Method and device for robust routing protocol in wireless sensor networks based on game theory and particle swarm optimization algorithm

Abstract

A method and device for a robust routing protocol for wireless sensor networks based on game theory and particle swarm optimization algorithm are proposed. The method adopts game theory and particle swarm optimization algorithm to select cluster heads, and combines the strategy of reusing faulty sensors to achieve low-energy consumption, long-life, and highly robust wireless sensor network routing. The method includes: A. deriving the theoretical optimal number of cluster heads based on the network model and communication energy consumption model; B. adopting a non-cooperative game method to select candidate cluster heads from sensors; C. adopting a particle swarm optimization algorithm to determine the final cluster head; D. Sensors are clustered in the nearest place, and the network enters the stable data transmission stage; E. deriving a strategy of reusing faulty sensors. When there is a sensor with a soft permanent fault, the faulty sensor is used as a relay node to achieve low-energy consumption, long-life, and highly robust wireless sensor network routing.

Description

Technical Field

[0001] The present invention belongs to the technical field of routing protocols, and in particular relates to a method and device for a robust routing protocol for a wireless sensor network based on game theory and a particle swarm optimization algorithm. Background Art

[0002] Structural health monitoring (SHM), enabling in-depth analysis of structural performance, is a key technology for ensuring the safety of civil engineering structures. Compared to traditional wired monitoring networks, wireless sensor networks (WSNs) have gained widespread application in SHM due to their low cost and flexible installation methods, providing effective monitoring data for structural condition assessment. However, due to energy consumption and harsh environmental conditions such as high temperature and corrosion, sensor nodes may run out of power or fail, impacting network stability and lifespan. Therefore, extending the network lifespan has become a key technology in WSN research.

[0003] Data transmission and reception are the primary sources of energy consumption in sensor nodes, while network routing protocols primarily focus on the paths used to transmit and receive data. Therefore, designing efficient routing protocols is crucial for reducing node energy consumption. Cluster routing protocols, due to their excellent scalability and low energy consumption, have become a research focus in WSNs. In existing cluster routing protocols, sensor nodes are divided into multiple clusters. Each cluster head coordinates the monitoring tasks of its members and, after receiving and aggregating the monitoring data from cluster members, forwards it to a base station or sink node. While these cluster routing protocols improve network lifespan to some extent, they still fail to fully consider factors such as sensor residual energy and the impact of sensor failures on the network, resulting in uneven node energy distribution and high overall network energy consumption. Therefore, it is necessary to develop a robust routing protocol that comprehensively considers factors such as residual energy, coverage, and node degree, and offers low-cost fault tolerance, in order to extend the network lifespan of WSNs. Summary of the Invention

[0004] The present invention aims to overcome the above-mentioned deficiencies of the prior art and provide a method and apparatus for a robust routing protocol for wireless sensor networks (FTRP-GPSO) based on game theory and particle swarm optimization algorithm to reduce the energy consumption of WSNs, extend the network life, and enhance the network robustness.

[0005] A first aspect of the present invention includes a method for a robust routing protocol for wireless sensor networks based on game theory and particle swarm optimization algorithm, comprising the following steps:

[0006] A. Derive the theoretical optimal number of cluster heads based on the network model and communication energy consumption model;

[0007] B. Use non-cooperative game method to select candidate cluster heads from sensors;

[0008] C. Use particle swarm optimization algorithm to determine the final cluster head;

[0009] D. Sensors are clustered nearby, and the network enters a stable data transmission phase;

[0010] E. Derivation of a faulty sensor reuse strategy (RFN). When there are sensors with soft permanent failures, the faulty sensors are used as relay nodes to achieve low-energy, long-life, and highly robust wireless sensor network routing.

[0011] Further, step A specifically includes:

[0012] A1. Obtain relevant parameters of the monitoring network. Assume that the number of sensors in the monitoring network is N and the monitoring area is m × n.

[0013] A2. Calculate cluster member energy consumption E based on the communication energy consumption model CM , cluster head energy consumption E CH , cluster energy consumption E Cluster , network energy consumption E network , the calculation formula is:

[0014]

[0015] Where: l is the length of the data packet (number of bits); k is the number of cluster heads; d toCH is the distance from the cluster member node to the cluster head; d toSN is the distance from the cluster head to the base station; ε elec is the energy consumption per bit of data sent or received; fs is the energy consumption coefficient under the free space model; ε mp is the energy consumption coefficient under the multipath fading model; ε DA is the energy consumption coefficient of data fusion;

[0016] A3. With the goal of minimizing network energy consumption, the theoretical optimal number of cluster heads k is derived based on the above formula. The calculation formula is:

[0017]

[0018] A4. Substitute the relevant parameters of the monitoring network into formula (2) to solve the theoretical optimal number of cluster heads k.

[0019] Further, step B specifically includes:

[0020] B1. Model the candidate cluster head selection problem as a clustering game model CG = {N, S, U}, where N represents the set of sensor nodes participating in the game, S = {H, NH} represents the strategy space of the sensor nodes, strategy H indicates that the sensor becomes a cluster head, and strategy NH indicates that the sensor becomes a cluster member;

[0021] B2. When all sensors choose strategy NH, they will send data directly to the base station, resulting in excessive energy consumption, so the utility is 0; when sensor i chooses strategy H to become the cluster head, its utility is defined as The utility of other cluster member nodes is defined as

[0022]

[0023] B3. Define the probability of sensor i selecting strategy H as p, calculate the Nash equilibrium probability p of the hybrid strategy; combine the residual energy, node degree and node importance to give the probability p of sensor i becoming a candidate cluster head i :

[0024]

[0025] Where: is the number of neighbor nodes; is the remaining energy; I i is the importance of the node; d toSN is the distance from the cluster head to the base station; α, β and γ are control parameters whose sum is 1; ω is the amplification factor;

[0026] B4. Each sensor negotiates with its neighboring nodes and calculates the probability p of itself becoming a cluster head i , and generate a random number rand i ∈[0,1];

[0027] B5. If rand i ≤p i , sensor i becomes a candidate cluster head and joins the candidate cluster head set C, otherwise sensor i becomes a cluster member.

[0028] Further, step C specifically includes:

[0029] C1. According to coverage C f , residual energy E f and importance I f The objective function F of the particle swarm optimization algorithm is defined as:

[0030]

[0031] Where: η, λ and δ are control parameters whose sum is 1;

[0032] C2. Use the particle swarm optimization algorithm to solve the objective function F and select the final cluster head close to the theoretical optimal number of cluster heads k from the candidate cluster head set C.

[0033] Further, step D specifically includes:

[0034] D1. According to the cluster head election results, the sensors that have not become cluster heads join the closest cluster head and become members of the cluster;

[0035] D2. The cluster head collects and organizes the monitoring data collected by the cluster members and transmits it to the base station;

[0036] D3. After a period of time, when the current round ends, a new cluster head is elected in the next round.

[0037] Further, step E specifically includes:

[0038] E1. Calculate the communication energy consumption E from the cluster head to the relay node based on the relationship between the distance d and the threshold distance d0 CtoR , communication energy consumption E from relay node to base station RtoS , calculate the communication energy consumption E from cluster head to base station CtoS :

[0039]

[0040] Where: d CtoR is the distance from the cluster head to the relay node; d RtoS is the distance from the relay node to the base station; d CtoS is the distance from the cluster head to the base station;

[0041] E2.E CtoR +E RtoS -E CtoS Indicates the energy saved by the relay node when transmitting data. If the value is less than 0, it means that the relay node effectively reduces the network energy consumption. CtoR d RtoS and d CtoS There are eight working conditions:

[0042]

[0043] E3. When a sensor experiences a soft permanent fault such as offset, drift, gain, or stuck fault that does not affect the sensor's communication function, the faulty sensor that satisfies formula (7) can act as a relay node to collect the monitoring data sent by the cluster head and forward it to the base station.

[0044] A second aspect of the present invention relates to an apparatus for a robust routing protocol for wireless sensor networks based on game theory and a particle swarm optimization algorithm, comprising a memory and one or more processors, wherein the memory stores executable code, and when the one or more processors execute the executable code, the apparatus is used for the method of the present invention for a robust routing protocol for wireless sensor networks based on game theory and a particle swarm optimization algorithm.

[0045] A third embodiment of the present invention relates to a computer-readable storage medium having a program stored thereon. When the program is executed by a processor, the method of the present invention for a robust routing protocol for wireless sensor networks based on game theory and particle swarm optimization algorithm is implemented.

[0046] The advantages of the present invention are:

[0047] (1) Based on the network model and energy consumption model, this paper derives the calculation formula for the theoretical optimal number of cluster heads and proposes a candidate cluster head selection method based on game theory. This method combines the mixed strategy Nash equilibrium with factors such as node residual energy, node degree, and node importance to make the candidate cluster heads selected more reasonable.

[0048] (2) The present invention proposes an optimal cluster head selection method based on particle swarm optimization and constructs an objective function that considers indicators such as the coverage rate, residual energy rate, and importance of the cluster head set, so that the final cluster heads are evenly distributed and their number is close to the theoretical optimal, effectively balancing the network energy consumption;

[0049] (3) The present invention proposes a fault sensor reuse strategy, which uses qualified soft permanent fault sensors as relay nodes, effectively utilizing the remaining energy of the fault sensors, maintaining the stability of the network topology, and extending the network life; BRIEF DESCRIPTION OF THE DRAWINGS

[0050] Figure 1 is a flow chart of the present invention;

[0051] Figure 2 is a network structure diagram of an embodiment of the present invention during a normal data transmission phase;

[0052] Figure 3 This is a graph showing the number of cluster heads in the FTRP-GPSO protocol of the present invention;

[0053] Figure 4 It is a graph showing the number of surviving nodes of the FTRP-GPSO protocol of the present invention and the comparison protocol;

[0054] Figure 5 It is the residual energy variation diagram of the FTRP-GPSO protocol of the present invention and the comparative protocol;

[0055] Figure 6 This is a network structure diagram of an embodiment of the present invention when a soft permanent fault occurs in a sensor;

[0056] Figure 7 It is a graph showing the number of cluster heads changing according to the RFN strategy of the present invention;

[0057] Figure 8 This is a graph showing the number of surviving nodes for the RFN strategy of the present invention and the comparison strategy;

[0058] Figure 9 It is the residual energy variation diagram of the RFN strategy of the present invention and the comparative strategy;

[0059] Figure 10 It is a schematic diagram of the device of the present invention. DETAILED DESCRIPTION

[0060] The present invention will be further described in detail below with reference to the accompanying drawings and a simulation example.

[0061] Example 1

[0062] This embodiment discloses a method for a robust routing protocol for wireless sensor networks based on game theory and particle swarm optimization algorithm, such as: Figure 1 As shown, it specifically includes the following steps:

[0063] A. Derivation of the theoretical optimal number of cluster heads based on the network model and communication energy consumption model, specifically including:

[0064] A1. Set the WSN monitoring area to 400m × 150m, with 71 randomly distributed measurement points. The simulation parameters are shown in Table 1.

[0065] Table 1. Network simulation parameters

[0066]

[0067]

[0068] A2. Calculate cluster member energy consumption E based on the communication energy consumption model CM , cluster head energy consumption E CH , cluster energy consumption E Cluster , network energy consumption E network ;

[0069] A3. By bringing in the relevant parameters of the monitoring network and taking minimizing the network energy consumption as the goal, it is deduced that the theoretical optimal number of cluster heads k is 10.

[0070] B. Use non-cooperative game method to select candidate cluster heads from sensors, including:

[0071] B1. Model the candidate cluster head selection problem as a clustering game model CG = {N, S, U}, and bring the network parameters into the utility function U(s i );

[0072] B2. Define the probability of sensor i selecting strategy H as p, calculate the Nash equilibrium probability p of the hybrid strategy; combine the residual energy, node degree and node importance to give the probability p of sensor i becoming a candidate cluster head i :

[0073] B3. Each sensor negotiates with its neighboring nodes and calculates the probability p of itself becoming a cluster head i , and generate a random number rand i ∈[0,1];

[0074] B4. If rand i ≤p i , sensor i becomes a candidate cluster head and joins the candidate cluster head set C, otherwise sensor i becomes a cluster member.

[0075] C. Use particle swarm optimization algorithm to determine the final cluster head, including:

[0076] C1. Bring in network parameters, according to coverage C f , residual energy E f and importance I f Define the objective function F of the particle swarm optimization algorithm;

[0077] C2. Use the particle swarm optimization algorithm to solve the objective function F and select the final cluster head close to the theoretical optimal number of cluster heads k from the candidate cluster head set C.

[0078] D. Sensors cluster nearby, and the network enters the stable data transmission phase, specifically including:

[0079] D1. According to the cluster head election results, the sensors that have not become cluster heads join the closest cluster head and become members of the cluster;

[0080] D2. The cluster head collects and organizes the monitoring data collected by the cluster members and transmits it to the base station, such as Figure 2 As shown;

[0081] D3. After a period of time, the current round ends and a new cluster head is elected in the next round;

[0082] D4. In the case of no sensor failure, the number of cluster heads of the proposed FTRP-GPSO protocol changes as compared with other protocols. Figure 3 As shown, the number of surviving nodes changes as Figure 4 As shown, the remaining energy changes as Figure 5 shown.

[0083] E. Derive the RFN strategy. When there are sensors with soft permanent failures, the faulty sensors are used as relay nodes to achieve low-energy, long-life, and highly robust wireless sensor network routing. Specifically, the strategy includes:

[0084] E1. When there is a sensor with soft permanent fault, calculate d CtoR d RtoS and d CtoS , and determine whether the RFN strategy is met;

[0085] E2. The faulty node that meets the RFN strategy acts as a relay node, collects the monitoring data sent by the cluster head, and forwards it to the base station, such as Figure 6 As shown;

[0086] E3. After a period of time, when the current round ends, a new cluster head is elected in the next round.

[0087] E4. In the case of a soft permanent fault in the sensor, the number of cluster heads of the FTRP-GPSO protocol using the RFN strategy changes compared with other strategies. Figure 7 As shown, the number of surviving nodes changes as Figure 8 As shown, the remaining energy changes as Figure 9 shown.

[0088] Example 2

[0089] like Figure 10 This embodiment relates to a device for a robust routing protocol for wireless sensor networks based on game theory and particle swarm optimization algorithm, including a memory and one or more processors. The memory stores executable code. When the one or more processors execute the executable code, they are used to implement the method for a robust routing protocol for wireless sensor networks based on game theory and particle swarm optimization algorithm of Example 1.

[0090] Example 3

[0091] This embodiment relates to a computer-readable storage medium having a program stored thereon. When the program is executed by a processor, the method of the wireless sensor network robust routing protocol based on game theory and particle swarm optimization algorithm of embodiment 1 is implemented.

[0092] The contents described in the implementation cases of this specification are merely an enumeration of the implementation forms of the inventive concept. The scope of protection of the present invention should not be regarded as limited to the specific forms described in the implementation cases. The scope of protection of the present invention also extends to equivalent technical means that can be conceived by those skilled in the art based on the inventive concept.

Claims

1. A method for a robust routing protocol for wireless sensor networks based on game theory and particle swarm optimization algorithm, comprising the following steps: A. Derive the theoretical optimal number of cluster heads based on the network model and communication energy consumption model; B. Use non-cooperative game method to select candidate cluster heads from sensors; C. Use particle swarm optimization algorithm to determine the final cluster head; D. Sensors are clustered nearby, and the network enters a stable data transmission phase; E. Derive a faulty sensor reuse strategy. When there are sensors with soft permanent failures, the faulty sensors are used as relay nodes to achieve low-energy, long-life, and highly robust wireless sensor network routing. Step A specifically includes: A1. Obtain relevant parameters of the monitoring network. Assume that the number of sensors in the monitoring network is N and the monitoring area is m × n. A2. Calculate cluster member energy consumption E based on the communication energy consumption model CM , cluster head energy consumption E CH , cluster energy consumption E Cluster , network energy consumption E network , the calculation formula is: Where: l is the length of the data packet (number of bits); k is the number of cluster heads; d toCH is the distance from the cluster member node to the cluster head; d toSN is the distance from the cluster head to the base station; ε elec is the energy consumption per bit of data sent or received; fs is the energy consumption coefficient under the free space model; ε mp is the energy consumption coefficient under the multipath fading model; ε DA is the energy consumption coefficient of data fusion; A3. With the goal of minimizing network energy consumption, the theoretical optimal number of cluster heads k is derived based on the above formula. The calculation formula is: A4. Substitute the relevant parameters of the monitoring network into formula (2) to solve the theoretical optimal number of cluster heads k.

2. The method for a robust routing protocol for wireless sensor networks based on game theory and particle swarm optimization algorithm according to claim 1, wherein: Step B specifically includes: B1. Model the candidate cluster head selection problem as a clustering game model CG = {N, S, U}, where N represents the set of sensor nodes participating in the game, S = {H, NH} represents the strategy space of the sensor nodes, strategy H indicates that the sensor becomes a cluster head, and strategy NH indicates that the sensor becomes a cluster member; B2. When all sensors choose strategy NH, they will send data directly to the base station, resulting in excessive energy consumption, so the utility is 0; when sensor i chooses strategy H to become the cluster head, its utility is defined as The utility of other cluster member nodes is defined as B3. Define the probability of sensor i selecting strategy H as p, calculate the Nash equilibrium probability p of the hybrid strategy; combine the residual energy, node degree and node importance to give the probability p of sensor i becoming a candidate cluster head i : Where: is the number of neighbor nodes; is the remaining energy; I i is the importance of the node; d toSN is the distance from the cluster head to the base station; α, β and γ are control parameters whose sum is 1; ω is the amplification factor; B4. Each sensor negotiates with its neighboring nodes and calculates the probability p of itself becoming a cluster head i , and generate a random number rand i ∈[0,1]; B5. If rand i ≤p i , sensor i becomes a candidate cluster head and joins the candidate cluster head set C, otherwise sensor i becomes a cluster member.

3. The method for a robust routing protocol for wireless sensor networks based on game theory and particle swarm optimization algorithm according to claim 2, wherein: Step C specifically includes: C1. According to coverage C f , residual energy E f and importance I f The objective function F of the particle swarm optimization algorithm is defined as: Where: η, λ and δ are control parameters whose sum is 1; Γ i is the coverage value of sensor i; C2. Use the particle swarm optimization algorithm to solve the objective function F and select the final cluster head close to the theoretical optimal number of cluster heads k from the candidate cluster head set C.

4. The method for a robust routing protocol for wireless sensor networks based on game theory and particle swarm optimization algorithm according to claim 1, wherein: Step D specifically includes: D1. According to the cluster head election results, the sensors that have not become cluster heads join the closest cluster head and become members of the cluster; D2. The cluster head collects and organizes the monitoring data collected by the cluster members and transmits it to the base station; D3. After a period of time, when the current round ends, a new cluster head is elected in the next round.

5. The method for a robust routing protocol for wireless sensor networks based on game theory and particle swarm optimization algorithm according to claim 1, wherein: Step E specifically includes: E1. Calculate the communication energy consumption E from the cluster head to the relay node based on the relationship between the distance d and the threshold distance d0 CtoR , communication energy consumption E from relay node to base station RtoS , calculate the communication energy consumption E from cluster head to base station CtoS : Where: d CtoR is the distance from the cluster head to the relay node; d RtoS is the distance from the relay node to the base station; d CtoS is the distance from the cluster head to the base station; E2.E CtoR +E RtoS -E CtoS Indicates the energy saved by the relay node when transmitting data. If the value is less than 0, it means that the relay node effectively reduces the network energy consumption. CtoR d RtoS and d CtoS There are eight working conditions: E3. When a sensor experiences a soft permanent fault such as offset, drift, gain, or stuck fault that does not affect the sensor's communication function, the faulty sensor that satisfies formula (7) can act as a relay node to collect the monitoring data sent by the cluster head and forward it to the base station.

6. A device for a robust routing protocol for wireless sensor networks based on game theory and particle swarm optimization algorithm, characterized in that: The method comprises a memory and one or more processors, wherein the memory stores executable code, and when the one or more processors execute the executable code, they are used to implement the method of a wireless sensor network robust routing protocol based on game theory and particle swarm optimization algorithm according to any one of claims 1 to 5.

7. A computer-readable storage medium, characterized in that A program is stored thereon, and when the program is executed by a processor, the method of a wireless sensor network robust routing protocol based on game theory and particle swarm optimization algorithm according to any one of claims 1 to 5 is implemented.

Images