CN1592254A - Method for bearing step selecting route in controlling layer - Google Patents

Abstract

The invention discloses a method for hop-by-hop routing selection in the bearer control layer, which mainly includes: when each bearer network resource manager in the bearer control layer only knows the topology structure of the management domain under its jurisdiction, or When each connection node only knows its own adjacent connection node, the bearer network resource manager or connection node only selects and determines the next-hop bearer network resource manager or the next-hop connection node. The method provided by the invention can make full use of the resources on each bearer network resource manager, rationally use the network resources, ensure the reliable establishment of routes, and improve the success rate of route selection.

Description

A method of hop-by-hop routing in the bearer control layer

technical field

The present invention relates to route selection technology, in particular to a method for hop-by-hop route selection in the bearer control layer.

Background technique

With the continuous expansion of the scale of the Internet, various network services emerge one after another, and advanced multimedia systems emerge in endlessly. Since real-time services are sensitive to characteristics such as network transmission delay and delay jitter, when there are services such as file transfer (FTP) or hypertext transfer (HTTP) containing image files on the network, the real-time service will fail. In addition, because the multimedia service will occupy a large amount of bandwidth, it will also make it difficult to obtain reliable transmission of key services that need to be guaranteed in the existing network. Therefore, in order to ensure the reliable transmission of key services, various Quality of Service (QoS, Quality of Service) technologies have emerged as the times require. The Internet Engineering Task Force (IETF, Internet Engineering Task Force) has proposed many service models and mechanisms to meet the requirements of QoS. At present, the industry generally recognizes that the integrated service (Int-Serv, Integrated Service) model is used at the access or edge of the network, and the differentiated service (Diff-serv, Differentiated Service) model is used at the core of the network.

The Diff-serv model guarantees QoS only by setting priority levels. Although this model has the characteristics of high line utilization, the specific effect is difficult to predict. Therefore, the industry has introduced an independent bearer control layer for the Diff-Serv model of the backbone network, established a set of special Diff-Serv QoS signaling mechanisms, and established a resource management layer for the Diff-Serv network to manage the network topology resources, this resource management Diff-Serv method is called the Diff-Serv model with an independent bearer control layer. Fig. 1 is the schematic diagram of this model, wherein, 101 is business server, belongs to service control layer, can realize functions such as soft switch; 102 is bearer network resource manager, belongs to bearer control layer; 103 is edge router (ER, Edge Router) , 104 is a core router, 105 is a border router (BR, Border Router); ER, core router, and BR all belong to the bearer network, and are collectively referred to as connection nodes (CN, Connection Node). In this model, the bearer network resource manager is responsible for configuring management rules and network topology, and allocating resources for customers' service bandwidth applications. The bearer network resource managers of each management domain transmit the customer's service bandwidth application request and result, and the path information allocated by each bearer network resource manager for the service application through signaling. When the bearer control layer processes the user's service bandwidth application, it will determine the path of the user's business, and the bearer network resource manager will notify the ER to forward the service flow according to the specified path.

The routes in the bearer network resource manager include signaling route and service route. Signaling route refers to the process of how each bearer network resource manager finds the next-hop bearer network resource manager; The process of how the resource manager finds a suitable bearer Label Switching Path (LSP) according to the service flow information, specifically includes intra-domain routing and inter-domain routing.

Usually, the bearer network forwards user traffic according to the specified route according to the path determined by the bearer control layer. At present, the industry mainly uses the multi-protocol label switching (MPLS) technology to reserve resources along the bearer control layer The specified service flow path establishes an LSP, or uses an explicit routing mechanism based on traffic engineering resource reservation protocol (RSVP-TE) or restricted routing label distribution protocol (CR-LDP) to establish an end-to-end LSP.

At present, there are already many methods for routing in the bearer network, such as the bandwidth proxy model of the service backbone experimental network (QBone, Quality-of-Service backbone). As shown in Figure 2, this model sets a corresponding bandwidth proxy 201 for each Diff-Serv management domain, which is responsible for processing bandwidth application requests from user hosts, business servers or network maintenance personnel, and according to the current The network resource reservation status and configuration strategy, as well as the business service level agreement (SLA, Service Level Agreement) signed with the user, determine whether to approve the user's bandwidth application. The bandwidth manager records a large amount of static or dynamic information such as various SLA configuration information, physical network topology information, router configuration information and policy information, user authentication information, current resource reservation information, and network occupation status information. At the same time, the bandwidth manager also needs to record routing information to establish the user's service flow path and the location of the cross-domain downstream bandwidth manager. In this bandwidth agent model, because the bandwidth manager directly manages the resources and configuration information of all routers in the area, there are problems of too complicated topology and management; at the same time, because the bandwidth manager needs to record the dynamic routing of the area Therefore, the routing table is updated frequently, resulting in unstable network reservation; moreover, the service route determined by the bandwidth manager based on the dynamic routing information in the area is also difficult to be consistent with the actual forwarding route of the service flow. Due to the problems mentioned above in the bandwidth manager model, this model has not been put into commercial use.

In addition, there is also a Rich QoS solution proposed by NEC Corporation. As shown in Figure 3, in this solution, the QoS server is used as a key component, and the solution also includes a policy server, a directory server, and a network management monitoring server that are matched with the QoS server. The policy server sets the parameters and configurations of the relevant routers according to the policy configuration information such as the QoS server and management interface; the directory server is a unified and centralized database for storing network device configuration information, user information and QoS information; and the network management monitors The server is responsible for collecting information such as the congestion status of each router and link in the bearer network, which can be used as a reference for the QoS server to select routes for service applications; while the QoS server is responsible for allocating qualified bearers for QoS service requests according to the topology and resource status of the bearer network path. In practical applications, it is necessary to pre-set the topology and bandwidth of the bearer network on the QoS server, and configure the routing rules. When the service server sends a bandwidth request to the QoS server, the QoS server records the resource request of the call, and allocates a bearer path that satisfies the requirements for the service request according to its QoS requirements, the current topology of the bearer network and the current resource status, and the allocated result Feedback business server. The QoS server can also issue a corresponding LSP policy modification command to the policy server according to the bandwidth occupation of the service, and the policy server configures the corresponding edge router according to the command of the QoS server. In this scheme, the edge router will use the explicit routing technology established by MPLS LSP, and re-establish or adjust the LSP according to the path specified by the QoS server.

In the above-mentioned Rich QoS solution, what the QoS server manages is still a relatively complex bearer network with a relatively large number of routers; at the same time, the QoS server and the policy server notify the edge routers to use the explicit routing MPLS LSP establishment technology to establish The end-to-end LSP model has poor scalability and limited network scale, and cannot meet the end-to-end service requirements of a national public network.

Contents of the invention

In view of this, the main purpose of the present invention is to provide a method of hop-by-hop routing selection for signaling routing in the bearer control layer, thereby improving the success of each bearer network resource manager in the bearer control layer to find the next hop CM. efficiency and ensure the reliability of resource allocation.

Another object of the present invention is to provide a method of hop-by-hop routing selection for service routing in the bearer control layer, thereby improving the success of each bearer network resource manager in the bearer control layer in selecting the LSP for the bearer service connection for the bearer layer efficiency and ensure the reliability of resource allocation.

In order to achieve the above object, technical solution of the present invention is achieved in that way:

A method for performing hop-by-hop signaling routing selection in a bearer control layer, comprising the following steps:

A1. The bearer network resource manager that currently receives the connection resource request message judges whether it is the target bearer network resource manager, and if so, completes the establishment of the signaling routing path and ends the current routing process; otherwise, executes step B1;

B1. The current bearer network resource manager selects the next-hop bearer network resource manager, and sends a connection resource request message to the next-hop bearer network resource manager; return to step A1.

Before the step A1, it further includes:

A01. The bearer network resource manager currently receiving the connection resource request message judges whether the connection between itself and the sender of the request message is available, and if available, execute step A1; otherwise, execute step A02;

A02. The bearer network resource manager that currently receives the connection resource request message returns a connection resource rejection message to the bearer network resource manager that sent the connection resource request message;

A03. The bearer network resource manager currently receiving the connection resource rejection message finds the corresponding connection resource request message received according to the connection resource rejection message; and reselects the next-hop bearer network resource management according to the connection resource request message and then send a connection resource request message to the newly selected next-hop bearer network resource manager; perform step A1.

Before completing the establishment of the signaling routing path in the step A1, it further includes:

A11. The current bearer network resource manager returns a connection resource response message to the bearer network resource manager that sent the request message according to the connection resource request message;

A12. The bearer network resource manager that has received the connection resource response message judges whether it is the initial bearer network resource manager, and if so, completes the establishment of the signaling routing path and ends the current routing process; otherwise, executes step A13;

A13. The current bearer network resource manager finds the corresponding received connection resource request message according to the connection resource response message, determines the previous hop bearer network resource manager, and returns to step A11.

The step A1 includes: the current bearer network resource manager judges whether it is the destination bearer network resource manager according to the destination address in the connection resource request message.

The selection described in step B1 is: the current bearer network resource manager selects the next-hop bearer network resource manager according to the destination address in the received connection resource request message; or the current bearer network resource manager selects the next-hop bearer network resource manager according to the service type and priority , the locally configured routing policy, the current network conditions and the destination address in the received connection resource request message to select the next-hop bearer network resource manager.

A set of searched bearer network resource managers is set, and the step B1 further includes before sending a connection resource request message to the selected next-hop bearer network resource manager: the current bearer network resource manager judges the selected Whether the next-hop bearer network resource manager has been added to the searched bearer network resource manager set; if yes, then give up the selection of the next-hop bearer network resource manager selected above, and return to step B1, Re-select the next-hop bearer network resource manager; otherwise, add its own information to the searched bearer network resource manager set.

The step A01 includes: the current bearer network resource manager judges whether the connection established between itself and the sender of the connection resource request message is available according to the unavailability or failure of its own resources.

The connection resource request message includes a session identifier, a quintuple, a service type, a quality of service parameter, and address information of an egress border router of the selected inter-domain label switching path.

After the signaling routing path is established, the initial bearer network resource manager will issue a flow mapping command to the corresponding ingress edge router in its domain according to the connection resource request message received above.

The flow mapping command includes session identification, flow information, quality of service parameters, flow descriptor and label stack of the entire path.

A method for hop-by-hop selection of a service route in a bearer control layer, comprising the following steps:

A2. The current bearer network resource manager finds the ingress connection node CN in the domain according to the received connection resource request message, and adds the information of the found ingress CN to the searched router set;

B2. The current bearer network resource manager selects an intra-domain label switching path according to the current entry CN;

C2. The current bearer network resource manager judges whether the egress CN of the label switching path in the selected domain is an edge server or a border server in the management domain of the current bearer network resource manager, and if so, completes the establishment of the service routing path and ends the current route selection process; otherwise, execute step D2;

D2. The resource manager of the current bearer network judges whether the current egress CN has been added to the above-mentioned set of searched routers, and if so, abandons the selection of the above-mentioned selected intra-domain label switching path, and returns to step B2; otherwise, proceed with The current egress CN is used as the current ingress CN, and the CN is recorded in the searched router set, and the execution returns to step B2.

The step A2 further includes: the current bearer network resource manager judges whether it is the initial bearer network resource manager, and if so, the current bearer network resource manager according to the connection resource request message received from the call agent The IP address information finds the ingress edge router in the domain as the ingress CN; otherwise, the current bearer network resource manager finds the ingress border router in the domain as the ingress CN according to the received connection resource request message.

The step B2 includes: randomly selecting an intra-domain label switching path according to the current ingress CN.

The step B2 includes: selecting an intra-domain label switching path according to the current ingress CN and according to the user address, or activation of the label switching path, or routing priority, or bandwidth requirement.

The CN is an edge router, or a border router, or a core router.

Before preparing to establish a service routing path, each bearer network resource manager will simulate a routing table for each CN in its domain.

The simulation of a routing table is: the current bearer network resource manager extracts the routing information stored in each CN in its domain to the bearer network resource manager.

The present invention provides a method for hop-by-hop routing selection in a bearer network. The method ensures the reliability of resource allocation through the mechanism of hop-by-hop application and confirmation and the method of rollback and reselection, and improves the reliability of each bearer in the bearer control layer. The success rate of inter-domain and intra-domain routing of the network resource manager. When using this method to select routes, even if some nodes or links in the network fail, the impact of the failure on providing accurate routing information will be reduced due to the mechanism of fallback and reselection in this method, and The hop-by-hop route selection method provided by the present invention does not have any special requirements and assumptions on the network structure, and can be adapted to any topological network, including various complex network situations.

Description of drawings

FIG. 1 is a schematic diagram of an independent bearer control layer network model in the prior art;

Fig. 2 is the schematic diagram of the bandwidth agent model of QBone in the prior art;

Fig. 3 is the schematic structural diagram of IP QoS scheme architecture in the prior art;

FIG. 4 is a flow chart of establishing a signaling routing path in the bearer control layer of the present invention;

FIG. 5 is a schematic diagram of message interaction between multiple bearer network resource managers when establishing a signaling routing path according to the present invention;

FIG. 6 is a schematic diagram of establishing a signaling routing path in the bearer control layer of the present invention;

Fig. 7 is a flow chart of establishing a service routing path in the bearer control layer of the present invention.

Detailed ways

In order to make the object, technical solution and advantages of the present invention more clearly, the present invention will be further described in detail below in conjunction with the accompanying drawings and specific embodiments.

The present invention provides a method for hop-by-hop routing selection in a bearer network, the main idea of which is: when each bearer network resource manager in the bearer control layer only knows the topology of the management domain it governs, Or when each CN only knows its own adjacent CN, the bearer network resource manager or CN only selects and determines the next-hop bearer network resource manager or next-hop CN.

For simplicity, in the following description, the bearer network resource manager is referred to as CM.

Fig. 4 shows the process of using the method for hop-by-hop routing of the present invention to set up a signaling routing path in the bearer control layer:

Step 401: The initial bearer network resource manager receives the connection resource request message sent by the CA;

Step 402: The bearer network resource manager currently receiving the connection resource request message judges whether its connection with the sender of the request message is available, and if available, executes step 403; otherwise, executes step 405;

Step 403: The current bearer network resource manager judges whether it is the destination bearer network resource manager according to the destination address in the connection resource request message, if yes, execute step 407; otherwise, execute step 404;

Step 404: The current bearer network resource manager selects a next-hop bearer network resource manager according to the destination address in the connection resource request message, and sends a connection resource request message to the selected next-hop bearer network resource manager ; Return to step 402;

Step 405: The bearer network resource manager that currently receives the connection resource request message returns a connection resource rejection message to the bearer network resource manager that sent the connection resource request message;

Step 406: The bearer network resource manager currently receiving the connection resource rejection message finds the corresponding previously received connection resource request message according to the connection resource rejection message; execute step 404;

Step 407: The current bearer network resource manager returns a connection resource response message to the bearer network resource manager that sent the request message according to the connection resource request message;

Step 408: The bearer network resource manager that has received the connection resource response message judges whether it is the initial bearer network resource manager, if yes, execute step 410; otherwise, execute step 409;

Step 409: The current bearer network resource manager finds the corresponding received connection resource request message according to the connection resource response message, and returns to step 407;

Step 410: Complete the establishment of the signaling routing path.

After the signaling routing path is established, the initial bearer network resource manager will issue a flow mapping command to the corresponding ingress ER in the domain of the initial bearer network resource manager according to the connection resource request message sent from the CA. The command contains the session identifier , flow information, QoS parameters, flow descriptor, and the label stack of the entire path.

The above-mentioned mechanism in which the bearer network resource manager sends connection resource response messages to its previous hop bearer network resource manager step by step can be called a confirmation mechanism. The previous hop bearer network resource manager sends a connection resource rejection message, and the previous hop bearer network resource manager reselects the next hop bearer network resource manager as a fallback mechanism. Adding the above-mentioned fallback mechanism and confirmation mechanism to the routing method can ensure the reliable establishment of the connection to the greatest extent possible and save network resources. Of course, the above-mentioned fallback cannot be unlimited fallback, and the number of fallbacks or a certain time limit can be set in advance to ensure the query efficiency hop by hop. For example, if the number of rollbacks exceeds the set number of rollbacks, or when a bearer network resource manager receives a connection resource rejection message, it finds that the time between the rejection message and the corresponding connection resource request message If the interval exceeds the preset time limit, it is considered that the routing selection failed this time, and the next routing selection is restarted; or the rollback flag is used, which only allows one rollback.

In addition, in practical applications, some networks, such as mesh networks, have complex structures. In the above route selection process, in order to prevent the selected bearer network resource manager The next-hop bearer network resource manager of a certain bearer network resource manager in the path, so to set a searched bearer network resource manager set, then in step 404, it should also include: the current bearer network resource manager in the direction Before the selected next-hop bearer network resource manager sends a connection resource request message, it should also determine whether the selected next-hop bearer network resource manager has been added to the above-mentioned searched bearer network resource manager set; if If yes, give up the selection of the selected next-hop bearer network resource manager, return to step 404, and re-select the next-hop bearer network resource manager; otherwise, add its own information to the searched bearer network resource manager In the resource manager set, send a connection resource request message to the selected next-hop bearer network resource manager.

FIG. 5 is a schematic diagram of message interaction between multiple bearer network resource managers according to the present invention. In this embodiment, the bearer network resource managers are referred to as CMs. In Figure 5, CM1 is the initial CM, CMn is the destination CM, and other CMs are intermediate CMs; the specific process is as follows:

When the initial CM, that is, CM1, receives the connection resource request message sent by the CA, it judges that the connection between itself and the sender of the request message, that is, CA, is available according to its own situation, and then judges that it is not the destination CM according to the request message. Then select the next-hop CM according to the service type, resource availability, priority, locally configured routing strategy, and the destination address in the received connection resource request message, and send the selected next-hop CM to the selected next-hop CM, namely CM2 sends a connection resource request message; after CM2 receives the connection resource request message sent by CM1, it judges itself and the sender of the request message according to its own situation, such as resource unavailability, self failure or other reasons, that is, the connection with CM1 is unavailable , then return a connection resource rejection message to CM1; CM1 reselects its next-hop CM according to the connection resource rejection message returned by CM2, and sends a connection resource request message to the re-selected next-hop CM, namely CM3; By analogy, after receiving the connection resource request message sent by its previous CM, each intermediate CM will judge whether its connection with the previous CM is available according to its own situation. Judging that the connection between itself and the previous hop CM is unavailable, it returns a connection resource rejection message to its previous hop CM; its previous hop CM reselects the next hop CM according to the returned connection resource rejection message, and sends the message to the previous hop CM. The re-selected next hop CM sends a connection resource request message; if an intermediate CM receives the connection resource request message sent by the previous hop CM, it judges the connection between itself and the previous hop CM according to its own situation. If it is available, and it is judged that it is not the destination CM according to the request message, it will select the next CM according to the service type, resource availability, priority, locally configured routing strategy, and the destination address in the received connection resource request message. Hop CM, and send a connection resource request message to the selected next hop CM; when the destination CM, that is, CMn receives the connection resource request message sent by the previous hop CM, it judges itself and the previous hop CM according to its own situation. The connection of the CMn is available, and then according to the request message, when judging that it is the destination CM, it returns a connection resource response message to the CM that sent the request message, that is, the last hop CM of CMn; The above-mentioned connection resource response message judges that it is not the initial CM, and then finds the corresponding connection resource request message once received according to the connection resource response message returned by CMn, and then sends the connection resource request message to the CM that sends the connection resource request message, that is, the upper CM of the current CM. One hop CM returns a connection resource response message; and so on, each intermediate CM returns a connection resource response message to its own previous hop CM according to the above method; when the initial CM, that is, CM1 receives its next hop CM return After receiving the connection resource response message, it will find the corresponding received connection resource request message according to the response message, and then return a connection resource response message to the CA that sent the connection resource request message, thus finally completing the entire signaling route Path establishment.

It can be seen from Figure 5 that each CM will give a connection resource response message or a connection resource rejection message for each connection resource request message it receives, and each CM will only send a connection resource response message or a connection resource rejection message when it receives the next-hop CM message. After receiving the connection resource response message or the connection resource rejection message, the connection resource response message or the connection resource rejection message will be sent to its previous hop CM. Through this confirmation mechanism and rollback mechanism, the reliability of resource allocation can be guaranteed, and resource waste can also be avoided.

FIG. 6 is a schematic diagram of establishing a signaling routing path by the bearer control layer of the present invention. In the present invention, the method for establishing a signaling routing path is as follows:

When the bearer network resource manager 601 receives the connection resource request message sent by the CA, the bearer network resource manager 601 judges that it is not the destination bearer network resource manager according to the request message, so the bearer network resource manager 601 The destination address in the message and the topology of other bearer network resource managers adjacent to bearer network resource manager 601 select all bearer network resource managers that can be used as their next-hop bearer network resource managers, as shown in Figure 6 Shown: the bearer network resource manager adjacent to bearer network resource manager 601 has bearer network resource manager 602 and bearer network resource manager 604, then bearer network resource manager 601 can combine bearer network resource manager 602 and bearer network resource The manager 604 serves as an optional next-hop bearer network resource manager, and then the bearer network resource manager 601 selects from the bearer network resource manager In 602 and the bearer network resource manager 604, a next-hop bearer network resource manager is selected. In this embodiment, assuming that the next-hop bearer network resource manager selected by the bearer network resource manager 601 is the bearer network resource manager 602, then the bearer network resource manager 601 sends a connection resource to the bearer network resource manager 602 A request message, which includes information such as the session identifier, quintuple, service type, QOS parameters, and the address of the egress BR of the selected inter-domain LSP, for the bearer network resource manager 602 to perform intra-domain routing; similarly , the bearer network resource manager 602 judges that it is not the destination bearer network resource manager according to the connection resource request message sent by the bearer network resource manager 601, so the bearer network resource manager 602 also follows the same rule as above to select the All bearer network resource managers of the next-hop bearer network resource manager, as shown in Figure 6, are bearer network resource manager 603 and bearer network resource manager 605, and then from bearer network resource manager 603 and bearer network The manager 605 selects the next-hop bearer network resource manager. In this embodiment, assume that the bearer network resource manager 602 selects the bearer network resource manager 603 as the next-hop bearer network resource manager, and sends the bearer network resource manager The manager 603 sends a connection resource request message; the bearer network resource manager 603 judges itself to be the target bearer network resource manager according to the connection resource request message sent by the bearer network resource manager 602, and returns a message to the bearer network resource manager 602. Connection resource response message; similarly, bearer network resource manager 602 returns a connection resource response message to bearer network resource manager 601; bearer network resource manager 601 returns a connection resource response message to CA, and finally completes the entire signaling routing The establishment of the path; then the bearer network resource manager 601 will issue a flow mapping command to the entrance ER606 in its domain according to the connection resource request message received above. The label stack for the entire path.

During the above process, if a certain bearer network resource manager in the signaling routing path finds that the connection between it and the previous hop bearer network resource manager is unavailable When used, the bearer network resource manager will send a connection resource rejection message to its last-hop bearer network resource manager, so that the last-hop bearer network resource manager can select another bearer network resource manager from other optional routing paths as the connection resource manager. The next hop bearer network resource manager to complete the connection. For example, in the above-mentioned embodiment, if the bearer network resource manager 602 finds that it has a failure or finds that the resources on the inter-domain LSP between itself and the bearer network resource manager 601 have been used up, Then the bearer network resource manager 602 will send a connection resource rejection message to the bearer network resource manager 601, and the bearer network resource manager 601 can reselect its next-hop bearer network resource manager according to the rejection message. , the bearer network resource manager 601 can select the bearer network resource manager 604 as the next-hop bearer network resource manager; The routing path of the device 603 completes the establishment of the signaling routing path.

The above-mentioned mechanism in which the bearer network resource manager sends connection resource rejection messages to its previous hop bearer network resource manager step by step is the fallback mechanism, through which the establishment of the connection can be guaranteed to the greatest extent possible. In practical applications, the backoff can be controlled by presetting the number of backoffs or the preset time, so as to ensure the query efficiency hop by hop.

The above is the process of establishing a signaling routing path using the hop-by-hop routing method of the present invention. The principle and mechanism of the method are also applicable to a certain bearer network resource manager when performing routing selection in the domain under its jurisdiction. , but in terms of specific implementation, it is no longer the message interaction between each bearer network resource manager, but each bearer network resource manager independently completes the route selection in its domain. In the management domain of the bearer network resource manager, each CN stores its own routing table, so in the method provided by the present invention, each bearer network resource manager simulates a routing table for each CN in its domain , that is, the bearer network resource manager extracts the routing information stored in the CN itself to the bearer network resource manager, and the bearer network resource manager selects intra-domain routes according to the routing table of each CN. When performing intra-domain routing selection, each bearer network resource manager can use the method of hop-by-hop routing selection to select routes between CNs, and select the LSP path set that bears the service flow.

Fig. 7 shows the flow chart of establishing service routing path in the bearer control layer in the present invention, and the specific steps are as follows:

Step 701: The bearer network resource manager currently performing intra-domain route selection determines whether it is the initial bearer network resource manager, and if so, finds an entry in the domain according to the IP address information in the connection resource request message received from the CA ER, as the entrance CN, and add the information of the entrance ER to the set of searched routers; otherwise, the current bearer network resource manager finds the entrance BR in this domain according to the information in the connection resource request message received, and serves as the entrance CN, and add the information of the ingress ER to the set of searched routers; wherein, any intra-domain LSP includes an ingress CN and an egress CN;

Step 702: Select an intra-domain LSP according to the current ingress CN; when selecting the intra-domain LSP, it can be selected randomly, or can be selected according to conditions such as user address, LSP activation status, routing priority, and bandwidth requirements;

Step 703: Determine whether the router at the other end of the selected LSP, that is, the egress CN, is the BR or ER in the management domain of the current bearer network resource manager. If yes, it means that the LSP in this domain has been selected successfully, and execute step 706; otherwise, execute step 706. 704;

Step 704: Determine whether the above-mentioned egress CN has been added to the above-mentioned set of searched routers, if yes, then give up the selection of the above-mentioned selected LSP, return to step 702, and re-select the LSP; otherwise, go to step 705;

Step 705: Use the egress CN described in step 703 as the ingress CN, return to step 702;

Step 706: Complete the establishment of the service routing path, and end the current routing process.

In summary, when selecting signaling routing paths between domains and selecting service routes within domains, the hop-by-hop routing method of the present invention can be used, or the hop-by-hop routing method of the present invention can be used when signaling routing is selected. A method for selecting a route, and use other algorithms when selecting a service route; or, use the method for selecting a route hop-by-hop according to the present invention when selecting a service route, and use another algorithm when selecting a signaling route; or The intra-domain routing of service routing uses this algorithm, and the inter-domain routing uses other algorithms; vice versa. The specific selection scheme can be determined according to network conditions and business requirements.

The above descriptions are only preferred embodiments of the present invention, and are not intended to limit the protection scope of the present invention. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principles of the present invention shall be included within the protection scope of the present invention.

Claims (17)

1. A method for carrying out signaling routing hop-by-hop selection in the bearer control layer, characterized in that it comprises the following steps: A1. The bearer network resource manager that currently receives the connection resource request message judges whether it is the target bearer network resource manager, and if so, completes the establishment of the signaling routing path and ends the current routing process; otherwise, executes step B1; B1. The current bearer network resource manager selects the next-hop bearer network resource manager, and sends a connection resource request message to the next-hop bearer network resource manager; return to step A1. 2. The method according to claim 1, characterized in that before the step A1, it further comprises: A01. The bearer network resource manager currently receiving the connection resource request message judges whether the connection between itself and the sender of the request message is available, and if available, execute step A1; otherwise, execute step A02; A02. The bearer network resource manager that currently receives the connection resource request message returns a connection resource rejection message to the bearer network resource manager that sent the connection resource request message; A03. The bearer network resource manager currently receiving the connection resource rejection message finds the corresponding connection resource request message received according to the connection resource rejection message; and reselects the next-hop bearer network resource management according to the connection resource request message and then send a connection resource request message to the newly selected next-hop bearer network resource manager; perform step A1. 3. The method according to claim 1, characterized in that, before completing the establishment of the signaling routing path in the step A1, further comprising: A11. The current bearer network resource manager returns a connection resource response message to the bearer network resource manager that sent the request message according to the connection resource request message; A12. The bearer network resource manager that has received the connection resource response message judges whether it is the initial bearer network resource manager, and if so, completes the establishment of the signaling routing path and ends the current routing process; otherwise, executes step A13; A13. The current bearer network resource manager finds the corresponding received connection resource request message according to the connection resource response message, determines the previous hop bearer network resource manager, and returns to step A11. 4. The method according to claim 1, wherein the step A1 comprises: the current bearer network resource manager judges whether it is the target bearer network resource manager according to the destination address in the connection resource request message. 5. The method according to claim 1, wherein the selection in step B1 is: the current bearer network resource manager selects the next-hop bearer network resource manager according to the destination address in the received connection resource request message or the current bearer network resource manager selects the next-hop bearer network resource manager according to the service type, priority, locally configured routing strategy, current network conditions, and the destination address in the received connection resource request message. 6. The method according to claim 1, wherein a set of searched bearer network resource managers is set, and before the step B1 sends a connection resource request message to the selected next-hop bearer network resource manager It further includes: the current bearer network resource manager judges whether the selected next-hop bearer network resource manager has been added to the set of searched bearer network resource managers; For the selection of a one-hop bearer network resource manager, return to step B1 and reselect the next-hop bearer network resource manager; otherwise, add its own information to the set of searched bearer network resource managers. 7. The method according to claim 2, wherein the step A01 comprises: the current resource manager of the bearer network judges whether the resource manager of the current bearer network is compatible with the sender of the connection resource request message according to the unavailability of its own resources or its failure. Whether the established connection is available. 8. The method according to claim 1, wherein the connection resource request message includes session identifier, quintuple, service type, quality of service parameters and the address of the egress border router of the selected inter-domain label switched path information. 9. The method according to claim 1, characterized in that: after the signaling routing path is established, the initial bearer network resource manager will send the flow to the corresponding ingress edge router in its domain according to the connection resource request message received Map command. 10. The method according to claim 9, wherein the flow mapping command includes session identifier, flow information, quality of service parameters, flow descriptor and label stack of the whole path. 11. A method for performing hop-by-hop selection of service routes in a bearer control layer, characterized in that it comprises the following steps: A2. The current bearer network resource manager finds the ingress connection node CN in the domain according to the received connection resource request message, and adds the information of the found ingress CN to the searched router set; B2. The current bearer network resource manager selects an intra-domain label switching path according to the current entry CN; C2. The current bearer network resource manager judges whether the egress CN of the label switching path in the selected domain is an edge server or a border server in the management domain of the current bearer network resource manager, and if so, completes the establishment of the service routing path and ends the current route selection process; otherwise, execute step D2; D2. The resource manager of the current bearer network judges whether the current egress CN has been added to the above-mentioned set of searched routers, and if so, abandons the selection of the above-mentioned selected intra-domain label switching path, and returns to step B2; otherwise, proceed with The current egress CN is used as the current ingress CN, and the CN is recorded in the searched router set, and the execution returns to step B2. 12. The method according to claim 11, wherein the step A2 further comprises: the current bearer network resource manager judges whether it is the initial bearer network resource manager, and if so, the current bearer network resource manager According to the IP address information in the connection resource request message received from the call agent, the manager finds the ingress edge router in the domain as the ingress CN; Find the ingress border router as the ingress CN. 13. The method according to claim 11, wherein the step B2 includes: randomly selecting an intra-domain label switching path according to the current ingress CN. 14. The method according to claim 11, wherein the step B2 includes: selecting an in-domain network according to the current ingress CN, user address, or label switched path activation, or routing priority, or bandwidth requirements. Label switching path. 15. The method according to claim 11, characterized in that: said CN is an edge router, or a border router, or a core router. 16. The method according to claim 11, characterized in that each bearer network resource manager simulates a routing table for each CN in its domain before preparing to establish a service routing path. 17. The method according to claim 16, wherein said simulating a routing table is: the current bearer network resource manager extracts the routing information stored in each CN in its domain to the bearer network resource manager device.

Images