Software-defined wireless Mesh network adaptive networking method and system based on multilink cooperation

Abstract

The invention relates to the technical field of mobile communication, in particular to a software-defined wireless Mesh network self-adaptive networking method based on multi-link cooperation, which comprises the following steps that: each node discovers neighbor nodes, and links of three different frequency bands are connected to form a Mesh network; an SDN controller is introduced, node state information is collected in real time, and an optimal route is selected as required according to different message types and link characteristics in combination with network topology changes and service requirements; and when the node has a fault or is off-network, combining with a node mobility management mechanism, enabling a routing protocol to converge quickly, and adjusting a transmission path at the same time. According to the invention, software is adopted to define routing, the controller uniformly distributes routing and link selection, and on-demand routing according to service types can be realized; the link weight is adaptively changed by collecting network state information, and a proper route and a proper link are selected according to the link condition and the service type; and an E-Dijkstra algorithm is adopted, so that rapid decision-making of routing and link selection can be realized, and the processing time delay is reduced.

Description

Technical Field

[0001] The present invention relates to the field of mobile communication technology, and in particular to a software-defined wireless Mesh network adaptive networking method and system based on multi-link collaboration. Background Art

[0002] With the rapid development of mobile Internet and wireless communication technologies, people's lifestyles have changed significantly. A variety of network services have significantly improved people's life experience. However, with the increasing diversification of needs, people have higher requirements for multimedia mobile communications and expect to provide more abundant service types. In order to meet these needs, many high-speed broadband communication technologies have emerged. As one of the most critical network architectures, wireless mesh networks have gradually become the focus because they provide users with convenient and economical Internet access services.

[0003] Wireless Mesh Networks have the characteristics of high speed, easy networking, self-healing and self-organization, and have attracted extensive attention from major global standardization research organizations and scientific research institutions, and have been widely used in practice, such as network deployment in campuses, enterprises and other temporary scenarios. However, traditional wireless Mesh Networks often face problems such as network congestion, instability and inability to guarantee QoS.

[0004] To solve these problems, introducing software-defined networking (SDN) technology into wireless mesh networks is an effective means. SDN technology enables the network controller to obtain a global view, thereby dynamically allocating routing strategies based on service types to meet the QoS requirements of services.

[0005] In addition, in Mesh networks, it is of great significance to use multi-band networking between nodes. It can improve bandwidth utilization, reduce signal interference, enhance network reliability, and optimize QoS according to business needs. Multi-band networking enables nodes to switch frequency bands to ensure communication continuity when the network is congested or interfered, and helps to expand network coverage and achieve efficient and stable network performance. Based on this, we propose a software-defined wireless Mesh network adaptive networking method and system based on multi-link collaboration. Summary of the invention

[0006] The purpose of the present invention is to overcome the above problems or at least partially solve the above problems, and to propose a software-defined wireless mesh network adaptive networking method and system based on multi-link collaboration.

[0007] To achieve the above object, the present invention provides the following technical solution: a software-defined wireless mesh network adaptive networking method based on multi-link collaboration, comprising the following steps:

[0008] Step S1: Network discovery: Each node discovers neighbor nodes. Each node builds links for different frequency bands, such as 1.4G, 2.4G, and 5G, and calculates the link quality. After the link quality evaluation is qualified, each node is connected through three different frequency band links to form a Mesh network.

[0009] Step S2: Adaptive networking, introducing SDN controller to communicate with Mesh nodes. Mesh nodes periodically provide status information to SDN controller. SDN controller collects node status information in real time and monitors messages reported by Mesh nodes. Combined with network topology changes and business needs, it selects the optimal route according to different message types and link characteristics.

[0010] Step S3, network maintenance: during network operation, when a node in the network fails or goes offline, the node mobility management mechanism is combined to enable the routing protocol to quickly reach a convergence state, and the transmission path is adaptively adjusted.

[0011] In a preferred embodiment, the step S1 comprises the following steps:

[0012] Step S1.1, each node loads the batman-adv routing protocol internally, and completes the establishment of 2.4G and 5G links through the OGM broadcast packet generated by the protocol;

[0013] Step S1.2: Each node is internally loaded with a 1.4G self-organizing network module, which starts the Mesh networking mode to realize neighbor discovery and networking, and complete the construction of the 1.4G link;

[0014] Step S1.3, links in different frequency bands maintain link-related information and calculate link quality in their own way;

[0015] In 2.4G and 5G links, the link quality is calculated based on the number of OGM broadcast packets sent and received. The link quality is divided into three parts: RQ (receive link status), EQ (loop link status) and TQ (transmit link status). In 1.4G links, the link quality is automatically evaluated by the self-organizing network module according to the internal algorithm.

[0016] In a preferred embodiment, the step S2 comprises the following steps:

[0017] Step S2.1, the SDN controller summarizes the status information of all nodes, and the communication link between the SDN controller and the Mesh node is a 1.4G link;

[0018] Step S2.2, the SDN controller continuously monitors the packet-in message reported by the Mesh node and determines the message type;

[0019] Step S2.3: If the message received by the SDN controller is Mesh node status information, the adjacency matrix, signal strength matrix, packet loss rate matrix, and mobility matrix of the three links will be maintained according to the reported status information, and finally the comprehensive weight matrix of the three links will be calculated;

[0020] Step S2.4: If the message received by the SDN controller is business data type information, the comprehensive weight matrix will be adjusted according to the business type, and the E-Dijkstra algorithm will be run to obtain the routing path and link selection plan. Finally, the flow table will be constructed and sent to the Mesh node.

[0021] In a preferred embodiment, the step S2.3 comprises the following steps:

[0022] Step S2.3.1: The SDN controller updates and maintains the state matrices of 1.4G, 2.4G and 5G links respectively according to the link state information reported by the Mesh node, and calculates the comprehensive weight matrices W of the three links respectively according to the updated link state matrices. 1.4G , W 2.4G and W 5G ;

[0023] Step S2.3.2: Elements in the adjacency matrix of three different links and The value is 0 or 1, where L ij Indicates the link adjacency relationship between node i and node j. If the value is 0, it means that there is no link in this frequency band between the nodes. If the value is 1, it means that there is a link in this frequency band between the nodes.

[0024] Step S2.3.3: Elements in the signal strength matrix of three different links and The value of in is the received signal strength of different frequency bands between node i and node j, where is the channel capacity of different frequency bands between node i and node j, is the bandwidth of different frequency bands between node i and node j, is the signal-to-noise ratio of different frequency bands between node i and node j, it can be proved and The value of is between 0 and 1;

[0025] Step S2.3.4: Elements in the mobility matrix of three different links and The value of Among them, M ij represents the mobility index between node i and node j, M nrepresents the mobility index of node n, represents the set of neighbor nodes of node n in different frequency bands at time t, Represents the set of neighbor nodes of n nodes in different frequency bands at time t-Δt, it can be proved that and The value of is between 0 and 1;

[0026] Step S2.3.5: Elements in the packet loss rate matrix of three different links and D ij represents the packet loss rate indicator between node i and node j, where and Calculated based on the TQ value of the batman-adv protocol, It is automatically evaluated by the self-organizing network module according to the internal algorithm, which can prove and The value of is between 0 and 1;

[0027] Step S2.3.6: Elements in the comprehensive weight matrix of three different links and a+b+c=1, where a, b and c are variable parameters used to control the weights of signal strength, packet loss rate and mobility in link evaluation. Finally, the comprehensive weight matrix W of three different links is obtained: 1.4G , W 2.4G and W 5G .

[0029] In a preferred embodiment, the step S2.4 comprises the following steps:

[0030] Step S2.4.1: When the SDN controller receives the service data type information, it first determines the service type, and then calculates the comprehensive weight matrix W of the three links according to different service types. 1.4G , W 2.4G and W 5G After modification, the E-Dijkstra algorithm is used as input to calculate the routing path and link selection;

[0031] Step S2.4.2: After the SDN controller calculates the final result, it sends the routing and link selection to the Mesh node in the form of a flow table. The Mesh node updates the routing according to the flow table and executes the corresponding strategy.

[0032] In a preferred embodiment, the step S3 comprises the following steps:

[0033] Step S3.1, the Mesh node regularly monitors its own status information and uploads it to the SDN controller;

[0034] Step S3.2: When the SDN controller receives the status information reported by the node, it calculates the link quality of the corresponding frequency band based on the status information. or The calculated link quality is compared with the link quality threshold or If the value is lower than the corresponding threshold, the adjacency matrix value corresponding to the link is Temporarily set to 0;

[0035] Step S3.3: When a link of a node fails, the value corresponding to the link of the node in the adjacency matrix will be set to 0 to prevent other nodes from continuing to transmit data through this link. At the same time, the controller sets up a recovery mechanism for the failed node;

[0036] Step S3.4: Each Mesh node calculates its own mobility index Adaptively adjust the broadcast frequency of OGM broadcast packets.

[0037] In a preferred embodiment, in step S3.4, the broadcast interval of the OGM broadcast packet is changed by calculating the node mobility index only for optimizing the 2.4G and 5G frequency band links in the mesh network, wherein the adjustment of the broadcast frequency of the OGM broadcast packet is based on the mobility index of the Mesh node itself, the specific adjustment method of the broadcast time interval is related to the node mobility index, and the broadcast time interval of the OGM broadcast packet is The unit is milliseconds.

[0038] In a preferred embodiment, RQ represents the number of OGM broadcast packets received from the neighbor interface. Taking into account the influence of asymmetric links, the local TQ calculation formula is: TQ local =TQ local ×(1-(1-RQ) 3 ).

[0039] In a preferred embodiment, based on the local TQ calculation, the influence of link quality in network transmission, especially the influence of hop count on link quality, is further considered. Each hop is multiplied by the hop count penalty coefficient hop_penalty, so that the calculation formula of the protocol TQ is obtained as follows: TQ = TQ local ×TQ OCM ×(1-(1-RQ) 3 )×hop_penalty.

[0040] The present invention also provides a software-defined wireless Mesh network adaptive networking system based on multi-link collaboration, including a processor and a memory, wherein instructions are stored in the memory, and when the instructions are executed by the processor, the system executes the networking method.

[0041] Compared with the prior art, the present invention has the following beneficial effects: the present invention adopts a software-defined routing approach, in which the controller uniformly allocates routing and link selection, and can realize on-demand routing according to the business type; by collecting network status information, the link weight is adaptively changed, and appropriate routes and links are selected according to the link status and business type; the present invention adopts the E-Dijkstra algorithm, which can realize rapid decision-making on routing and link selection and reduce processing delay. BRIEF DESCRIPTION OF THE DRAWINGS

[0042] Figure 1 This is a rendering of the implementation of the software-defined wireless mesh network adaptive networking algorithm based on multi-link collaboration provided by the present invention;

[0043] Figure 2 is a flow chart of the protocol network discovery in the present invention;

[0044] Figure 3 It is a flow chart of the adaptive networking in the present invention;

[0045] Figure 4 It is a flowchart of the protocol networking maintenance in the present invention;

[0046] Figure 5 It is a flow chart of mobility management in the present invention. DETAILED DESCRIPTION

[0047] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0048] like Figure 1 As shown, the present invention provides a software-defined wireless mesh network adaptive networking algorithm based on multi-link collaboration. Considering the mesh network of three different frequency bands, and adding the SDN network controller, the on-demand routing according to different service types is realized through the software-defined routing method. The controller collects the status information of all nodes in the network, and adopts different paths and links for different services from aspects such as node mobility, link signal strength, and link packet loss rate, so as to realize the combination of link diversity, network dynamics and service dynamics, and ensure the QoS of the service.

[0049] The network discovery process of the present invention is as follows Figure 2 As shown, the following steps are included:

[0050] Step S1: Network discovery: Each node discovers neighbor nodes. Each node builds links for different frequency bands, such as 1.4G, 2.4G, and 5G, and calculates the link quality. After the link quality evaluation is qualified, each node is connected through three different frequency band links to form a Mesh network.

[0051] Step S1 also includes the following steps:

[0052] Step S1.1, each node loads the batman-adv routing protocol internally, and completes the construction of 2.4G and 5G links through the OGM broadcast packet generated by the protocol. The content of the OGM broadcast packet includes ttl, sequence number, MAC address of the source node, MAC address of the previous hop node, and link transmission quality;

[0053] Step S1.2: Each node is internally loaded with a 1.4G self-organizing network module, which starts the Mesh networking mode to realize neighbor discovery and networking, and complete the construction of the 1.4G link;

[0054] Step S1.3: For 2.4G and 5G links, the node performs an update operation after receiving the OGM broadcast packet. The update content includes the link quality between the local node and the one-hop neighbor, and the list of source nodes reachable by the local node; for 1.4G links, the node directly maintains the neighbor relationship and the list of reachable nodes through the self-organizing network module, and calculates the link quality;

[0055] In 2.4G and 5G links, the link quality is calculated based on the number of OGM broadcast packets sent and received. The link quality is divided into three parts: RQ (receive link status), EQ (loopback link status) and TQ (transmit link status). In 1.4G links, the link quality is automatically evaluated by the self-organizing network module according to the internal algorithm.

[0056] RQ indicates the number of OGM broadcast packets received from the neighbor interface. When this node receives a broadcast packet sent by a neighbor node, it first checks whether the sequence number in the message is the latest. If it is, it puts the message into the sliding window. The sliding window moves forward, and the latest position is '1'. Messages outside the sliding window will be discarded. Within the specified time, the number of '1's in the sliding window is used to count the number of OGM messages received from the neighbor during this time period, and the RQ value of all neighbor interfaces can be calculated. The calculation of EQ is similar to RQ, and also uses a sliding window for calculation. After obtaining EQ and RQ, the local TQ value can be calculated. The local TQ calculation formula is: In addition, the impact of asymmetric links is also considered, and the formula is: TQlocal =TQ local ×(1-(1-RQ) 3 ).

[0057] Then comes the calculation of the transmission quality of a path. When the source node sends a message, the TQ field is set. This field is 8 bits and the maximum value is 255. After the local node receives the broadcast packet sent by the neighboring node, it calculates the local transmission quality and extracts the TQ field in the message. When rebroadcasting, the local node recalculates a global transmission quality based on the local transmission quality and the extracted TQ field, and puts it into the rebroadcast message for outward transmission. The formula is: TQ = TQ local ×TQ OGM Finally, consider the impact of hop count on link quality. Each hop is multiplied by the hop count penalty coefficient hop_penalty. In summary, the calculation formula of the protocol metric TQ is: TQ = TQ local ×TQ OCM ×(1-(1-RQ) 3 )×hop_penalty.

[0058] The adaptive networking process of the present invention is as follows Figure 3 As shown. In this process, the SDN controller is introduced to collect the status information of all nodes in real time, combine the network topology changes and business needs, and select the optimal route according to different business types and link characteristics. The specific steps are as follows:

[0059] Step S2: Adaptive networking, introducing SDN controller to communicate with Mesh nodes. Mesh nodes periodically provide status information to SDN controller. SDN controller collects node status information in real time and monitors messages reported by Mesh nodes. Combined with network topology changes and business needs, it selects the optimal route according to different message types and link characteristics.

[0060] Specifically, in step S2.1, the SDN controller summarizes the status information of all nodes. The link between the SDN controller and the Mesh node is a 1.4G link. Since the 1.4G link has a large coverage range and good penetration, it can ensure that the link between the Mesh node and the controller is as direct as possible. In addition, the 1.4G frequency band has little interference and a relatively small bandwidth, which is more suitable for transmitting such high-reliability data.

[0061] Install RYU software in the SDN controller and install and compile open Vswitch software in the Mesh node to implement data interaction between the controller and ordinary nodes. The data interaction adopts the openflow protocol. Ordinary nodes periodically upload their own status information to the controller. Among them, the neighbor node list, TQ value, RSSI, SNR and neighbor node set changes of different links are uploaded to the controller every 5 seconds. Ordinary nodes continuously monitor business information and report business type data to SDN.

[0062] Step S2.2, the SDN controller continuously monitors the packet-in message reported by the Mesh node and determines the message type;

[0063] Step S2.3: If the message received by the SDN controller is Mesh node status information, the adjacency matrix, signal strength matrix, packet loss rate matrix, and mobility matrix of the three links will be maintained according to the reported status information, and finally the comprehensive weight matrix of the three links will be calculated;

[0064] More specifically, in step S2.3.1, the SDN controller needs to update and maintain the state matrices of 1.4G, 2.4G and 5G links respectively according to the link state information reported by the Mesh nodes, where the state matrix includes the adjacency matrix, signal strength matrix, packet loss rate matrix and mobility matrix. Finally, the comprehensive weight matrix W of the three links is calculated according to the updated link state matrix. 1.4G , W 2.4G and W 5G ;

[0065] Step S2.3.2: Elements in the adjacency matrix of three different links and The value is 0 or 1, where L ij Indicates the link adjacency relationship between node i and node j. If the value is 0, it means that there is no link in this frequency band between the nodes. If the value is 1, it means that there is a link in this frequency band between the nodes.

[0066] Step S2.3.3: Elements in the signal strength matrix of three different links and The value of in is the received signal strength of different frequency bands between node i and node j, where is the channel capacity of different frequency bands between node i and node j, is the bandwidth of different frequency bands between node i and node j, is the signal-to-noise ratio of different frequency bands between node i and node j, it can be proved and The value of is between 0 and 1;

[0067] Step S2.3.4: Elements in the mobility matrix of three different links and The value of Among them, M ij represents the mobility index between node i and node j, M n represents the mobility index of node n, represents the set of neighbor nodes of node n in different frequency bands at time t, Represents the set of neighbor nodes of n nodes in different frequency bands at time t-Δt, it can be proved that and The value of is between 0 and 1;

[0068] Step S2.3.5: Elements in the packet loss rate matrix of three different links and D ij represents the packet loss rate indicator between node i and node j, where and Calculated based on the TQ value of the batman-adv protocol, It is automatically evaluated by the self-organizing network module according to the internal algorithm, which can prove and The value of is between 0 and 1;

[0069] Step S2.3.6: Elements in the comprehensive weight matrix of three different links and a+b+c=1, where a, b and c are variable parameters used to control the weights of signal strength, packet loss rate and mobility in link evaluation. Finally, the comprehensive weight matrix W of three different links is obtained: 1.4G , W 2.4G and W 5G ;

[0071] Step S2.4: If the message received by the SDN controller is a service data type information, the comprehensive weight matrix will be adjusted according to the service type, and the E-Dijkstra algorithm will be run to obtain the routing path and link selection scheme, and finally the flow table will be constructed and sent to the Mesh node;

[0072] More specifically, in step S2.4.1, when the SDN controller receives the service data type information, it first determines the service type, and then calculates the comprehensive weight matrix W of the three links according to different service types. 1.4G , W 2.4G and W 5GModify it and run the E-Dijkstra algorithm as input to calculate the routing path and link selection. For example, if the received data type is a high-bandwidth service such as high-definition video, W 5G Properly enlarge (W 5G The matrix as a whole is multiplied by a coefficient greater than 1), so that the route calculated by the E-Dijkstra algorithm will tend to select the 5G link, thereby better meeting the QoS;

[0073] Step S2.4.2: After the SDN controller calculates the final result, it sends the route and link selection to the Mesh node in the form of a flow table. The Mesh node updates the route and executes the corresponding strategy based on the flow table. The Mesh node will continue to listen to the packet-in message sent by the controller. Once the flow table information sent by the controller is detected, the route will be updated and the strategy will be executed immediately according to the flow table.

[0074] The E-Dijkstra algorithm is an extension of the standard Dijkstra algorithm, which is used to handle the routing problems of path selection and frequency selection at the same time. The goal of the E-Dijkstra algorithm is to determine the optimal path between nodes in the network while also selecting the best frequency channel for each hop, thereby minimizing the total cost of network transmission. This algorithm is particularly suitable for this hardware platform that supports multi-band communication;

[0075] The routing algorithm networking maintenance process of the present invention is as follows Figure 4 As shown, the following steps are included:

[0076] Step S3: Network maintenance: During network operation, when a node in the network fails or goes offline, the node mobility management mechanism is combined to enable the routing protocol to quickly reach a convergence state, and the transmission path is adaptively adjusted;

[0077] Specifically, in step S3.1, each Mesh node regularly monitors its own status information, including the transmission quality parameters (such as link quality, throughput, received signal strength (RSSI), signal-to-noise ratio (SNR)) of the currently connected link (such as 1.4G, 2.4G, 5G link), and uploads this information to the centralized SDN controller;

[0078] Step S3.2: When the SDN controller receives the status information (packet-in message) reported by the node, it first determines which frequency band the reported information belongs to, and then calculates the link quality based on the status information. or The calculated link quality is compared with the link quality threshold or If the value is lower than the corresponding threshold, the adjacency matrix value corresponding to the link is Temporarily set to 0;

[0079] Step S3.3: When a link of a node fails, the value corresponding to the link of the node in the adjacency matrix It will be set to 0 to prevent other nodes from continuing to transmit data through this link. At the same time, the controller sets a recovery time limit (200 seconds) for the failed node. If the broadcast packet of the node is received again during this period, the node will not be deleted and will be re-added to the Mesh network.

[0080] Step S3.4: Each Mesh node will calculate its own mobility index Adaptively adjust the broadcast frequency of the OGM broadcast packet. The calculated value is too high, indicating that the relative position of the node is relatively stable. In this case, the broadcast interval of the OGM broadcast packet should be appropriately increased. If the calculated value is too low, it means that the relative position of the node is unstable and is moving rapidly. In this case, the broadcast time interval of the OGM broadcast packet should be appropriately reduced.

[0081] More specifically, in step S3.4, in the mesh network composed of three different frequency bands, the 2.4G and 5G frequency band links are established by broadcasting OGM broadcast packets through the batman-adv routing protocol, so changing the broadcast interval of the OGM broadcast packet by calculating the node mobility index is only used to optimize the 2.4G and 5G frequency band links in the mesh network;

[0082] The broadcast interval of the OGM broadcast packet The unit is ms. Use the batctl tool of the batman-adv protocol to modify the broadcast interval of the OGM broadcast packet. The modification command is batctl meshifbat0 orig_interval

[0083] Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. A software-defined wireless mesh network adaptive networking method based on multi-link collaboration, characterized in that: The following steps are involved: Step S1: Network discovery: Each node discovers neighbor nodes. Each node builds links for different frequency bands, such as 1.4G, 2.4G, and 5G, and calculates the link quality. After the link quality evaluation is qualified, each node is connected through three different frequency band links to form a Mesh network. Step S2: Adaptive networking, introducing SDN controller to communicate with Mesh nodes. Mesh nodes periodically provide status information to SDN controller. SDN controller collects node status information in real time and monitors messages reported by Mesh nodes. Combined with network topology changes and business needs, it selects the optimal route according to different message types and link characteristics. Step S3, network maintenance: during network operation, when a node in the network fails or goes offline, the node mobility management mechanism is combined to enable the routing protocol to quickly reach a convergence state, and the transmission path is adaptively adjusted.

2. According to claim 1, a software-defined wireless mesh network adaptive networking method based on multi-link collaboration is characterized in that: The step S1 comprises the following steps: Step S1.1, each node loads the batman-adv routing protocol internally, and completes the establishment of 2.4G and 5G links through the OGM broadcast packet generated by the protocol; Step S1.2: Each node is internally loaded with a 1.4G self-organizing network module, which starts the Mesh networking mode to realize neighbor discovery and networking, and complete the construction of the 1.4G link; Step S1.3, links in different frequency bands maintain link-related information and calculate link quality in their own way; In 2.4G and 5G links, the link quality is calculated based on the number of OGM broadcast packets sent and received. The link quality is divided into three parts: RQ (receive link status), EQ (loop link status) and TQ (transmit link status). In 1.4G links, the link quality is automatically evaluated by the self-organizing network module according to the internal algorithm.

3. According to claim 2, a software-defined wireless mesh network adaptive networking method based on multi-link collaboration is characterized in that: The step S2 comprises the following steps: Step S2.1, the SDN controller summarizes the status information of all nodes, and the communication link between the SDN controller and the Mesh node is a 1.4G link; Step S2.2, the SDN controller continuously monitors the packet-in message reported by the Mesh node and determines the message type; Step S2.3: If the message received by the SDN controller is Mesh node status information, the adjacency matrix, signal strength matrix, packet loss rate matrix, and mobility matrix of the three links will be maintained according to the reported status information, and finally the comprehensive weight matrix of the three links will be calculated; Step S2.4: If the message received by the SDN controller is business data type information, the comprehensive weight matrix will be adjusted according to the business type, and the E-Dijkstra algorithm will be run to obtain the routing path and link selection plan. Finally, the flow table will be constructed and sent to the Mesh node.

4. According to claim 3, a software-defined wireless mesh network adaptive networking method based on multi-link collaboration is characterized in that: The step S2.3 comprises the following steps: Step S2.3.1: The SDN controller updates and maintains the state matrices of 1.4G, 2.4G and 5G links respectively according to the link state information reported by the Mesh node, and calculates the comprehensive weight matrices W of the three links respectively according to the updated link state matrices. 1.4G , W 2.4G and W 5G ; Step S2.3.2: Elements in the adjacency matrix of three different links and The value is 0 or 1, where L ij Indicates the link adjacency relationship between node i and node j. If the value is 0, it means that there is no link in this frequency band between the nodes. If the value is 1, it means that there is a link in this frequency band between the nodes. Step S2.3.3: Elements in the signal strength matrix of three different links and The value of in is the received signal strength of different frequency bands between node i and node j, where is the channel capacity of different frequency bands between node i and node j, is the bandwidth of different frequency bands between node i and node j, is the signal-to-noise ratio of different frequency bands between node i and node j, it can be proved and The value of is between 0 and 1; Step S2.3.4: Elements in the mobility matrix of three different links and The value of Among them, M ij represents the mobility index between node i and node j, M n represents the mobility index of node n, represents the set of neighbor nodes of node n in different frequency bands at time t, Represents the set of neighbor nodes of n nodes in different frequency bands at time t-Δt, it can be proved that and The value of is between 0 and 1; Step S2.3.5: Elements in the packet loss rate matrix of three different links and D ij represents the packet loss rate indicator between node i and node j, where and Calculated based on the TQ value of the batman-adv protocol, It is automatically evaluated by the self-organizing network module according to the internal algorithm, which can prove and The value of is between 0 and 1; Step S2.3.6: Elements in the comprehensive weight matrix of three different links and a+b+c=1, where a, b and c are variable parameters used to control the weights of signal strength, packet loss rate and mobility in link evaluation. Finally, the comprehensive weight matrix W of three different links is obtained: 1.4G , W 2.4G and W 5G .

5. According to claim 4, a software-defined wireless mesh network adaptive networking method based on multi-link collaboration is characterized in that: The step S2.4 comprises the following steps: Step S2.4.1: When the SDN controller receives the service data type information, it first determines the service type, and then calculates the comprehensive weight matrix W of the three links according to different service types. 1.4G , W 2.4G and W 5G After modification, the E-Dijkstra algorithm is used as input to calculate the routing path and link selection; Step S2.4.2: After the SDN controller calculates the final result, it sends the routing and link selection to the Mesh node in the form of a flow table. The Mesh node updates the routing according to the flow table and executes the corresponding strategy.

6. The method for adaptive networking of software-defined wireless mesh networks based on multi-link collaboration according to claim 5, characterized in that: The step S3 comprises the following steps: Step S3.1, the Mesh node regularly monitors its own status information and uploads it to the SDN controller; Step S3.2: When the SDN controller receives the status information reported by the node, it calculates the link quality of the corresponding frequency band based on the status information. or The calculated link quality is compared with the link quality threshold or If the value is lower than the corresponding threshold, the adjacency matrix value corresponding to the link is Temporarily set to 0; Step S3.3: When a link of a node fails, the value corresponding to the link of the node in the adjacency matrix will be set to 0 to prevent other nodes from continuing to transmit data through this link. At the same time, the controller sets up a recovery mechanism for the failed node; Step S3.4: Each Mesh node calculates its own mobility index Adaptively adjust the broadcast frequency of OGM broadcast packets.

7. The method for adaptive networking of software-defined wireless mesh networks based on multi-link collaboration according to claim 6, characterized in that: In step S3.4, the broadcast interval of the OGM broadcast packet is changed by calculating the node mobility index, which is only used to optimize the 2.4G and 5G frequency band links in the mesh network, wherein the adjustment of the broadcast frequency of the OGM broadcast packet is based on the mobility index of the Mesh node itself, the specific adjustment method of the broadcast time interval is related to the node mobility index, and the broadcast time interval of the OGM broadcast packet is The unit is milliseconds.

8. The method for adaptive networking of software-defined wireless mesh networks based on multi-link collaboration according to claim 7, characterized in that: RQ represents the number of OGM broadcast packets received from the neighbor interface. Taking into account the influence of asymmetric links, the local TQ calculation formula is: TQ local =TQ local ×(1-(1-RQ)).

9. The method for adaptive networking of software-defined wireless mesh networks based on multi-link collaboration according to claim 8, characterized in that: On the basis of local TQ calculation, the influence of link quality on network transmission is further considered, especially the influence of hop count on link quality. Each hop is multiplied by the hop count penalty coefficient hop_penalty, so the calculation formula of protocol TQ is obtained as follows: TQ = TQ local ×TQ OGM ×(1-(1-RQ) 3 )×hop_penalty.

10. A software-defined wireless mesh network adaptive networking system based on multi-link collaboration, characterized in that: The system comprises a processor and a memory, wherein instructions are stored in the memory, and when the instructions are executed by the processor, the system executes the networking method according to any one of claims 1 to 9.

Images