Wireless Mesh network reliability guarantee method based on adaptive network coding

Abstract

The invention provides a wireless Mesh network reliability guarantee method based on adaptive network coding. The method comprises the steps of routing initialization, dynamic weight adjustment, link selection, network coding and data transmission. The method comprises the following steps: firstly, in a routing initialization step, each node discovers a neighbor node and establishes a Mesh network structure; in the dynamic weight adjustment step, a weight matrix is adjusted according to different service types (such as video streams and audio streams) so as to optimize the transmission reliability and the resource utilization rate; in the link selection step, an optimal link path is selected by constructing a state matrix and adopting an E-D algorithm, so that the transmission reliability and efficiency are improved; in the network coding and data transmission steps, the optimal coding degree and redundancy are calculated through a Gaussian process (GP) model so as to adapt to the dynamic change of the network environment and improve the transmission robustness. According to the invention, the software defined network (SDN) controller is utilized to manage the routing in a unified manner, and the link selection strategy can be flexibly adjusted according to service requirements. By collecting the state information of each node in the network, the system adaptively adjusts the coding parameters and weight configuration of the nodes, and selects the optimal route according to the performance of the nodes. According to the method, the packet loss rate and delay in the data packet transmission process are effectively reduced, and the transmission efficiency and reliability of the multi-hop wireless Mesh network are improved.

Description

Technical Field

[0001] The present invention belongs to the field of wireless network communication technology, and in particular relates to a wireless Mesh network reliability assurance method and system based on adaptive network coding. The method realizes efficient and reliable data transmission of a multi-hop wireless Mesh network through cross-layer information perception, dynamic weight adjustment and network coding. Background Art

[0002] With the rapid development of the Internet of Things and wireless communication technologies, application scenarios such as smart homes, smart transportation, and telemedicine are constantly emerging, and users' demand for efficient, stable, and low-latency data transmission on wireless networks is growing. As a multi-hop self-organizing network structure, wireless mesh networks have become an important network architecture to meet these needs due to their high bandwidth, flexibility, and self-healing properties. Wireless mesh networks are widely used in smart cities, emergency communications, industrial monitoring, and other fields, providing reliable network access support for various devices.

[0003] Although wireless mesh networks have certain flexibility and adaptability, they are often unstable due to channel interference, network congestion, and node failure during multi-hop transmission, and cannot meet the requirements of high-quality services (QoS). Traditional routing protocols are difficult to respond quickly and adapt to the needs of different services in a dynamic environment, and often cannot provide sufficient transmission reliability.

[0004] The introduction of software-defined networking (SDN) technology can significantly improve the flexibility and controllability of wireless mesh networks. Through the global view of the SDN controller, the network can dynamically adjust the routing strategy based on information such as service type and link status to adapt to different QoS requirements. However, how to achieve fast and accurate path selection and reliable data transmission in mesh networks is still a key technical problem.

[0005] As an effective technology for anti-interference and improving data transmission efficiency, network coding can further improve the reliability of multi-hop Mesh networks. By introducing network coding technology in data transmission, the packet loss rate can be significantly reduced, the transmission efficiency can be improved, and the robustness of the network in complex environments can be enhanced. Therefore, combining the centralized control advantages of SDN and the efficient transmission characteristics of network coding, developing a wireless Mesh network reliability assurance method based on adaptive network coding is of great significance for improving the transmission performance of multi-hop wireless networks. Summary of the invention

[0006] Purpose of the invention: The invention aims to improve the reliability and transmission efficiency of wireless Mesh networks. By combining software defined networks (SDN) and network coding technology, introducing network dynamics and service demand dynamics, integrating multi-dimensional cross-layer information perception into the routing selection process, and performing on-demand routing selection according to different service types, a wireless Mesh network reliability assurance method based on adaptive network coding is provided.

[0007] To achieve the above objectives, the present invention provides a wireless multi-hop routing fast decision method based on cross-layer information perception, comprising the following steps:

[0008] Step S1, initialization routing and node status discovery step: each node discovers neighboring nodes, builds a Mesh network, and records the status information of each node;

[0009] Step S2, dynamic weight adjustment and comprehensive matrix calculation step: according to different requirements of business types, adjust the weights of each matrix and calculate the comprehensive matrix to provide a basis for link selection;

[0010] Step S3, link selection step: using the ED algorithm to select the optimal transmission path according to the link state matrix;

[0011] Step S4, network coding matrix generation and data coding step: calculate the optimal coding degree k and redundancy r through the Gaussian process (GP) model, generate the coding matrix and encode the data;

[0012] Step S5, data transmission and reception step: data transmission is performed based on the encoded data packet, and decoding is performed at the receiving end to restore the original data.

[0013] A further improvement of the present invention is that step S1 further includes the following steps:

[0014] Step S1.1, each node periodically uploads its own status information to the network controller, including link quality (TQ value), signal strength and adjacency relationship, etc.;

[0015] Step S1.2, the network controller maintains the adjacency matrix, signal strength matrix and TQ value matrix according to the information uploaded by the node;

[0016] Step S1.3: The controller maintains the above matrix on different frequency bands respectively to support link selection requirements of different frequency bands.

[0017] A further improvement of the present invention is that step S2 further includes the following steps:

[0018] Step S2.1, the network controller assigns different weights to the adjacency matrix, signal strength matrix and TQ value matrix according to the service type to reflect the priority of the service demand;

[0019] Step S2.2: For different service types (such as video streaming, audio streaming, and file transfer), the weight matrix is ​​dynamically adjusted to meet the different service requirements for reliability and delay;

[0020] Step S2.3: Calculate a comprehensive matrix based on the adjusted weights to serve as basic data for link selection.

[0021] A further improvement of the present invention is that step S3 further includes the following steps:

[0022] Step S3.1, observe the state matrix of each frequency band to select the frequency band;

[0023] Step S3.2: Use the ED algorithm to evaluate the link state matrix and select the optimal link path that meets the transmission requirements.

[0024] A further improvement of the present invention is that step S4 further includes the following steps:

[0025] Step S4.1, calculating the coding degree k and redundancy r by using a Gaussian process (GP) model to optimize the objective function, so as to determine the key parameters of the network coding;

[0026] Step S4.2, data partitioning, generating a k×k identity matrix and an r×k random matrix, which are used to construct a coding matrix required for network coding;

[0027] Step S4.3, synthesizing a coding matrix for coding the data packet, wherein the coding matrix is ​​obtained by a linear combination of a unit matrix and a random matrix or a matrix operation;

[0028] Step S4.4: Encode the data packet to be transmitted according to the encoding matrix and send it to the destination node.

[0029] A further improvement of the present invention is that step S5 further includes the following steps:

[0030] Step S5.1, after receiving the encoded data packet, the receiving node extracts the packet header information and records the transmission path and link quality of the data packet;

[0031] Step S5.2: If the number of data packets is less than k, re-request to ensure the reliability of data transmission;

[0032] Step S5.3: The receiving node decodes the data packet using the coding matrix to restore the original data.

[0033] A further improvement of the present invention is that in step S4.1, the objective function is: f(x)=max(λ 1 P success -λ 2B-λ 3 t delay ), which contains the following input features: Coding degree k: refers to the number of coding blocks after the data packet is divided, which affects the reliability of transmission; Redundancy r: refers to the number of redundant data packets generated by network coding, which is used to enhance the reliability of transmission; Packet loss rate P loss : Used to calculate the reception success rate: Bandwidth B: used to evaluate the transmission capacity of the link; delay t delay : used to measure the delay performance of the link; weight parameter λ 1 , 2 , 3 : They correspond to the importance of coding degree, redundancy, reception success rate, bandwidth and delay in the objective function, and are used to dynamically adjust the link selection strategy in different business scenarios to meet the needs of specific businesses.

[0034] A further improvement of the present invention is that in step S4, the GP model is used to dynamically calculate the coding degree k and the redundancy r, and these two parameters are optimized by the RBF kernel function and the expected improvement (EI) criterion to ensure the best coding efficiency and reliability under different network conditions and business requirements.

[0035] A further improvement of the present invention is that the calculation of the GP model comprises the following steps:

[0036] Step S4.1.1: Define the kernel function of the GP model as where α 0 and t are the amplitude and length parameters of the kernel function, which are used to measure the similarity between input features;

[0037] Step S4.1.2, assuming that the mean function m(x)=0 to simplify the model calculation;

[0038] Step S4.1.3, based on the maximization of the acquisition function, the optimal values ​​of the coding degree k and the redundancy r are calculated, and the expected improvement (EI) and the expected reduction (ED) are maximized to improve the coding redundancy and transmission reliability of the network.

[0039] In order to achieve the purpose of the invention, the present invention also provides a system for implementing the wireless mesh network reliability assurance method based on adaptive network coding as described above.

[0040] The beneficial effects of the present invention are as follows: the software-defined network technology is adopted to enable the controller to centrally manage routing, and the network coding technology is introduced to dynamically adjust the coding parameters through the Gaussian process model to improve the stability of data transmission. Based on cross-layer information perception and dynamic weight adjustment, the system realizes on-demand routing for different business needs, reduces latency, and improves the reliability and efficiency of data packet transmission. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] Figure 1 This is a rendering of the wireless Mesh network reliability algorithm based on adaptive network coding provided by the present invention.

[0042] Figure 2 It is a specific process diagram of network coding.

[0043] Figure 3 It is a flow chart of the algorithm provided by the present invention, which shows the overall process of routing initialization, maintenance, adaptive coding and data transmission in the network. DETAILED DESCRIPTION

[0044] In order to make the purpose, technical solutions and advantages of the present invention more clear, the present invention is described in detail below with reference to the accompanying drawings and specific embodiments.

[0045] It should be emphasized that in the process of describing the present invention, various formulas and constraints are distinguished by consistent labels, but the use of different labels to mark the same formula and / or constraints is not excluded. The purpose of this setting is to more clearly illustrate the characteristics of the present invention.

[0046] like Figure 1 As shown in the figure, the present invention provides a wireless mesh network reliability algorithm based on adaptive network coding. A network controller is added to the traditional mesh network, and the network global information is collected through the network controller. Then, the network coding technology is introduced to improve the transmission reliability and efficiency of the network. The addition of network coding enables the data packet to carry not only its own information but also some redundant information during the transmission process. Figure 2 As shown, when data packet loss or network congestion occurs, the receiving node can use redundant information to recover data, reducing the number of retransmissions. This method greatly reduces the possibility of data loss and enhances the robustness of the network in complex and dynamic environments. Specifically, the network controller dynamically calculates the optimal coding degree and redundancy of each path based on the collection of node status information of the entire network to meet the needs of different business types. For example, for IoT and voice services with high reliability requirements, the controller will increase redundancy to ensure data integrity; and for video services that require high throughput, it will maintain a certain reliability while improving transmission efficiency. In this way, through the combination of network coding and on-demand routing, the present invention effectively solves the common packet loss, delay and network congestion problems in multi-hop wireless mesh networks, and comprehensively guarantees the quality of service (QoS) of various types of services.

[0047] The adaptive network coding algorithm process described in the present invention is as follows Figure 2 As shown, the following steps are included:

[0048] The first part is the initialization routing, maintenance and link selection part:

[0049] Step S1, initialization routing and node status discovery step: each node discovers neighboring nodes, builds a Mesh network, and records the status information of each node; specifically:

[0050] Step S1.1, each node periodically uploads its own status information to the network controller, including link quality (TQ value), signal strength and adjacency relationship, etc.;

[0051] Step S1.2, the network controller maintains the adjacency matrix, signal strength matrix and TQ value matrix according to the information uploaded by the node;

[0052] Step S1.3: The controller maintains the above matrix on different frequency bands respectively to support link selection requirements of different frequency bands.

[0053] Step S2, dynamic weight adjustment and comprehensive matrix calculation step: according to different requirements of business types, adjust the weights of each matrix and calculate the comprehensive matrix to provide a basis for link selection; specifically:

[0054] Step S2.1, the network controller assigns different weights to the adjacency matrix, signal strength matrix and TQ value matrix according to the service type to reflect the priority of the service demand;

[0055] Step S2.2: For different service types (such as video streaming, audio streaming, and file transfer), the weight matrix is ​​dynamically adjusted to meet the different service requirements for reliability and delay;

[0056] Step S2.3: Calculate a comprehensive matrix based on the adjusted weights to serve as basic data for link selection.

[0057] Step S3, link selection step: using the ED algorithm to select the optimal transmission path according to the link state matrix; specifically:

[0058] Step S3.1, observe the state matrix of each frequency band to select the frequency band;

[0059] Step S3.2: Use the ED algorithm to evaluate the link state matrix and select the optimal link path that meets the transmission requirements.

[0060] The second part is the adaptive network coding part:

[0061] Step S4, network coding matrix generation and data coding step: the coding degree k and redundancy r are calculated by Gaussian process (GP) model to optimize the objective function, generate a coding matrix and encode the data; specifically:

[0062] Step S4.1, calculating the coding degree k and redundancy r by using a Gaussian process (GP) model so that the objective function is optimized to determine the key parameters of network coding;

[0063] When the controller receives a transmission request from a source node, it first determines the transmission service type, and then determines whether the node already has coding degree and redundancy. If not, the initialized coding degree and redundancy are directly sent to the source node; otherwise, the controller selects the feasible range of coding degree and redundancy and the threshold of the objective function according to the service type. If the objective function value f(x)=max(λ 1 P success -λ 2 B-λ 3 t delay ) reaches the threshold, then the information of maintaining the original coding degree and redundancy is directly sent down; otherwise, the Gaussian process (GP) model is used to calculate the coding degree k and redundancy r. The specific process is to calculate the acquisition function values ​​of all possible points based on the current Gaussian process model (different feasible sets are selected according to different business types), find the largest point as the output (this point has the greatest possibility of improving the objective function), send the found maximum expected coding degree and redundancy to the source node, record its objective function value after the data transmission is completed, and based on the current transmission result (X new ,f new ), update the Gaussian process model.

[0064] Step S4.2, data partitioning, generating a k×k unit matrix and an r×k random matrix, which are used to construct the coding matrix required for network coding;

[0065] After the source node receives the data stream from the terminal, it divides the data into blocks according to the data threshold T for encoding and transmission. The source node continuously receives data and puts the received data into a buffer. Whenever the data in the buffer reaches the threshold T, it extracts T bytes of data and regards it as a data block. The data block is further divided into multiple original data packets according to the size of the data packet. When the source node receives the message sent by the controller, it generates a k×k unit matrix according to the coding degree and an r×k random matrix according to the redundancy to encode the data packet. The data range in the random matrix is ​​GF(256).

[0066] Step S4.3, synthesizing a coding matrix for coding the data packet, wherein the coding matrix is ​​obtained by a linear combination of a unit matrix and a random matrix or a matrix operation;

[0067] Step S4.4: Encode the data packet to be transmitted according to the encoding matrix and send it to the destination node.

[0068] The data packets to be transmitted are encoded according to the coding matrix, and the header of each coded packet needs to be attached with metadata (coding coefficients (row vectors of matrix G), packet sequence number, data block sequence number) before it can be transmitted.

[0069] The third part is the data transmission and reception part:

[0070] Step S5, data transmission and reception step: data transmission is performed on the basis of the encoded data packet, and decoding is performed at the receiving end to restore the original data; specifically:

[0071] Step S5.1, after receiving the encoded data packet, the receiving node extracts the packet header information and records the transmission path and link quality of the data packet;

[0072] Step S5.2: If the number of data packets is less than k, re-request to ensure the reliability of data transmission;

[0073] Step S5.3: The receiving node decodes the data packet using the coding matrix to restore the original data.

[0074] Embodiment 2: A wireless Mesh network reliability assurance system based on adaptive network coding, the system comprising an SDN controller and distributed nodes, each node having three wireless communication modules (1.4G, 2.4G, 5G); the controller is used to collect and summarize the status information of each node, select the optimal link path by constructing a state matrix and combining the ED algorithm, and generate a network coding matrix by calculating the coding degree and redundancy through a Gaussian process model, so as to realize the network coding and link selection strategy driven by the service type.

[0075] The above embodiments are only used to illustrate the technical solution of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solution of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical solution of the present invention.

Claims

1. A wireless mesh network reliability assurance method based on adaptive network coding, wherein the assurance system includes an SDN controller and distributed nodes, each node has three wireless communication modules; characterized in that: The following steps are involved: Step S1, initialization routing and node status discovery step: each node discovers neighboring nodes, builds a Mesh network, and records the status information of each node; Step S2, dynamic weight adjustment and comprehensive matrix calculation step: according to different requirements of business types, adjust the weights of each matrix and calculate the comprehensive matrix to provide a basis for link selection; Step S3, link selection step: using the ED algorithm to select the optimal transmission path according to the link state matrix; Step S4, network coding matrix generation and data coding step: calculate the optimal coding degree k and redundancy r through the Gaussian process (GP) model, generate the coding matrix and encode the data; Step S5, data transmission and reception step: data transmission is performed based on the encoded data packet, and decoding is performed at the receiving end to restore the original data.

2. The wireless mesh network reliability assurance method based on adaptive network coding according to claim 1 is characterized in that: Step S1 also includes the following sub-steps: Step S1.1, each node periodically uploads its own status information to the network controller, including link quality (TQ value), signal strength and adjacency; Step S1.2, the network controller maintains the adjacency matrix, signal strength matrix and TQ value matrix according to the information uploaded by the node; Step S1.3: The controller maintains the above matrix on different frequency bands respectively to support link selection requirements of different frequency bands.

3. The wireless mesh network reliability assurance method based on adaptive network coding according to claim 2 is characterized in that: Step S2 includes the following steps: Step S2.1, the network controller assigns different weights to the adjacency matrix, signal strength matrix and TQ value matrix according to the service type to reflect the priority of the service demand; Step S2.2: For different service types, including video streaming, audio streaming, and file transfer, the weight matrix is ​​dynamically adjusted to meet the different service requirements for reliability and latency; Step S2.3: Calculate a comprehensive matrix based on the adjusted weights to serve as basic data for link selection.

4. The wireless mesh network reliability assurance method based on adaptive network coding according to claim 3 is characterized by: Step S3 also includes the following steps: Step S3.1, observe the state matrix of each frequency band to select the frequency band; Step S3.2: Use the ED algorithm to evaluate the link state matrix and select the optimal link path that meets the transmission requirements.

5. The wireless mesh network reliability assurance method based on adaptive network coding according to claim 4 is characterized in that: Step S4 also includes the following steps: Step S4.1, calculating the coding degree k and redundancy r by using a Gaussian process (GP) model to optimize the objective function, so as to determine the key parameters of the network coding; Step S4.2, data partitioning, generating a k×k identity matrix and an r×k random matrix, which are used to construct a coding matrix required for network coding; Step S4.3, synthesizing a coding matrix for coding the data packet, wherein the coding matrix is ​​obtained by a linear combination of a unit matrix and a random matrix or a matrix operation; Step S4.4: Encode the data packet to be transmitted according to the encoding matrix and send it to the destination node.

6. The wireless mesh network reliability assurance method based on adaptive network coding according to claim 5 is characterized by: Step S5 also includes the following steps: Step S5.1, after receiving the encoded data packet, the receiving node extracts the packet header information and records the transmission path and link quality of the data packet; Step S5.2: If the number of data packets is less than k, re-request to ensure the reliability of data transmission; Step S5.3: The receiving node decodes the data packet using the coding matrix to restore the original data.

7. The wireless mesh network reliability assurance method based on adaptive network coding according to claim 4 is characterized in that: The objective function in step S4.1 is: f(x) = max(λ1P success -λ2B-λ3t delay ), which contains the following input features: encoding Degree k: refers to the number of coding blocks after the data packet is divided, which affects the reliability of transmission; redundancy r: refers to the number of redundant data packets generated by network coding, which is used to enhance the reliability of transmission; packet loss rate P loss :Used to calculate the reception success rate: Bandwidth B: used to evaluate the transmission capacity of the link; delay t delay : Used to measure the delay performance of the link; weight parameters λ1, λ2, λ3: correspond to the importance of coding degree, redundancy, reception success rate, bandwidth and delay in the objective function, respectively, and are used to dynamically adjust the link selection strategy in different business scenarios to meet the needs of specific businesses.

8. The wireless mesh network reliability assurance method based on adaptive network coding according to claim 5, characterized in that: In step S4, the GP model is used to dynamically calculate the coding degree k and redundancy r, and these two parameters are optimized through the RBF kernel function and the expected improvement (EI) criterion to ensure the best coding efficiency and reliability under different network conditions and service requirements.

9. The wireless mesh network reliability assurance method based on adaptive network coding according to claim 8, characterized in that: The calculation of the GP model includes the following steps: Step S4.1.1: Define the kernel function of the GP model as Where α0 and t are the amplitude and length parameters of the kernel function, which are used to measure the similarity between input features; Step S4.1.2, assuming that the mean function m(x)=0 to simplify the model calculation; Step S4.1.3, based on the maximization of the acquisition function, the optimal values ​​of the coding degree k and the redundancy r are calculated, and the expected improvement (EI) and the expected reduction (ED) are maximized to improve the coding redundancy and transmission reliability of the network.

10. A wireless mesh network reliability assurance system based on adaptive network coding, characterized in that: It includes a device and a controller for executing any one of the methods of claims 1-9, wherein the controller is used to collect and summarize the status information of each node, select the optimal link path by constructing a state matrix and combining it with an ED algorithm, and generate a network coding matrix through the coding degree and redundancy calculated by a Gaussian process model, so as to implement a business type-driven network coding and link selection strategy.

Images