WO2007076661A1

WO2007076661A1 - A congestion controlling method and network device for real time multicasting service.

Info

Publication number: WO2007076661A1
Application number: PCT/CN2006/002065
Authority: WO
Inventors: Hui Liu; Hongke Zhang; Hui Zhang; Daofei Zhu; Jianfeng Guan
Original assignee: Huawei Technologies Co., Ltd.; Beijing Jiaotong University
Priority date: 2005-12-31
Filing date: 2006-08-15
Publication date: 2007-07-12
Also published as: CN1996933A; CN1996933B

Abstract

A congestion controlling method and network device for real time multicasting service. The method includes: each multicasting transmit node forecasts the congestion of itself; when the multicasting transmit node confirms that congestion will appear, it carries out congestion control for real time multicasting service according to the priority of all active multicasting route which had passed itself. The present invention can assure high quality service for real time multicasting service with different levels. The adjustment of active multicasting route is carried out before the congestion status appears, so the smooth transition of active multicasting route is realized, delay and rate of lost packet can be reduced, and the congestion control has better robustness. The present invention can realize load leveling and resource optimization of network, thereby increase the performance of MANET network. In addition present invention does not adopt complicated congestion algorithm so it reduces the overhead and avoids the wasting of MANET network resources.

Description

Method and network device for congestion control of real-time multicast service

The present invention relates to the field of communications, and in particular, to congestion control and network equipment for multicast services. Background of the invention

MANET (Mobile Ad hoc Network) is a distributed wireless mobile multi-hop network that is a core component of the B3G (Beyond 3G) network. Since the bandwidth of each terminal in the MANET network is limited, and the multicast technology can effectively save bandwidth, the application of multicast technology in the network environment of the MANET, including instant messaging such as streaming media, video, multipoint communication, etc., will have Very broad application space.

MANET has high link error rate, limited bandwidth, limited energy, multi-hop, and highly dynamic changes in the topology. Therefore, communication nodes in MANET are prone to congestion. The appearance of congestion indicates that the bandwidth resources of the congested node have been Exhausted, no longer able to withstand more load, nor a fair resource allocation for each active route. Therefore, the emergence of congestion will increase the delay of data packets and even the loss of data packets. If the network resources are always in a congested state, it may cause the final collapse of the network, which will inevitably have a greater impact on the performance of the MANET network. Therefore, the corresponding congestion control strategy must be adopted.

Traditional congestion control strategies use TCP (Transmission Control Protocol) congestion control algorithm and queue scheduling algorithm. The congestion control algorithm is divided into two different mechanisms: congestion avoidance and congestion control. Congestion avoidance is a "prevention" mechanism. Its goal is to prevent the network from entering a congested state and to operate the network in a high-throughput, low-latency state. Congestion control is a "recovery" mechanism that restores a congested node to a normal state. The queue scheduling algorithm performs policy scheduling on the data packets in the sending queue in the case of congestion, including policies such as discarding and prioritizing forwarding. The traditional congestion control strategy can solve the network congestion to a certain extent and improve the performance of network transmission. However, multicast packets are generally encapsulated in UDP (User Datagram Protocol). UDP is a Best-Effort connectionless service protocol, so there is no built-in congestion mechanism to prevent the multicast stream from exhausting the link bandwidth or Its key resources. Therefore, in a MANET network where the pure multicast service or the multicast service is dominant, the traditional congestion control strategy is difficult to meet the real-time multicast service sensitive to delay and packet loss rate.

The prior art related to the present invention is a congestion control scheme applicable to the multi-rate multicast service of the MANET, and the core thereof is: in the MANET, multiple multicast layers are used to forward multicast data, and in the multicast forwarding structure A strategy for implementing congestion control at the congestion bottleneck. When performing congestion control, it uses the traffic information of the packets flowing through the node to compare with a predetermined threshold to prevent or release high-level data of some multicast services, rather than joining and exiting layers. In order to distribute the bandwidth fairly with TCP, it adjusts the bandwidth occupied by multicast traffic and TCP traffic at the multicast bottleneck to achieve fairness. In order to achieve the stability of multicast traffic, a rate control factor is introduced during the interruption and recovery of the multicast session to achieve a smooth transition of the rate.

It can be seen from the technical solution of the prior art that it has the following defects:

The prior art adopts a layered multicast technology in MANET, which increases the complexity of the congestion algorithm and increases the resource consumption of the node.

The prior art 2 related to the present invention is a method for congestion control of a multicast service in a MANET network. The core is: Keep a list of record receiver IPs at the multicast source node to record the nodes that are congested (that is, the nodes that reply NACK to the multicast source). When the list is not empty, the multicast source node enters a congestion control state, and the multicast source node performs a control policy until the list is empty.

It can be seen from the technical solution of the prior art 2 that it has the following defects:

The scheme is post-reaction congestion control, that is, the multicast source node determines whether congestion occurs according to the unreceived condition in the ACK acknowledgement message, and once it is confirmed to be congested, the packet has been delayed and lost at this time, Real-time multicast services have a large impact. Summary of the invention

The object of the present invention is to provide a method and a network device for performing congestion control on a real-time multicast service, and the present invention can ensure real-time multicast of different levels (high level and low level). The service can obtain high-quality services; and realizes a smooth transition of active multicast routing, minimizes delay and packet loss rate, and makes congestion control have better robustness. Furthermore, the present invention can implement network Load balancing and resource optimization improve the performance of the MANET network. In addition, the present invention does not use a complicated congestion algorithm, reduces overhead, and avoids the consumption of valuable resources of the MANET network.

The object of the invention is achieved by the following technical solutions:

The present invention provides a method for congestion control of a real-time multicast service, which includes -

A. Each multicast forwarding node predicts congestion of its own node;

B. When the multicast forwarding node confirms that it is about to be congested, it establishes a new multicast forwarding structure that does not pass its own node according to the priority of all active multicast routes through its own node, and the upstream node according to the new The multicast forwarding structure forwards multicast services to perform congestion control on real-time multicast services.

The step A specifically includes:

Al, each multicast forwarding node performs congestion prediction by detecting the resource utilization of its own node; and/or,

A2. Each multicast forwarding node performs congestion prediction by detecting the energy of its own node. The step A1 specifically includes:

Each multicast forwarding node detects whether the bandwidth utilization of the own node reaches a congestion threshold, and if so, confirms that the node is about to be congested; otherwise, confirms that the node is in a normal working state.

The step A2 specifically includes:

Each multicast forwarding node detects whether the energy of the own node falls to a security threshold, and if so, confirms that the node is about to be congested; otherwise, the node is confirmed to be in a normal working state.

The step B specifically includes:

Bl, when the multicast forwarding node confirms that it is about to be congested, selects the active multicast route with the lowest priority from all active multicast routes through its own node, and routes the downstream node to the selected active multicast group. Send a congestion message; After receiving the congestion packet, the downstream node sends a multicast group join request to the neighbor node according to the multicast group ID information carried in the congestion packet, and receives the response of the neighbor node. After the message, it is confirmed that it has successfully grafted to the forwarding structure of the multicast group, and then sends the information that the route dynamic adjustment succeeds to the congestion node;

B3, the fast-moving node sends the multicast group route invalidation message to its upstream node;

B4. The upstream node determines the multicast group routing information of the newly established multicast forwarding structure according to the received multicast group routing invalidation message, and uses the routing information to forward the corresponding multicast service.

In the step B1:

If there are multiple active multicast routes with the lowest priority, select the active multicast route with a small bandwidth.

The step B2 further includes: after receiving the multicast group join request message, the neighboring node determines whether it meets the condition of joining the multicast group, and if yes, sends a response message to the downstream node; otherwise, discards The multicast group joins the request message.

The conditions for joining the multicast group include:

After the multicast forwarding node establishes a multicast route, it will not cause congestion to occur, and the route to the upstream node of the node to be congested will not pass through the node that is congested, nor will it pass through the node that is about to be congested.

The step B further includes:

If a new route request discovery message arrives during the process of performing congestion control, the near-congested node determines whether it is the forwarding node of the multicast routing request message, and if so, discards; otherwise, receives the discovery message. And join the multicast group.

Before the step A, the method further includes:

Cl, assigning a corresponding priority to the nodes according to the importance of each node joined to the mobile ad hoc network MANET;

C2. Determine, according to the priority of the multicast source node in the multicast group, the active route of the multicast group. priority.

The step CI specifically includes:

Cl l, each node requests the authentication of the CA of the certification center before joining the mobile ad hoc network MANET;

C12. After the security authentication succeeds, the CA allocates a corresponding priority according to the importance of the node in the MANET.

The step C2 specifically includes:

The node C2 initiates a multicast routing request as a multicast source, and embeds its own priority into the multicast routing request discovery process, and establishes a corresponding multicast forwarding structure according to the multicast routing request discovery process;

C22. The intermediate forwarding node in the multicast forwarding structure records the priority of the multicast source into its multicast routing information.

The present invention further provides a network device, where the network device is a multicast forwarding node, and the network device is provided with a storage module, a congestion prediction module, and a multicast forwarding structure module;

The storage module is configured to store the priority of all active multicast routes through the node where the node is located; the congestion prediction module is configured to predict the congestion of the node, and notify the multicast forwarding structure module when it is confirmed that the node is about to be congested. ;

The multicast forwarding structure module is configured to: after receiving the notification, establish a new multicast forwarding structure that does not go through the node where the node is located according to the priority stored in the storage module, so that the upstream node performs group according to the new multicast forwarding structure. Broadcast service forwarding to perform congestion control on real-time multicast services.

The multicast forwarding structure module includes:

The initiating sub-module is configured to: after receiving the notification, select an active multicast route with the lowest priority from all active multicast routes passing through the node, and send a congestion report to the downstream node of the selected active multicast group route Text

The grafting sub-module is configured to: after receiving the congestion message transmitted by the other node, send a multicast group join request to the neighboring node according to the multicast group ID information carried in the congestion packet, and receive the packet After the response message to the neighboring node is confirmed, the user is successfully grafted to the forwarding structure of the multicast group, and the node that sends the congestion packet is replied to the information that the route dynamic adjustment succeeds;

The invalidation sub-module is configured to send a multicast group route invalidation message to the upstream node of the node where the node is located, so that the upstream node determines the multicast group route of the newly established multicast forwarding structure according to the received multicast group route invalidation message. Information, and use the routing information to forward the corresponding multicast service.

The multicast forwarding structure module is further configured with a multicast join determination sub-module and a response sub-module. The multicast join determination sub-module is configured to determine whether the node is in compliance after receiving the multicast group join request message received by the node. The conditions for joining the multicast group and notifying the response submodule;

The response sub-module is configured to: when receiving the notification of the compliance, send a response message to the node that sends the request message to the multicast group, and discard the multicast group join request message when receiving the notification of the non-conformity.

The network device is further provided with a routing request judging module and an adding module;

The routing request judging module is configured to: when a new routing request discovery message arrives in the congestion control process of the node, determine whether the node is the forwarding node of the multicast routing request message, and notify the joining module;

The module is added to notify the forwarding node according to the routing request judgment module, discards the route request discovery message, and joins the multicast group according to the notification sent by the routing request judgment module that is not the forwarding node.

It can be seen from the technical solution provided by the present invention that, in the present invention, each multicast forwarding node first predicts congestion of its own node; when it confirms that congestion is about to occur, it is based on all active multicasts passing through the node. The level of routing controls congestion for real-time multicast services. The present invention can ensure that high-quality services can be obtained for real-time multicast services of different levels (higher level and lower level); and the adjustment of the active multicast routing of the present invention is performed under the premise of impending congestion. Therefore, a smooth transition of the active multicast routing is implemented, the delay and the packet loss rate are minimized, so that the congestion control is more robust, and the invention can implement load balancing and resource optimization of the network, thereby Improve the performance of the network. In addition, the present invention does not employ a complicated congestion algorithm, which reduces overhead and avoids MANET. The consumption of valuable resources of the network. BRIEF DESCRIPTION OF THE DRAWINGS

Figure 1 is a flow chart of a first embodiment of the present invention;

2 is a schematic diagram of a GPCCA congestion control process in a first embodiment of the present invention; FIG. 3 is a schematic diagram of GPCCA congestion control results in a first embodiment provided by the present invention.

Mode for carrying out the invention

The present invention provides a method for performing congestion control on a real-time multicast service, the core of which is: each multicast forwarding node predicts congestion of its own node; when the multicast forwarding node confirms that congestion is about to occur, according to the The level of all active multicast routes of the node controls the congestion of the real-time multicast service.

The present invention solves the congestion problem of real-time multicast services in MANET through GPCCA (Group Priority based Congestion Control Architecture). The priority of multicast is determined by the importance of the multicast source. GPCCA uses it to implement congestion control in the multicast protocol. This method is compatible with existing congestion control methods.

The main idea of the first embodiment provided by the present invention is: First, each multicast forwarding node performs congestion prediction by detecting the resource utilization of the own node; when the multicast forwarding node confirms that congestion is about to occur, according to the The level of all active multicast routes of the node controls the congestion of the real-time multicast service. The specific implementation process is shown in Figure 1, including:

Step 101: Determine a priority of the multicast group active route according to a priority of the multicast source node in the multicast group.

Each node joining the MANET network must first pass the security verification of the CA (Certificate Authority). After the security verification succeeds, the CA assigns a corresponding priority according to the importance of the nodes joining the network in the MANET. The higher the importance of the node, the higher the corresponding priority, and the network at the node. The priority of this node remains unchanged during the lifetime.

When a node in the network wants to send multicast data, but there is no multicast route at this time, The node initiates a multicast routing request as a multicast source, and embeds its own priority in the process of multicast routing request discovery (Route Request), and establishes a corresponding multicast forwarding structure through the multicast routing request discovery process. . After the multicast forwarding structure is established, the intermediate forwarding node of the multicast forwarding structure records the priority of the multicast source into its multicast routing information, and the real-time multicast data begins to establish a good multicast forwarding structure. Forwarded on.

Step 102: Each multicast forwarding node detects whether the bandwidth utilization of the own node reaches a congestion threshold. If yes, step 103 is performed. In step 103, it is confirmed that the multicast forwarding node is about to be congested, and then step 104 is performed; otherwise, Step 105 is performed to confirm that the node is in a normal working state, and then proceed to step 102.

Each multicast forwarding node in the present invention monitors the resource utilization of its own nodes for congestion prediction. In steps 102 and 103, the multicast forwarding node may confirm that the multicast forwarding node is about to become congested by detecting whether the bandwidth utilization of the own node exceeds a congestion threshold. Of course, in the steps 102 and 103, the above two detection methods can also be used simultaneously to confirm that the multicast forwarding node is about to be congested.

Step 104: The node that is congested performs congestion control on the real-time multicast service according to the priority of all active multicast routes.

During the congestion control process, the node that is about to be congested is predicted to discover the newly received route request message, and first determines whether it is the receiver of the route request discovery message, if it is not the multicast that is the route request discovery message. The receiver discards the processing, that is, guarantees that no new services are added, and congestion control is performed on the congested nodes. The message is received if it is determined to be the recipient of the route request discovery message. The specific congestion control process is as follows:

Step 11: The camping node selects an active multicast route with the lowest priority from all the active multicast routes passing through the congestion node, and sends a congestion packet to the downstream node of the selected active multicast group route.

If the same priority is used to adjust the multicast routing object, the bandwidth usage is small. The dynamic multicast route is adjusted, and a congestion message is sent to the downstream node of the selected active multicast group route.

Step 12: The downstream node sends a multicast group join request message to its neighbor node according to the multicast group ID information carried in the congestion packet.

Step 13: After receiving the multicast group join request message, the neighboring node determines whether it meets the conditions for joining the multicast group, that is, after determining that the multicast route is established, the neighboring node does not cause itself to be congested, and The route of the upstream node of the congested node does not pass through the node to be congested, and does not pass through the congested node. If yes, the response message is sent back to the downstream node; otherwise, the multicast group join request message is discarded.

Step 14: After receiving the response message of the neighboring node, the downstream node confirms that it has been grafted to the forwarding structure of the multicast, and then sends information about successful routing dynamic adjustment to the node to be congested.

Step 15: The initiating congestion node sends the multicast group route invalidation message to the upstream node of the adjusted active multicast route. The multicast group route invalidation packet includes the group address to be adjusted and the upcoming congestion node. The downstream node address.

Step 16: After receiving the multicast group route invalidation message, the upstream node forwards the corresponding multicast service according to the multicast group routing information in the newly established multicast structure, that is, the upstream of the congestion node After receiving the multicast group route invalidation packet, the node extracts the address of the downstream node of the congestion node from the multicast group route invalidation message, and then joins the downstream node of the congestion node to the multicast group, and avoids The node is about to be congested. As can be seen from the above description, the multicast group route invalidation message at this time can be regarded as the multicast join request message of the downstream node of the congested node. Therefore, the multicast forwarding structure of the downstream node of the congested node can be re-joined into the multicast group.

It can be seen from the congestion control process in steps 11 to 16 that the congestion node first selects the active multicast route with the lowest priority from its own multicast routing table as the adjustment object, and then approaches the congestion node to the downstream of the multicast route. The node sends congestion packets, and the downstream nodes are based on multicast. The group ID information is sent to the neighboring node to send a multicast join request message. After the downstream node receives the multicast join response message returned by the neighboring node, the downstream node is successfully grafted into the multicast group forwarding structure. That is, the downstream node actively joins the multicast group as a new multicast receiver, and the route from the downstream node to the multicast source avoids the upcoming congestion node, and the multicast on the new multicast forwarding structure. The forwarding node cannot be congested. At this time, the downstream node sends a route dynamic adjustment success to the to-be-congested node, and the congestion node needs to send the multicast group route invalidation packet to the upstream node corresponding to the multicast route, and it is noted that The link between the upstream node and the to-be-congested node also maintains connectivity, so the connectivity of the link is not changed due to dynamic adjustment of active multicast routing in the congestion control process. After the upstream node receives the multicast group route invalidation message, the upstream node will not send the multicast group data to the upcoming congestion node, but forward the corresponding multicast data through the newly established multicast route. It can be seen that during the dynamic adjustment of the multicast routing, the multicast data is forwarded by the congested node. Only after the new multicast routing is established, the multicast data is forwarded through the new multicast routing, thus ensuring the active group. A smooth transition of the broadcast route. Since the early congestion prediction mechanism is adopted, the impact on real-time multicast services is very small.

The specific implementation process of the first embodiment provided by the present invention will be described below by taking the network topology shown in FIG. 2 as an example. In the topology shown in Figure 2, assuming that the multicast forwarding node B predicts that its bandwidth utilization exceeds the congestion threshold, it immediately starts the congestion control policy and performs congestion control on its own multicast service. The control process is as follows: :

Step 21: That is, the congested Node B only receives the new multicast routing request as the multicast member, and the other multicast routing request message is automatically discarded, so that no new load is added; Step 22, the congested Node B is about to be added. First, the active multicast route with the lowest priority is found in the multicast routing table, and then the congestion packet is sent to the downstream node E corresponding to the route.

If the same priority is used to adjust the multicast routing object, the active multicast routing with small bandwidth is adjusted to adjust the impact of the active routing adjustment on the network performance to a minimum.

Step 23: After receiving the congestion packet, the downstream node E sends a multicast join to all its neighbor nodes. a request message, the request message includes the multicast group ID, and the IP of the neighboring node to be avoided; Step 24, each neighbor node, such as the node F, after receiving the multicast join request message, determines whether the join condition is met, That is, after establishing a multicast route with itself, it will not cause congestion, and the route to the multicast source S will not pass through the node to be congested, nor will it pass through the congestion node. When it is confirmed that the above conditions are met, the node F is The multicast routing request message is answered.

Step 25: After receiving the multicast routing response message of the node F, the downstream node E can receive the multicast data through the node F, and then the downstream node E sends the multicast graft success message to the node F. At the same time, the multicast node adjustment success packet is sent to the node B, and the downstream node E can receive the multicast group data through the node F. The downstream node E discards the same multicast routing response message received later, and only establishes a route with the sender of the first received multicast response message. .

As shown in FIG. 3, the adjusted multicast group forwarding structure avoids the congestion of the node B and receives the multicast data sent by the multicast source S from the non-congested nodes C and F.

Step 26: After receiving the adjustment success message sent by the downstream node E, the congestion node B sends the multicast route invalidation message to the upstream node, that is, the multicast source S node, and the multicast source S to the node B, and The multicast route between Node B and Node E is invalid. The multicast source S forwards the corresponding multicast data through the newly established multicast routing structure as shown in FIG.

When the congestion node B detects that it has returned to the normal forwarding state, the congestion control process ends. If it is detected that it is still in a state of congestion, the same strategy is used to continue the adjustment, that is, the above 22 to 27 process is repeated until it returns to normal.

The second embodiment of the present invention is different from the first embodiment in that: in this embodiment, each multicast forwarding node performs congestion prediction by detecting the energy of the own node. The specific prediction process includes:

Step 201: Each multicast forwarding node detects whether the energy of the own node falls to a security threshold, and if yes, confirms that the node is about to be congested; otherwise, confirms that the node is in a normal working state. When a node detects that its own energy is lower than the security threshold, it will adjust all its active multicast routes (except that the node acts as a multicast source or multicast receiver). The adjustment process is the same as in the first embodiment. The related description is the same, and will not be described in detail here.

There are many indicators in the link layer that can be used to measure network congestion. In the above embodiment, only the multicast forwarding node uses bandwidth utilization and energy as the standard for measuring congestion. When using other links in the link layer, When the indicator measures whether the multicast forwarding node is congested, the specific implementation process is similar to the related description in the foregoing embodiment, and is not described in detail.

The present invention also provides a network device that is a multicast forwarding node in a multicast network. The network device of the present invention is provided with a storage module, a congestion prediction module, a multicast forwarding structure module, a route request judging module, and an adding module.

The storage module is primarily used to store the priority of all active multicast routes through the node it is on. Each node in the present invention is set with a priority, and the priority of the node may be allocated according to the importance of each node joined to the mobile ad hoc network MANET. The priority of the node can be assigned by the certificate authority CA. For example, each node requests the authentication of the certificate authority CA before joining the mobile ad hoc network MANET. After the security authentication succeeds, the CA assigns the corresponding node according to its importance in the MANET. Priority. The priority of the multicast group active route in the present invention is determined according to the priority of the multicast source node in the multicast group, for example, the node initiates a multicast routing request as a multicast source, and embeds its own priority into the group. During the discovery of the broadcast route request, the node establishes a corresponding multicast forwarding structure according to the multicast routing request discovery process, and the intermediate multicast forwarding node in the multicast forwarding structure records the priority of the multicast source into its multicast routing information. . Specifically, it is described in the above method.

The congestion prediction module is mainly used to predict the congestion of the node where it is located, and notify the multicast forwarding structure module when it is confirmed that the node where it is located is about to be congested. The congestion prediction module can perform congestion prediction by detecting the resource utilization of the node where it is located. The congestion prediction module can also perform congestion prediction by detecting the energy of the node where it is located, and can also perform resource consumption and energy by detecting the resource utilization of the node. Congestion prediction. Specifically, it is described in the above method. The multicast forwarding structure module is mainly configured to: after receiving the notification of the congestion prediction module, establish a new multicast forwarding structure that does not go through the node where the node is located according to the priority stored in the storage module, so that the upstream node is configured according to the new multicast forwarding structure. Perform multicast service forwarding to perform congestion control on real-time multicast services.

The multicast forwarding structure module is composed of an initiator submodule, a graft submodule, a dead submodule, a multicast join judging submodule, and a response submodule.

The initiating sub-module is mainly configured to: after receiving the notification of the congestion prediction module, select an active multicast route with the lowest priority from all active multicast routes of the node, and route the downstream node to the selected active multicast group. Send a congestion message. When the active multicast route with the lowest priority is multiple, the initiating submodule selects an active multicast route with a small bandwidth occupation, as described in the foregoing method. When the initiating submodule is a multicast forwarding node that is about to be congested, the above operation is performed.

The grafting sub-module is configured to send a multicast group join request to the neighboring node according to the multicast group ID information carried in the congestion packet after receiving the congestion packet transmitted by the other node, and receive the response message of the neighbor node. After that, it is confirmed that the user has successfully grafted to the forwarding structure of the multicast group, and replies to the node that sends the congestion message with the information that the route dynamic adjustment succeeds. The grafting sub-module only establishes a route with the sending node of the first received reply message, and discards the reply message of the same multicast route that is received later. Specifically, it is described in the above method. When the node where the grafting submodule is located is the downstream node of the multicast forwarding node to be congested, the above operation is performed.

The multicast join judgment sub-module is mainly used to determine whether the node meets the conditions for joining the multicast group after the multicast group join request message received by the node, and notifies the response sub-module. The conditions for joining a multicast group here may be as follows: After the node establishes a multicast route, the node does not cause congestion of the node, and the route from the node to the upstream node of the node to be congested does not pass through the node that is congested. Will not pass the node that is about to be congested. Specifically, it is as described in the above method. When the multicast join judgment node is the neighbor node of the downstream node of the multicast forwarding node to be congested, the above operation is performed. The response sub-module is mainly used to send a response message sent by the node that sends the multicast group to the request message when receiving the notification of the compliance sent by the multicast join judgment sub-module, and the non-conformity transmitted by the multicast join judgment sub-module is received. When the notification is made, the multicast group join request message is discarded. When the node where the response submodule is located is the neighbor node of the downstream node of the multicast forwarding node to be congested, the above operation is performed.

The invalidation sub-module is mainly used to send a multicast group route invalidation packet to the upstream node of the node where the node is located when receiving the information that the route dynamic adjustment succeeds. At this time, the multicast route between the upstream node and the node where the failed submodule is located is invalid. In this way, the upstream node can determine the multicast group routing information of the newly established multicast forwarding structure according to the received multicast group routing invalidation message, and use the routing information to forward the corresponding multicast service. When the node where the failed submodule is located is the multicast forwarding node to be congested, the above operation is performed.

The routing request judging module is mainly used to determine whether the node is the multicast forwarding node of the multicast routing request message when the new routing request discovery message arrives during the congestion control process of the node, that is, whether the node is located The receiver of the routing request discovery message notifies the joining module of the judgment result of the recipient or not the recipient. Specifically, it is described in the above method. When the node where the route request judgment module is located is the multicast forwarding node to be congested, the above operation is performed.

The adding module is mainly used to discard the routing request discovery message according to the notification that the routing request is sent by the module, and the routing request judging module sends the notification of the receiver to join the multicast group. When the node where the route request judgment module is located is the multicast forwarding node to be congested, the above operation is performed.

Through the above operation processing of the routing request judging module and the joining module, it is ensured that there is no more new multicast service, and further congestion control is performed on the congested node.

As can be seen from the above specific embodiments of the present invention, the present invention has the following beneficial effects:

1. It has good adaptability to the adjustment of active multicast routing, realizes load balancing of the network, improves the throughput of the network, realizes the optimization of network resource allocation, and also adapts to MANET. An environment in which the topology changes dynamically;

2. The adjustment of active multicast routing is based on the premise of congestion. Therefore, the smooth transition of active multicast routing will minimize the delay and packet loss rate. The impact on real-time multicast will be minimal, so The real-time multicast service is well supported and greatly avoids congestion.

3. At the expense of any active routing, the adjustment of the low-priority real-time multicast service can obtain more effective resources to ensure the quality of service, and the remaining unadjusted multicast services are also improved due to congestion conditions. Better quality of service;

4. The energy of the node is used as a factor of congestion avoidance, which can prolong the survival time of the node in the network, and can also provide better service quality for the multicast service, and has good application value in military applications;

5. No complicated congestion algorithm is adopted, which reduces overhead and avoids the consumption of precious resources of the MANET network;

6. It has good security. The external node obtains the priority issued by the CA after it has been verified by the security of the CA. Therefore, it has security protection and has little change to the existing multicast routing protocol. It has good deployment. '

The above is only a preferred embodiment of the present invention, but the scope of the present invention is not limited thereto, and any person skilled in the art can easily think of changes or within the technical scope disclosed by the present invention. Alternatives are intended to be covered by the scope of the present invention. Therefore, the scope of protection of the present invention should be determined by the scope of protection of the claims.

Claims

Rights request

A method for performing congestion control on a real-time multicast service, comprising:

A. Each multicast forwarding node predicts congestion of its own node;

The method according to claim 1, wherein the step A specifically includes: Al, each multicast forwarding node performs congestion prediction by detecting resource utilization of the own node; and/or,

A2. Each multicast forwarding node performs congestion prediction by detecting the energy of its own node.

The method according to claim 2, wherein the step A1 specifically includes: each multicast forwarding node detects whether the bandwidth utilization of the own node reaches a congestion threshold, and if so, confirms that the node is about to become congested. ; Otherwise, confirm that the node is in normal working condition.

The method according to claim 2, wherein the step A2 specifically includes: each multicast forwarding node detects whether the energy of the own node falls to a security threshold, and if so, confirms that the node is about to be congested; otherwise , confirm that the node is in normal working condition.

The method according to claim 1, wherein the step B specifically includes: Bl, when the multicast forwarding node confirms that it is about to be congested, selecting a priority from all active multicast routes through its own node. The lowest active multicast route, and sending a congestion message to the downstream node of the selected active multicast group route;

After receiving the congestion packet, the downstream node sends a multicast group join request to the neighbor node according to the multicast group ID information carried in the congestion packet, and receives the response of the neighbor node. After the message, it is confirmed that it has successfully grafted to the forwarding structure of the multicast group, and then sends the information that the route dynamic adjustment succeeds to the congestion node;

B3, the fast-moving node sends the multicast group route invalidation message to its upstream node; B4. The upstream node determines, according to the received multicast group route invalidation message, the multicast group routing information of the newly established multicast forwarding structure, and uses the routing information to forward the corresponding multicast service.

The method according to claim 5, wherein in the step B1, if there are multiple active multicast routes with the lowest priority, the active multicast routing with a small bandwidth occupation is selected.

The method according to claim 5, wherein the step B2 further comprises: after receiving the multicast group join request message, the neighboring node determines whether it meets the condition for joining the multicast group, if If yes, the response message is sent back to the downstream node; otherwise, the multicast group join request message is discarded.

The method according to claim 7, wherein the condition for joining the multicast group comprises: - the multicast forwarding node does not make itself impending congestion after establishing the multicast route, and goes to the upcoming congestion node by itself The route of the upstream node does not pass through the node that is about to be congested, nor does it pass through the congested node.

The method according to any one of claims 1 to 8, wherein the step B further comprises:

If a new route request discovery message arrives during the process of performing congestion control, the neighboring node determines whether it is the forwarding node of the multicast routing request message, and if so, discards; otherwise, receives the discovery message and Join the multicast group.

10. The method according to claim 1, further comprising, before said step A, -

C2: Determine a priority of the active routing of the multicast group according to a priority of the multicast source node in the multicast group.

The method according to claim 10, wherein the step C1 specifically includes: C11. Before joining the mobile ad hoc network MANET, each node requests security authentication of the CA at the certification center;

The method according to claim 10, wherein the step C2 specifically includes: C21, the node acts as a multicast source to initiate a multicast routing request, and embeds its own priority into the multicast routing request discovery. In the process, a corresponding multicast forwarding structure is established according to the multicast routing request discovery process;

A network device, which is a multicast forwarding node, wherein the network device is provided with a storage module, a congestion prediction module, and a multicast forwarding structure module;

,

The network device according to claim 13, wherein the multicast forwarding structure module comprises:

The grafting sub-module is configured to: after receiving the congestion message transmitted by the other node, send a multicast group join request to the neighboring node according to the multicast group ID information carried in the congestion packet, and receive the neighbor After the node responds to the message, it confirms that it has successfully grafted the forwarding to the multicast group. Structurally, the node that sends the congestion message is replied to the information that the route dynamic adjustment succeeds; the failure sub-module is configured to send the multicast group route invalidation message to the upstream node of the node where the node is located, so that the upstream node is based on the received group. The multicast routing failure message determines the multicast group routing information of the newly established multicast forwarding structure, and uses the routing information to forward the corresponding multicast service.

The network device according to claim 14, wherein the multicast forwarding structure module further includes a multicast join determination submodule and a response submodule;

The multicast join judgment sub-module is configured to determine, according to the multicast group join request message received by the node, whether the node meets the conditions for joining the multicast group, and notify the response sub-module;

The network device according to claim 13 or 14, wherein the network device is further provided with a routing request judging module and an adding module;