Multi-path QoS (Quality of Service) guarantee WSN (Wireless Sensor Network) routing control method based on grey decision

Abstract

The invention discloses a grey decision-based multipath QoS (Quality of Service) guarantee WSN (Wireless Sensor Network) routing control method, which is used for improving the reliability and integrity of data transmission, and comprises the following steps of: trust evaluation of cluster head nodes, construction of a plurality of paths between a source node and a base station, and selection of a standby routing path with the highest QoS index in a grey decision mode. According to the method, a certain improvement is made in the aspect of multiple paths, the gray decision method has a good effect in multi-factor decision, the path decision is used as a gray system process, a linguistic variable is set to represent a gray number, and factors such as transmission delay and energy consumption are subjected to different ratings to complete the decision. The gray decision model can select a plurality of standby routing paths with high quality, so that the energy consumption of the network can be effectively reduced, the conditions of channel blockage and the like can be avoided, and the threat of malicious nodes to data in the network communication process is reduced.

Description

Technical Field

[0001] The invention belongs to the field of wireless sensor network routing security, and in particular relates to a multi-path QoS guaranteed WSN routing control method based on grey decision. Background Art

[0002] Wireless Sensor Network (WSN) is a distributed sensing system consisting of a large number of low-cost, miniature sensor nodes that are interconnected through wireless communication and work together to monitor and collect various information in the environment. These nodes usually have limited computing power, storage space and battery life. They are deployed in the monitoring area and use built-in sensors to sense various physical phenomena such as temperature, humidity, light, sound, vibration, etc.

[0003] In view of the security and energy consumption issues in routing control of wireless sensor networks, domestic and foreign researchers have proposed many improvement methods. Single-path routing protocols may face challenges in some cases. For example, when network conditions are unstable or under security threats, such protocols may find it difficult to ensure the reliability and integrity of data transmission. For example, packet loss or external interference may cause transmission failure, and the base station lacks sufficient real-time information to make necessary adjustments in this case. In contrast, multi-path routing protocols have potential advantages. They can disperse data traffic to multiple paths, thereby reducing the load of a single path and avoiding network congestion. If a path fails, the data can be automatically switched to other backup paths, improving the reliability of data transmission. Therefore, in order to better ensure the quality of service (QoS), more and more protocols have begun to adopt multi-path routing strategies. Gerla et al. proposed a routing protocol based on ant colony algorithm. The protocol detects multiple paths by sending ant detection packets, which are screened by the base station, while increasing the bandwidth of the main channel to prevent channel congestion and ensure service quality. The routing protocol designed by B.Deb et al. improves the reliability of the network. This protocol can effectively deal with the end-to-end security issues that other protocols may face. This protocol pays special attention to the importance of data in the data packet and makes corresponding adjustments when there are problems with channel transmission. To enhance network stability, the protocol allows source nodes to transmit data in parallel through multiple paths when sending data to the base station, but it does not fully consider the energy consumption issue.

[0004] In summary, designing a safe and efficient routing control method has become a research hotspot in wireless sensor networks in recent years. How to balance energy consumption, ensure service quality, and take into account node security has become an urgent problem to be solved in this field. Summary of the invention

[0005] In order to solve the technical problems mentioned in the above background technology, the present invention proposes a multi-path QoS guaranteed WSN routing control method based on grey decision-making.

[0006] In order to achieve the above technical objectives, the technical solution of the present invention is:

[0007] A multi-path QoS guaranteed WSN routing control method based on grey decision-making includes the following steps:

[0008] (1) The base station adopts a trust evaluation mechanism to identify malicious nodes in the network. Through the trust evaluation model, the trust of sensor nodes in the network is comprehensively scored for cluster head selection, and malicious nodes are identified and removed.

[0009] (2) Each cluster head selects n backup paths and transmits the information to the source node. The source cluster head generates explore num *n detection data packets are sent along n backup paths for path detection.

[0010] (3) After receiving the path detection data packet, the base station uses the gray decision method to comprehensively evaluate all backup paths, obtains m paths according to the value ranking from high to low, and calculates the energy loss of these m paths. If the energy consumption of a path is greater than the average energy consumption of each backup path, the path will not be enabled. After screening, s available backup paths are obtained, and the base station sends these path information to the corresponding cluster head.

[0011] (4) After receiving all the data in the cluster, the cluster head performs data fusion and then sends it to the next hop cluster head along the optimal path until it reaches the base station. If a path fails, the nearest node in the remaining backup paths is found as the next hop.

[0012] Furthermore, in step (1), the trust evaluation mechanism identifies malicious nodes as follows:

[0013] (101) The base station first performs an in-depth analysis of the collected data packets to identify and classify those data packets that have not been successfully received into corresponding clusters. For those clusters with a data loss rate exceeding 10%, the base station will include them in the list of suspected clusters.

[0014] (102) The base station performs a one-to-one trust evaluation on the cluster head node of the suspected cluster. If the cluster head node is confirmed to be reliable, it will be authorized to perform trust evaluation on its cluster members and transmit the results back to the base station. If the cluster head node fails the trust evaluation and is judged to be suspicious or malicious, the neighboring nodes will jointly perform a many-to-one trust evaluation on it.

[0015] (103) The base station calculates the comprehensive trust value of each node and identifies malicious nodes. The formula is as follows:

[0016]

[0017] Among them, w1 = 0.7, w2 = 0.3, TTL = 1. In the calculation of the comprehensive trust of the node, if λ ≥ γ, it means that the node is evaluated as a trusted node, if λ ≤ ξ, it means that it is a malicious node, and the rest indicates that the node is suspected. γ and ξ refer to the trusted trust threshold and the malicious trust threshold respectively. i,j (t) represents the number of packets successfully received by evaluation node i from suspected malicious node j, L i,j (t) represents the number of packets that the evaluation node i has not successfully received from the suspected malicious node j. ETy is a number that takes the value of 0 or 1, which is 1 when the value of the node is greater than the threshold and 0 otherwise.

[0018] Furthermore, in step (2), the steps of selecting n backup paths are as follows:

[0019] (201) Construct a path decision weight matrix D[i][j] to express the path weight between two nodes, i, j = 0, 1, 2, ..., n. It is stipulated that the cluster head node data is transmitted from the outer layer to the inner layer, and the path weight between each node is calculated by the following formula:

[0020]

[0021] in, is the energy threshold in the energy model, and d is the distance from node i to node j. resj is the residual energy of node j, a, b, c, d and m, n, p, q are all weight coefficients. θ is the angle between sensor node i and j, and α is the angle between sensor node j and the base station. level is the level label of each sensor node, recording the layer where the node is located. The cluster head node transfers data from the outermost layer to the innermost layer, and selects the node with the highest decision value as the next hop routing node.

[0022] (202) Adjust the path weight matrix. For each edge in the determined optimal path, adjust its weight value to infinity, so as to ensure that these edges will not be selected again when searching for alternative paths.

[0023] (203) Each source cluster head node has an independent path decision matrix, and a set of backup paths is independently generated for each source cluster head node so that data can be transmitted effectively when the main path is unavailable.

[0024] Furthermore, in step (3), the specific steps of using the grey decision-making method to evaluate the backup path are as follows:

[0025] (301) Let S = {S1, S2, ..., S n} are m independent alternative paths, Q = {Q1, Q2, ..., Q n} is a set of n attributes of alternative paths, ω={ω1,ω2,...,ω n} is the weight vector of each attribute. In order to select an alternative routing path with high service quality, the three parameters of transmission delay, energy consumption and packet loss rate are passed into the decision model. The decision maker directly gives the weight of the attribute. Assuming there are k decision makers, Q j The attribute weights can be described as:

[0026]

[0027] Among them, w j k (j=1,2,3,...,n) is the attribute weight of the kth decision maker, represented by gray numbers.

[0028] (302) The attribute is converted into a language variable according to the membership function to generate an attribute rating value. j The calculation formula is:

[0029]

[0030] Among them, G ij k (i=1,2,...,m,j=1,2,...,n) is the attribute rating value of the k-th decision maker and can be represented by the grey number express.

[0031] (303) Establish a grey decision matrix:

[0032]

[0033] (304) Normalized grey decision matrix:

[0034]

[0035] Benefit attributes It can be expressed as:

[0036]

[0037] Cost Attributes It can be expressed as:

[0038]

[0039] Normalize the data so that the normalized gray number is in the range of [0,1];

[0040] (305) Create a grey weighted normalized decision matrix. The weighted normalized decision matrix can be described by the following matrix:

[0041]

[0042] in

[0043] (306) Take the ideal solution as an alternative, and for m alternative paths S = {S1, S2, ..., S m}, the ideal alternative path is:

[0044]

[0045] (307) Calculate the ideal alternative path S max Grey possibility degree between the candidate backup path S and the comparison:

[0046]

[0047] (308) Sort the candidate backup paths. i ≤S max The smaller the value is, the higher the ranking of the candidate backup path is, and vice versa.

[0048] The beneficial effects brought by adopting the above technical solution are:

[0049] (1) The present invention introduces a trust evaluation mechanism and a multi-path transmission concept. By giving priority to cluster head nodes with higher trust values, the possibility of malicious nodes becoming cluster heads is avoided, and the utilization efficiency of backup paths is improved. When the main path fails, it can be quickly adjusted, greatly enhancing the robustness of network transmission.

[0050] (2) Based on the single hop between clusters, the present invention adopts the gray decision method. According to the node transmission delay, node energy and packet loss rate of each cluster area, three parameters are input into the decision model. Combined with the idea of ​​backup multi-path, when the path is unavailable, the multi-path fragment is utilized to select the next hop node with a shorter distance, which effectively reduces the transmission delay and improves the data packet transmission rate, thus achieving a trade-off between the three QoS service indicators of packet loss rate, transmission delay and energy consumption. BRIEF DESCRIPTION OF THE DRAWINGS

[0051] Figure 1 It is a schematic diagram of cluster head node trust evaluation in the present invention;

[0052] Figure 2 It is a schematic diagram of multi-path transmission in the present invention; DETAILED DESCRIPTION

[0053] The technical solution of the present invention will be described in detail below with reference to the accompanying drawings.

[0054] A multi-path QoS guaranteed WSN routing control method based on grey decision-making includes the following steps:

[0055] (1) The base station adopts a trust evaluation mechanism to identify malicious nodes in the network. Through the trust evaluation model, the trust of sensor nodes in the network is comprehensively scored for cluster head selection, and malicious nodes are identified and removed.

[0056] (2) Each cluster head selects n backup paths and transmits the information to the source node. The source cluster head generates explore num *n detection data packets are sent along n backup paths for path detection.

[0057] (3) After receiving the path detection data packet, the base station uses the gray decision method to comprehensively evaluate all backup paths, obtains m paths according to the value ranking from high to low, and calculates the energy loss of these m paths. If the energy consumption of a path is greater than the average energy consumption of each backup path, the path will not be enabled. After screening, s available backup paths are obtained, and the base station sends these path information to the corresponding cluster head.

[0058] (4) After receiving all the data in the cluster, the cluster head performs data fusion and then sends it to the next hop cluster head along the optimal path until it reaches the base station. If a path fails, the nearest node in the remaining backup paths is found as the next hop.

[0059] In this embodiment, the following preferred solution can be used to implement the above (1):

[0060] (101) The base station first performs an in-depth analysis of the collected data packets to identify and classify those data packets that have not been successfully received into corresponding clusters. For those clusters with a data loss rate exceeding 10%, the base station will include them in the list of suspected clusters.

[0061] (102) The base station performs a one-to-one trust evaluation on the cluster head node of the suspected cluster. If the cluster head node is confirmed to be reliable, it will be authorized to perform trust evaluation on its cluster members and transmit the results back to the base station. If the cluster head node fails the trust evaluation and is judged to be suspicious or malicious, the neighboring nodes will jointly perform a many-to-one trust evaluation on it.

[0062] (103) The base station updates a list of trusted cluster heads, which includes cluster heads whose data loss rate in the current communication cycle is lower than the preset standard. For those cluster heads not included in the list, the system will arrange them into a queue of cluster heads to be evaluated from near to far according to their physical distance from the base station. The evaluation process will start with the cluster head closest to the base station in the order of the queue and evaluate the trust one by one. When the distance between the base station and the cluster head is less than d0 of the first-order radio model in Chapter 3, the base station will directly send data packet forwarding instructions to them to perform trust evaluation. For cluster heads to be evaluated that are located d0 away from the base station, due to the unlimited energy of the base station, the base station directly sends a packet forwarding request to the cluster head to be evaluated. At the same time, the base station will refer to the cluster heads in the trusted cluster head list to build a reliable transmission path, and send the path information together with the packet forwarding request to the cluster head to be evaluated. The cluster head to be evaluated transmits the information required by the base station to the base station in a multi-hop routing manner according to the trusted transmission path given by the base station, such as Figure 1 shown.

[0063] (104) The base station calculates the comprehensive trust value of each node and identifies malicious nodes. The formula is as follows:

[0064]

[0065] Among them, w1 = 0.7, w2 = 0.3, TTL = 1. In the calculation of the comprehensive trust of the node, if λ ≥ γ, it means that the node is evaluated as a trusted node, if λ ≤ ξ, it means that it is a malicious node, and the rest indicates that the node is suspected. γ and ξ refer to the trusted trust threshold and the malicious trust threshold respectively. i,j (t) represents the number of packets successfully received by evaluation node i from suspected malicious node j, L i,j (t) represents the number of packets that the evaluation node i has not successfully received from the suspected malicious node j. ETy is a number that takes the value of 0 or 1, which is 1 when the value of the node is greater than the threshold and 0 otherwise.

[0066] In this embodiment, the following preferred solution can be used to implement the above (2):

[0067] (201) Construct a path decision weight matrix D[i][j] to express the path weight between two nodes, i, j = 0, 1, 2, ..., n. It is stipulated that the cluster head node data is transmitted from the outer layer to the inner layer, and the path weight between each node is calculated by the following formula:

[0068]

[0069] in, is the energy threshold in the energy model, and d is the distance from node i to node j. resjis the residual energy of node j, a, b, c, d and m, n, p, q are all weight coefficients. θ is the angle between sensor node i and j, and α is the angle between sensor node j and the base station. level is the level label of each sensor node, recording the layer where the node is located. The cluster head node transfers data from the outermost layer to the innermost layer, and selects the node with the highest decision value as the next hop routing node.

[0070] (202) Adjust the path weight matrix. For each edge in the determined optimal path, adjust its weight value to infinity, so as to ensure that these edges will not be selected again when searching for alternative paths.

[0071] (203) Each source cluster head node has an independent path decision matrix, and a set of backup paths is independently generated for each source cluster head node so that data can be transmitted effectively when the main path is unavailable.

[0072] In this embodiment, the following preferred solution can be used to implement the above (3):

[0073] (301) Let S = {S1, S2, ..., S n} are m independent alternative paths, Q = {Q1, Q2, ..., Q n} is a set of n attributes of alternative paths, ω={ω1,ω2,...,ω n} is the weight vector of each attribute. In order to select an alternative routing path with high service quality, the three parameters of transmission delay, energy consumption and packet loss rate are passed into the decision model. The decision maker directly gives the weight of the attribute. Assuming there are k decision makers, Q j The attribute weights can be described as:

[0074]

[0075] Among them, w k j (j=1,2,3,...,n) is the attribute weight of the kth decision maker, represented by gray numbers.

[0076] (302) The attribute is converted into a language variable according to the membership function to generate an attribute rating value. j The calculation formula is:

[0077]

[0078] Among them, G ij k (i=1,2,...,m,j=1,2,...,n) is the attribute rating value of the k-th decision maker and can be represented by the grey number express.

[0079] (303) Establish a grey decision matrix:

[0080]

[0081] (304) Normalized grey decision matrix:

[0082]

[0083] Benefit attributes It can be expressed as:

[0084]

[0085] Cost Attributes It can be expressed as:

[0086]

[0087] Normalize the data so that the normalized gray number is in the range of [0,1];

[0088] (305) Create a grey weighted normalized decision matrix. The weighted normalized decision matrix can be described by the following matrix:

[0089]

[0090] in

[0091] (306) Take the ideal solution as an alternative, and for m alternative paths S = {S1, S2, ..., S m}, the ideal alternative path is:

[0092]

[0093] (307) Calculate the ideal alternative path S max Grey possibility degree between the candidate backup path S and the comparison:

[0094]

[0095] (308) Sort the candidate backup paths. i ≤S max The smaller the value is, the higher the ranking of the candidate backup path is, and vice versa.

[0096] In this embodiment, the following preferred solution can be used to implement the above (4):

[0097] (401) Figure 2As shown in Figure 1, during data transmission, if the data transmission from node A to node B fails, according to the backup multipath idea of ​​this paper, it should be resent from the source node. However, this wastes the energy of the source node transmitting to node A. Therefore, an improvement is made. When the transmission fails, the node closest to the failed node in the backup path set is found. Figure 2 Node C is selected as the next hop node. This method can greatly reduce energy consumption. At the same time, even if the main path fails, the backup path can ensure reliable data transmission.

Claims

1. A multi-path QoS guaranteed WSN routing control method based on grey decision making, characterized in that: The following steps are involved: (1) The base station uses a trust evaluation mechanism to identify malicious nodes in the network. Through the trust evaluation model, the trust of sensor nodes in the network is comprehensively scored for cluster head selection. Malicious nodes are identified and removed. The specific steps are as follows: (101) The base station first conducts an in-depth analysis of the collected data packets to identify and classify those data packets that have not been successfully received into corresponding clusters. For those clusters with a data loss rate exceeding 10%, the base station will list them in the suspect cluster list; (102) The base station performs a one-to-one trust evaluation on the cluster head node of the suspected cluster. If the cluster head node is confirmed to be reliable, it will be authorized to perform trust evaluation on its cluster members and transmit the result back to the base station. If the cluster head node fails the trust evaluation and is judged to be suspicious or malicious, the neighboring nodes will jointly perform a many-to-one trust evaluation on it. (103) The base station calculates the comprehensive trust value of each node and identifies malicious nodes. The formula is as follows: Among them, w1=0.7, w2=0.3, TTL=1. In the calculation of the comprehensive trust of the node, if λ≥γ, it means that the node is evaluated as a trusted node. If λ≤v represents a malicious node, the rest indicates that the node is suspected. γ and ξ refer to the trusted trust threshold and the malicious trust threshold respectively. f i,j (t) represents the number of packets successfully received by evaluation node i from suspected malicious node j, L i,j (t) represents the number of packets that the evaluation node i has not successfully received from the suspected malicious node j. ETy is a number with a value of 0 or 1. It is 1 when the value of the node is greater than the threshold, otherwise it is 0; (2) Each cluster head selects n backup paths and transmits the information to the source node. The source cluster head generates explore num *n detection data packets are sent along n backup paths for path detection. The steps for selecting backup paths are as follows: (201) Construct a path decision weight matrix D[i][j] to express the path weight between two nodes, i, j = 0, 1, 2, ..., n. It is stipulated that the cluster head node data is transmitted from the outer layer to the inner layer, and the path weight between each node is calculated by the following formula: in, is the energy threshold in the energy model, d is the distance from node i to j, E resj is the residual energy of node j, a, b, c, d and m, n, p, q are all weight coefficients, θ is the angle between sensor node i and j, α is the angle between sensor node j and base station, level is the level label of each sensor node, recording the layer where the node is located, the cluster head node transfers data from the outermost layer to the innermost layer, and selects the node with the highest decision value as the next hop routing node; (202) Adjust the path weight matrix, and adjust the weight value of each edge in the determined optimal path to infinity, so as to ensure that these edges will not be selected again when searching for alternative paths; (203) Each source cluster head node has an independent path decision matrix, and a set of backup paths are independently generated for each source cluster head node so that data can be effectively transmitted when the primary path is unavailable; (3) After receiving the path detection data packet, the base station uses the grey decision method to comprehensively evaluate all backup paths, obtains m paths according to the value ranking from high to low, and calculates the energy loss of these m paths. If the energy consumption of a path is greater than the average of the energy consumption of each backup path, the path is not enabled. After screening, s available backup paths are obtained. The base station sends these path information to the corresponding cluster head. The specific steps of using grey decision to evaluate backup paths are as follows: (301) Let S = {S1, S2, ..., S n } are m independent alternative paths, Q = {Q1, Q2, ..., Q n } is a set of n attributes of alternative paths, ω={ω1,ω2,...,ω n } is the weight vector of each attribute. In order to select an alternative routing path with high service quality, the three parameters of transmission delay, energy consumption and packet loss rate are passed into the decision model. The decision maker directly gives the weight of the attribute. Assuming there are k decision makers, Q j The attribute weights can be described as: Among them, w j k (j=1,2,3,...,n) is the attribute weight of the kth decision maker, represented by gray numbers. (302) The attribute is converted into a language variable according to the membership function to generate an attribute rating value. j The calculation formula is: Among them, G ij k (i=1,2,...,m,j=1,2,...,n) is the attribute rating value of the k-th decision maker and can be represented by the grey number express; (303) Establish a grey decision matrix: (304) Normalized grey decision matrix: Benefit attributes It can be expressed as: Cost Attributes It can be expressed as: Normalize the data so that the normalized gray number is in the range of [0,1]; (305) Create a grey weighted normalized decision matrix. The weighted normalized decision matrix can be described by the following matrix: in (306) Take the ideal solution as an alternative, and for m alternative paths S = {S1, S2, ..., S m }, the ideal alternative path is: (307) Calculate the ideal alternative path S max Grey possibility degree between the candidate backup path S and the comparison: (308) Sort the candidate backup paths. i ≤S max The smaller the value is, the higher the ranking of the candidate backup path is, otherwise, the lower the ranking is; (4) After receiving all the data in the cluster, the cluster head performs data fusion and then sends it to the next hop cluster head along the optimal path until it reaches the base station. If a path fails, the nearest node in the remaining backup paths is found as the next hop.

Images