on interface 1. The data of MN and CN1 is as shown in this step as MN MAGl-MN MAG-CNl ^CNl.

 Step 1603, the MN performs data communication with CN2 on interface 2. The data of MN and CN2 is as shown in this step as MN MAG2-MN MAG-CN2 CN2.

 Step 1604: MAG1-MN or MAG2-MN prepares to initiate flow migration from MAG1 to MAG2 based on load sharing, signal strength, user willingness, network planning, or operation and maintenance.

Step 1605: The MAG1-MN or the MAG2-MN sends a flow migration indication message to the MAPS-MN. The message needs to carry the MN identity (MN ID) and source interface information. Step 1606: The MAPS-MN determines the migrated target MAG (MAG2) and the flow selection policy, prepares to initiate a flow migration from MAG1 to MAG2 or prepares to modify the flow selection policy.

 Step 1607 and step 1608, the MAPS-MN sends a policy modification request message to the policy server, where the message carries the MN ID and the modified flow selection policy. The Policy Server updates the MN-related flow selection policy and returns a policy modification response message. This step occurs after 1606, and there are no other steps in the other steps.

 The purpose of the policy modification request message is to notify the MAPS to update the flow selection policy related to the MN, which may be implemented by adding a new message, or by modifying an existing message, and the name of the message may be another name, as long as the message implements the above The function is OK.

 The purpose of the policy modification response message is to respond to the policy modification request message, which can be implemented by adding a new message, or by modifying an existing message. The name of the message can be other names, as long as the message implements the above functions. .

 Step 1609, an optional step. When the MN registers with the MAPS-MN, the MAPS-MN interacts with the policy server to obtain a flow selection policy or obtains a connection attribute, and then performs this step. To prevent the CN from actively transmitting packets, select the interface that has been migrated (that is, on interface 1), MAPS-MN needs to update the MN registration information table, and change from HNP1-source interface information-connection attribute 1 to HNP1-target. Interface information - connection attribute 1. This step can be performed after step 1606, and there is no order for other steps.

 Step 1610: The MAPS-MN notifies the MAG of the opposite end to update the address mapping and the connection attribute. This step can be performed after step 1606, and there is no order with other steps.

 The address mapping refers to the mapping between the HNP of the MN and the MAG-MN address mapping or HNP and CoA.

The MAPS-MN can obtain all the communication peer CNs of the MN in the communication peer table maintained locally or from the communication peer table obtained by the MAGI-MN, and query the corresponding MAG-CN address or the CN of the CN according to the CN, and then Sending a CoA update message to the MAG-CN (MAG-CN1 in this embodiment), where the message carries the HNP of the MN, the source MAG address (ie, the MAGI-MN address) or the source CoA (ie, CoAl), and the target MAG address (ie, The MAG2-MN address) or the target CoA (ie, CoA2) also carries the connection attribute of the migrated stream. The communication peer table obtains all the communication peer CNs of the MN, and obtains the MAPS-CN corresponding to the CN, and then sends a CoA update message to the MAPS-CN, where the message carries the HNP of the MN, and the source MAG address (ie, the MAG1-MN address) Or the source CoA (ie, CoAl), the target MAG address (ie, the MAG2-MN address) or the target CoA (ie, CoA2), and also carries the connection attribute of the migrated stream. The CoA update message is forwarded by the MAPS-CN to the MAG of the CN.

 The purpose of the CoA update message is to send the address mapping of HNP1 and MAG2-MN or the HNP1 and CoA2 mapping and connection attributes to the MAG-CN. The name of the message can be another name as long as the message carries the above information.

 Step 1611: After receiving the message, the MAG-CN1 obtains the MAG2-MN address in the message or

CoA2 and connection attributes, update the locally stored MN address mapping table, and change the HNPl-CoAl (or MAG1-MN address)-connection attribute to the HNPl-CoA2 (or MAG2-MN address) country connection attribute. MAG-CN1 returns a CoA update response message. This step can be performed after step 1610, and there is no order with other steps.

 The purpose of the CoA update response message is to respond to the CoA update message of step 1611, and the name of the message may be other names as long as the message completes the above functions.

 Step 1612: The MAPS-MN returns a flow migration response message to the MAGI-MN or the MAG2-MN, and carries the target interface information and the modified flow selection policy.

 The purpose of the flow migration response message is to respond to the flow migration indication message of step 1605, and the name of the message may be another name as long as the message completes the above functions.

 This step can be performed after completing 1606, and there is no order with other steps.

 Step 1613: The MAG1-MN or the MAG2-MN sends a flow migration indication message to the MN. The message carries the target interface information and the modified flow selection policy. The optional source interface information and the flow information to be migrated are optional.

 The target interface information refers to related information of the target interface to which the flow needs to be migrated, and includes at least one of the following: a link local address of the terminal interface 2, a MAC address of the MAG2, a CoA2, an interface identifier of the terminal interface 2, and a terminal interface 2 MAC address.

Source interface information refers to information about the source interface that needs to be migrated out, including at least the following One: the linklocal address of the terminal interface 1, the MAC address of the MAG1, the CoAl, the interface identifier of the terminal interface 1, and the MAC address of the terminal interface 1.

 The flow information to be migrated refers to the flow related information that needs to be migrated, and is used to determine a specific flow, which may be a five-tuple or other indication information. When you need to migrate individual flows on an interface, you need to carry this information. If all the flows on the interface are migrated, you do not need to carry this field.

 The purpose of the flow migration indication message is to notify the MN to perform flow migration, and to carry related interface information and flow information, which may be implemented by adding a new message, or by tampering with an existing message. The name of the message may be Other names, as long as the message implements the above functions.

 In step 1614, the flow migration response message is returned to the MAG1-MN or MAG2-MN.

 The purpose of the flow migration response message is to respond to the flow migration indication message of step 1613, and the name of the message may be another name as long as the message completes the above functions.

 In step 1615, after the stream migration is completed, data communication with CN1 is performed on interface 2.数据 The data message with CN1 is MN MAG2-MN MAG-CNl CNl as shown in this step.

 This example uses the flow migration initiated by the network as an example to describe the flow of the flow selection policy and the address mapping relationship caused by the flow migration. When a multi-access terminal is switched, that is, when the multi-access terminal changes the MAG of an interface due to the movement, or when an interface of the multi-access terminal fails for some reason (such as a failure, etc.) When serving the terminal, all the flows on the interface where the switching/failure occurs need to be switched to other interfaces of the terminal. The flow of this embodiment can still be used. For handover scenarios, steps 1605 and 1613 may be handover request messages.

 There are two ways to modify the flow selection policy of the policy server:

 Update Policy Server Method One:

 As mentioned in Figure 16, step 1607 and step 1608 are performed after the MAPS-MN obtains the modified stream selection policy.

 Update Policy Server Method 2:

After obtaining the modified flow selection policy information, the MAG1-MN or the MAG2-MN sends a policy modification request message to the policy server, where the message carries the MN ID and the modified flow selection policy. The policy server updates the flow selection policy associated with the MN and returns a policy modification response message to replace step 1607 and step 1608 above. There are two methods for modifying the flow selection policy of the terminal as follows:

 Update terminal method one:

 As mentioned in Figure 16, step 1612 and step 1613 are performed after the MAPS-MN determines to modify the stream selection policy.

 Update terminal method two:

 After the new flow selection policy is obtained by the policy server, the flow migration indication message is directly sent to the MAG1-MN or the MAG2-MN to replace the step 1612 in FIG. The message carries the MN identifier (MN ID), the target interface information, and the modified flow selection policy. The optional source interface information and the flow information to be migrated are optional. If sent to MAG1-MN, MAGI-MN can delete the associated stream or connection attribute.

 There are two ways to modify the MAG-CN stored flow selection policy:

 Update MAG-CN Method 1:

 As mentioned in Figure 16, step 1610 and step 1611 are performed after the MAPS-MN decides to perform the stream selection policy modification.

 Update MAG-CN Method 2:

 After the MAG1-MN or the MAG2-MN or the policy server determines or obtains the modified flow selection policy information, the MAG of the opposite end is notified to update the address mapping and the connection attribute to replace the foregoing step 1610. At the same time, after receiving the message, the MAG-CN obtains the MAG2-MN address or CoA2 and the connection attribute in the message, updates the locally stored MN address mapping table, and sets the HNP1-CoAl (or MAG1-MN address)-connection attribute. Change to the HNPl-CoA2 (or MAG2-MN address) country connection attribute. The MAG-CN1 returns a CoA Update Response message to the MAG1-MN or MAG2-MN or the Policy Server to replace the above step 1611.

After the network side decides to change the flow selection policy, it needs to perform policy update on the terminal, MAPS-MN, policy server, and MAG-CN respectively, and perform flow selection on each network element (terminal, MAPS-MN, policy server, and MAG-CN). The policy update has the above multiple manners. In the update, various modes can be combined in any way to complete the information modification of the relevant network element, and no longer, for example, repeated description. In addition, if the policy related to the MN is not stored on the policy server, the operation of the flow selection policy update may not be performed on the policy server. Example twelve

 When the network side (MAPS-MN, policy server, MAG1-MN, or MAG2-MN) is based on load sharing, signal strength, user willingness, network planning, or operation and maintenance, etc., after deciding to change the flow selection policy, the terminal may be notified to perform the flow. Migration (If the policy server or MAPS-MN decides to change the flow selection policy, the notification message can be sent to the MN via MAGI-MN or MAG2-MN, carrying the updated flow selection policy), and the terminal decides whether to initiate the policy modification process. If the terminal initiates a policy update, the operation is similar to the processing of FIG. 6, and the description will not be repeated here.

 In the foregoing embodiment of the present invention, the MAPS/MAG-MN/policy server notifies the peer MAG to update the address mapping table, that is, the update location relationship (CoA update procedure), and the connection attribute field is carried in the CoA update message. When the flow selection policy is stored on the peer MAG, that is, when the address mapping table in the peer MAG is a HNP-CoA-flow selection policy, the modified flow selection policy needs to be carried in the CoA update message instead of the connection. Attributes.

 In the process of all the foregoing embodiments of the present invention, the policy update process of the terminal, the MAPS policy, the connection attribute, the MN registration information table update process, the policy server policy update process, the address mapping of the peer MAG, and the policy update are all described. Process, but the update process of the four network elements does not need to be implemented in the process. For different scenario problems, the update process of the four network elements may have different combinations.

Combination method 1: update the terminal and the peer MAG, or update the terminal, the peer MAG and the MAPS. When the terminal/MAPS/MAG only needs to initiate the migration or handover of the current flow for some reason, and the subsequent newly initiated flow still selects the path according to the existing flow selection policy (ie, selects the location information), all the above processes can still be used. In this scenario, it is no longer necessary to modify the flow selection policy process on the Policy Server. In this way, the flow that the MN has established can select the path or location information according to the modified flow selection policy, and the new flow created by the MN will select a different path because of the original policy. For example, if the original path of stream 1 of the MN is MN MAG1-MN o MAG-CN1 CN1 , and the MN/network side decides to migrate the stream to MAG2-MN for some reason, the path of stream 1 becomes MN MAG2-MN MAG. -CNl CNl, because tampering with the information of the MN and the peer MAG, stream 1 can maintain the path, and since the policy server is not modified, the MN newly created stream 2 according to the policy on the policy server, the selected path will still be By MAG1-MN, not MAG2-MN. Combination 2: Update the terminal and policy server, or update the terminal, policy server and MAPS. When the combination mode is used, the peer MAG needs to correctly select the location/MAG of the multiple access terminal when forwarding the data packet sent by the peer end, and the peer MAG needs to send the data packet to the MAPS each time. - MN queries the location information of the MN. For the traditional PMIP system, this combination is more optimized because there is no need to update the MAG of the communication peer.

 Combination mode 3: Update the terminal, policy server and peer MAG, or update the terminal, policy server, peer MAG and MAPS.

 Obviously, those skilled in the art should understand that the above modules or steps of the present invention can be implemented by a general-purpose computing device, which can be concentrated on a single computing device or distributed over a network composed of multiple computing devices. Alternatively, they may be implemented by program code executable by the computing device so that they may be stored in the storage device by the computing device, or they may be separately fabricated into individual integrated circuit modules, or Multiple modules or steps are made into a single integrated circuit module. Thus, the invention is not limited to any particular combination of hardware and software.

 The above description is only the preferred embodiment of the present invention, and is not intended to limit the present invention, and various modifications and changes can be made to the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and scope of the present invention are intended to be included within the scope of the present invention.

Industrial and practical

 The embodiment of the invention provides a method and a system for managing a policy in a flow movement. When there are multiple access gateways, the peer end learns the correct location information of the terminal to implement message transmission.

Claims

Claim  1. A method for managing a policy in a flow movement, comprising:  When the terminal initiates a flow migration or handover, the terminal sends the modified flow selection policy to the network side;  And the network side sends the modified flow selection policy to the policy server; or, when the flow server selects a flow selection policy based on the terminal, the network side uses the modified flow selection policy Sent to the policy server.  2. The method of claim 1, wherein the method further comprises  After the policy server receives the modified flow selection policy, when the flow server selects a flow selection policy based on the terminal, the policy server updates and the modified flow selection policy according to the modified a terminal-related flow selection policy; when there is no flow selection policy based on the terminal on the policy server, the policy server uses the modified flow selection policy as a new flow selection policy of the terminal.  3. The method of claim 1, wherein  The terminal sends the modified flow selection policy to the network side, and the network side sends the modified flow selection policy to the policy server, including:  The terminal sends a flow migration indication message to the source access gateway or the target access gateway of the migrating flow of the terminal, where the flow migration indication message carries the modified flow selection policy;  Transmitting, by the source access gateway or the target access gateway, the flow migration indication message to a location server of the terminal;  The location server sends a policy modification request message to the policy server, and carries the modified flow selection policy.  The method of claim 1, wherein the terminal sends the modified flow selection policy to the network side, and the network side sends the modified flow selection policy to the policy server, including: the terminal Sending a flow migration indication message to the source access gateway or the target access gateway of the migrating flow of the terminal, where the flow migration indication message carries the modified flow selection policy; The source access gateway or the target access gateway sends a policy modification request message to the policy server, Carrying the modified stream selection policy.  5. The method of claim 4, wherein the method further comprises:  The source access gateway or the target access gateway sends a flow migration indication message to the location server of the terminal, where the flow migration indication message carries the modified flow selection policy;  The location server updates the local flow selection policy associated with the terminal according to the modified flow selection policy.  6. The method of claim 3, wherein the method further comprises:  After receiving the modified flow selection policy, the location server updates the local flow selection policy related to the terminal according to the modified flow selection policy.  The method according to claim 3, 5 or 6, wherein the method further comprises: when the terminal registers with the location server, the location server acquires connection information or a flow selection policy from the policy server, and establishes the method a registration information table of the terminal, the registration information table of the terminal includes an identity of the terminal, a mapping relationship between the interface information currently accessed by the terminal, and a connection attribute; the location server returns the location server to the access gateway And sending, by the access gateway, the identity identification information of the terminal, or sending the identity identification information of the terminal and the location information allocated by the access gateway to the terminal to The terminal.  8. The method according to claim 7, wherein the method further comprises:  After receiving the modified flow selection policy, the location server modifies the registration information table of the terminal, and modifies the mapping relationship between the terminal identifier, the source interface information and the connection attribute of the migration flow to the target identifier and the destination of the migration flow. The mapping between interface information and connection attributes.  A flow management system, including a network side and a policy server, wherein: the network side is configured to: when the terminal initiates a flow migration or handover, receiving a modified flow selection sent by the terminal Strategy; and, Sending the modified flow selection policy to the policy server; or sending the modified flow selection policy to the policy server when there is a flow selection policy based on the terminal on the policy server. 10. The system of claim 9, wherein  The policy server is further configured to: after receiving the modified flow selection policy, when there is a flow selection policy based on the terminal on the policy server, update and describe the modified flow selection policy according to the a terminal-related flow selection policy; when the flow selection policy based on the terminal does not exist on the policy server, the modified flow selection policy is used as a new flow selection policy of the terminal.  11. The system of claim 9, wherein  The network side includes a source access gateway or a target access gateway, and a location server, where: the source access gateway or the target access gateway is configured to: receive a flow migration indication message sent by the terminal, and the flow migration indication Transmitting, by the message, the modified flow selection policy, and forwarding the flow migration indication message to a location server of the terminal;  The location server is configured to: send a policy modification request message to the policy server, and carry the modified flow selection policy.  The system of claim 9, wherein the network side comprises a source access gateway or a target access gateway, where:  The source access gateway or the target access gateway is configured to: receive a flow migration indication message sent by the terminal, where the flow migration indication message carries the modified flow selection policy; and send a policy modification request message to the policy server Carrying the modified stream selection policy.  13. The system of claim 12, wherein the system further comprises a location server, wherein:  The source access gateway or the target access gateway is further configured to: send a flow migration indication message to the location server of the terminal, where the flow migration indication message carries the modified flow selection policy; And updating a local flow selection policy related to the terminal according to the modified flow selection policy.  14. The system of claim 11 wherein: The location server is further configured to: after receiving the modified flow selection policy, update the local flow selection policy related to the terminal according to the modified flow selection policy. 15. The system of claim 11, 13 or 14, wherein  The location server is further configured to: when the terminal registers with the access gateway, obtain a connection information or a flow selection policy from the policy server, and establish a registration information table of the terminal, where the registration information table of the terminal includes the a terminal identifier of the terminal, a mapping relationship between the interface information currently accessed by the terminal, and a connection attribute; and returning, to the access gateway, identity information that is allocated by the location server to the terminal, where the access gateway The identity information of the terminal is sent, or the identity information of the terminal and the location information allocated by the access gateway to the terminal are sent to the terminal.  16. The system of claim 15 wherein  The location server is further configured to: after receiving the modified flow selection policy, modify the registration information table of the terminal, and modify the mapping relationship between the terminal identity identifier, the source interface information of the migration flow, and the connection attribute to the terminal identity. The mapping relationship between the target interface information of the migration stream and the connection attribute.