Soil humidity monitoring wireless sensor network coverage optimization method based on chaos adaptive dung beetle optimization algorithm

Abstract

The invention relates to the technical field of wireless sensor networks, is applied to a soil humidity monitoring scene in agriculture, and particularly relates to a soil humidity monitoring wireless sensor network coverage optimization method based on a chaos adaptive dung beetle optimization algorithm, which comprises the following steps: constructing a soil humidity monitoring wireless sensor network coverage model of a monitoring area; an improved dung beetle optimization algorithm is adopted to carry out coverage optimization on the soil humidity monitoring wireless sensor network coverage model, an optimized coverage result is obtained, and the improved dung beetle optimization algorithm is obtained through improvement of a Logistic chaos operator and an adaptive quasi-reverse learning strategy. According to the method, the coverage rate and robustness of the agricultural soil humidity monitoring wireless sensor network can be improved, local optimum can be avoided while the convergence speed is guaranteed and global search is improved, and the optimal coverage effect is achieved.

Description

Technical Field

[0001] The invention relates to the technical field of wireless sensor networks, and is applied to soil moisture monitoring scenarios in agriculture, in particular to a soil moisture monitoring wireless sensor network coverage optimization method based on a chaotic adaptive dung beetle optimization algorithm. Background Art

[0002] As agriculture becomes increasingly dependent on advanced technologies to improve productivity and sustainability, soil moisture monitoring wireless sensor networks (SMMWSNs) are widely used in agricultural production as one of the key technologies to provide farmers with real-time soil moisture data to support precision irrigation and soil health monitoring. Since traditional soil moisture monitoring methods often face problems such as single-point sampling, time delay, and data accuracy, it is difficult to meet the needs of timely monitoring. Therefore, the development of SMMWSNs technology has gradually become an effective means to solve these problems. SMMWSNs are composed of multiple sensor nodes with communication and computing capabilities. These nodes are flexible, reliable, and low-cost. Network node coverage technology is a key factor to ensure network stability and normal operation, which directly affects network quality and overall performance. Therefore, it is of great significance to study the coverage optimization problem of SMMWSNs.

[0003] In order to deal with the problem of network coverage optimization, various swarm intelligence optimization algorithms have been proposed in recent years to replace traditional optimization algorithms. These algorithms include the snake optimization algorithm (SO), the manta ray foraging optimization algorithm (MRFO), the whale optimization algorithm (WOA) and the sparrow search algorithm (SSA). Among them, SO can effectively avoid premature convergence to the local optimum, but its global search ability is limited. MRFO shows strong global search ability, but often has the problem of slow convergence. WOA can achieve fast convergence in the initial stage, but may face the dilemma of premature convergence. SSA can maintain diversity well during the search process, which is crucial to avoid early convergence and ensure more comprehensive network coverage. However, its global search efficiency may be limited in some cases, affecting the reliability and stability of search results.

[0004] In the problem of optimizing the coverage of soil moisture monitoring wireless sensor networks (SMMWSNs), many studies have made important contributions to the field and explored various methods, covering both non-heuristic and heuristic methods. Tossa F. et al. proposed a coverage maximization method based on connectivity constraints in sensor networks. The method effectively balances coverage and connectivity, ensuring that sensor nodes maintain reliable communication links while maximizing the monitoring area. However, the algorithm faces the challenge of increased computational complexity in large-scale networks, which may lead to inefficiencies in execution time and resource allocation. Yang J. et al. proposed an improved Cuckoo algorithm to optimize the coverage and routing of wireless sensor networks. The algorithm benefits from enhanced global search capabilities and efficient exploration of the solution space, thereby improving coverage and routing performance. Despite these advantages, the algorithm may exhibit slow convergence when dealing with highly complex or large networks, which affects its real-time applicability.

[0005] However, most coverage algorithms have low coverage, which is usually caused by the limitations of sensor deployment strategies. Reduced coverage can lead to errors in data collection, which affects the accuracy and completeness of SMMWSNs monitoring data; in addition, most heuristic coverage algorithms have high computational complexity, resulting in excessive energy consumption of SMMWSNs and shortened network life, which greatly limits their applicability for long-term deployment in agricultural environments; and most heuristic coverage algorithms have problems such as slow convergence and easy to fall into local optimality, which greatly reduces the effectiveness of the algorithm. These limitations lead to poor network performance and decision delays, affecting the overall efficiency of SMMWSNs.

[0006] Therefore, optimizing network node coverage is of great significance to improving the effect of SMMWSNs. Seeking better coverage while reducing computational complexity, ensuring faster convergence speed, and enhancing the ability to escape local optimality are problems that the current algorithm needs to solve. Summary of the invention

[0007] The purpose of the present invention is to provide a soil moisture monitoring wireless sensor network coverage optimization method based on a chaotic adaptive dung beetle optimization algorithm, which solves the limitations of the prior art by providing higher coverage, reducing computational complexity, ensuring faster convergence speed, and enhancing the ability to jump out of local optimality.

[0008] To achieve the above object, the present invention provides the following solutions:

[0009] A soil moisture monitoring wireless sensor network coverage optimization method based on chaotic adaptive dung beetle optimization algorithm, comprising:

[0010] Construct a wireless sensor network coverage model for soil moisture monitoring in the monitoring area;

[0011] The improved dung beetle optimization algorithm is used to optimize the coverage of the soil moisture monitoring wireless sensor network coverage model to obtain the optimized coverage result, wherein the improved dung beetle optimization algorithm is obtained by improving the Logistic chaos operator and the adaptive quasi-reverse learning strategy.

[0012] Optionally, constructing a soil moisture monitoring wireless sensor network coverage model for the monitoring area includes:

[0013] Establishing a two-dimensional plane model of the monitoring area;

[0014] A number of monitoring target points and sensor nodes are randomly set in the two-dimensional plane model to construct a soil moisture monitoring wireless sensor network coverage model with full coverage connection.

[0015] Optionally, the full coverage connection of the soil moisture monitoring wireless sensor network coverage model is:

[0016]

[0017] Among them, f i represents the fitness function for optimizing the coverage of wireless sensor networks for soil moisture monitoring, S i,j represents the sensor node p i For monitoring target point q j The coverage of (x i ,y i ) represents the sensor node p i The position of (x j ,y j ) represents the monitoring target point q j The position of s represents the sensor node p i Perception radius.

[0018] Optionally, using the improved dung beetle optimization algorithm to optimize the coverage of the soil moisture monitoring wireless sensor network coverage model includes:

[0019] Step 1: Set initialization parameters to initialize the sensor node positions in the soil moisture monitoring wireless sensor network coverage model;

[0020] Step 2: Initialize the population position through the Logistic chaos operator and calculate the individual fitness value of the population;

[0021] Step 3: Use the group linear decrease method to update the population size, and update the position through the behaviors of rolling dung beetles, breeding dung beetles, foraging dung beetles and stealing dung beetles in the dung beetle optimization algorithm to obtain the optimal individual and the best position;

[0022] Step 4: Use an adaptive quasi-reverse learning strategy to replace individuals, update the global optimal fitness and the global optimal position, until the maximum number of iterations is reached, and output the global optimal fitness. Otherwise, return to step 1.

[0023] Optionally, the initialization parameters in step 1 include: population size, individual numbers of rolling dung beetles, breeding dung beetles, foraging dung beetles and stealing dung beetles, maximum number of iterations, dimension, and upper and lower limits.

[0024] Optionally, the method of initializing the population position by using the Logistic chaos operator in step 2 is:

[0025]

[0026] Among them, w avg is the average fitness of the population, wb is the fitness of individual b, and w is used to limit σ 2 The size of σ 2 An indicator for evaluating whether a dung beetle population converges to a local optimum prematurely.

[0027] Optionally, the method of updating the population size using the population linear decrease method in step 3 is:

[0028]

[0029] Among them, pop' represents the updated population size, pop max and pop min They represent the maximum and minimum population sizes respectively, N is the total number of iterations, and P represents the total number of dung beetle populations, that is, the number of sensor nodes.

[0030] Optionally, the method for updating the position in step 3 through the behaviors of the rolling dung beetle, the breeding dung beetle, the foraging dung beetle and the stealing dung beetle in the dung beetle optimization algorithm is:

[0031] Update the position of the rolling dung beetle behavior:

[0032]

[0033] Among them, n represents the current iteration number, x j n represents the position of the jth dung beetle at the nth iteration; k is the deflection coefficient; B represents a fixed value in the range of (0,1); C is a natural coefficient, X mrepresents the global worst position, and Δx is used to simulate the intensity change of light;

[0034] Update the position of the behavior of breeding dung beetles:

[0035]

[0036] Update the location of the foraging dung beetle behavior:

[0037]

[0038] Update the location of the stealing dung beetle:

[0039]

[0040] Where β and h are the parameters of the improved variable spiral search strategy, β = e qn ×cos(2πq), q is a random number uniformly distributed in the range (0,1), h is affected by the cosine function and changes with the number of iterations; n max is the maximum number of iterations, x* is the current local optimal value under the dung beetle breeding state, and x n is the current local optimal value of the dung beetle in the foraging state, x q is the current local optimal value of the dung beetle in the stealing state, hq* and uq* represent the upper and lower limits of the breeding area respectively, hq q and uq q They represent the upper and lower limits of the foraging area respectively, g is a random vector of size 1×D that conforms to the normal distribution, and D represents the dimension of the optimization challenge.

[0041] Optionally, in step 4, an adaptive quasi-reverse learning strategy is used to perform individual replacement, and a method for updating the global optimal fitness is:

[0042]

[0043] Among them, x j Q represents the position of the jth dung beetle in the Qth dimension, hq is used to guide the movement direction of the dung beetle in the qth dimension to help the individual move towards a better solution, uq is used to control the displacement amplitude of the dung beetle in the qth dimension to adapt to the search needs at different stages, and x j represents the position of the jth dung beetle, rand represents a random number used to introduce randomness, and its value range is [0,1].

[0044] Optionally, in step 4, the global optimal position is updated by using a dimension-by-dimension mutation strategy, which is:

[0045]

[0046] Among them, x q new (i) represents the updated value of the global best position in the i-th dimension, x q (i) represents the i-th dimension of the optimal individual, n max is the maximum number of iterations, n is the current number of iterations, x random (i) represents a random position in the i-th dimension, and cauchy represents the Cauchy distribution, which is used to generate a random variable that obeys the Cauchy distribution.

[0047] The beneficial effects of the present invention are:

[0048] The present invention combines Logistic chaotic mapping, adaptive operator, variable spiral search strategy and dimension-by-dimensional mutation strategy to improve the dung beetle algorithm, and proposes a soil moisture monitoring wireless sensor network coverage optimization method based on the chaotic adaptive dung beetle optimization algorithm to optimize the coverage of SMMWSNs. While ensuring the optimal convergence speed and reducing the risk of falling into local optimality, the computational complexity is kept low, thereby effectively improving the coverage efficiency and robustness of SMMWSNs. BRIEF DESCRIPTION OF THE DRAWINGS

[0049] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0050] Figure 1 A soil moisture monitoring wireless sensor network coverage optimization method based on a chaotic adaptive dung beetle optimization algorithm according to an embodiment of the present invention;

[0051] Figure 2 1 is a comparison diagram of optimized node distribution and average coverage optimization results of five algorithms in an 80m×80m area of ​​an embodiment of the present invention, wherein (a) is a CADBO optimized node distribution diagram, (b) is a SO optimized node distribution diagram, (c) is a SSA optimized node distribution diagram, (d) is a CWOA optimized node distribution diagram, (e) is a MRFO optimized node distribution diagram, and (f) is a coverage comparison diagram of the five algorithms;

[0052] Figure 31 is a comparison diagram of optimized node distribution and average coverage optimization results of five algorithms in a 160m×160m area of ​​an embodiment of the present invention, wherein (a) is a CADBO optimized node distribution diagram, (b) is a SO optimized node distribution diagram, (c) is a SSA optimized node distribution diagram, (d) is a CWOA optimized node distribution diagram, (e) is a MRFO optimized node distribution diagram, and (f) is a coverage comparison diagram of the five algorithms;

[0053] Figure 4 1 is a comparison diagram of optimized node distribution and average coverage optimization results of five algorithms in a 240m×240m area of ​​an embodiment of the present invention, wherein (a) is a CADBO optimized node distribution diagram, (b) is a SO optimized node distribution diagram, (c) is a SSA optimized node distribution diagram, (d) is a CWOA optimized node distribution diagram, (e) is a MRFO optimized node distribution diagram, and (f) is a coverage comparison diagram of the five algorithms;

[0054] Figure 5 4 is a comparison chart of the convergence speeds of the five algorithms according to the embodiments of the present invention. DETAILED DESCRIPTION

[0055] 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.

[0056] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the present invention is further described in detail below with reference to the accompanying drawings and specific embodiments.

[0057] This embodiment provides a soil moisture monitoring wireless sensor network coverage optimization method based on a chaotic adaptive dung beetle optimization algorithm, including:

[0058] Construct a wireless sensor network coverage model for soil moisture monitoring in the monitoring area;

[0059] The improved dung beetle optimization algorithm is used to optimize the coverage model of soil moisture monitoring wireless sensor network to obtain the optimized coverage result. The improved dung beetle optimization algorithm is obtained by improving the Logistic chaos operator and the adaptive quasi-reverse learning strategy.

[0060] Specifically, this embodiment combines Logistic chaotic mapping, adaptive operators, variable spiral search strategy, and dimension-by-dimensional mutation strategy to improve the dung beetle algorithm, and proposes a soil moisture monitoring wireless sensor network coverage optimization method based on the chaotic adaptive dung beetle optimization algorithm to optimize the coverage of the soil moisture monitoring wireless sensor network coverage model SMMWSNs. While ensuring the optimal convergence speed and reducing the risk of falling into local optimality, it maintains a low computational complexity, thereby effectively improving the coverage efficiency and robustness of SMMWSNs.

[0061] Furthermore, the construction of the soil moisture monitoring wireless sensor network coverage model in the monitoring area includes:

[0062] Establish a two-dimensional plane model of the monitoring area;

[0063] Several monitoring target points and sensor nodes are randomly set in the two-dimensional plane model to construct a soil moisture monitoring wireless sensor network coverage model with full coverage connection.

[0064] Specifically, the SMMWSNs coverage model is constructed as follows: First, a two-dimensional plane monitoring area A is established, and n sensor nodes are randomly placed in the area. The sensing radius of the sensor node is r s , the communication radius is r c The position p of each sensor node i Use (x i ,y i ), where 1≤i≤n, and the acquisition points of the monitoring target point are P={p1,p2,...,p n}. Each point Q has coordinates x′ j ,y′ j Logo.

[0065] This embodiment assumes an ideal wireless channel, in which the communication between the sensor node and the sink node is reliable, without transmission errors and without the need for retransmission.

[0066] The full coverage connections of SMMWSNs are:

[0067]

[0068] in, represents the sensor node p i With the monitoring target point q j If the distance is less than r s , S i,j =1 indicates sensor p i Successfully covered the target point q j .

[0069] The fitness function for soil moisture monitoring wireless sensor network coverage optimization is:

[0070]

[0071] Furthermore, the improved dung beetle optimization algorithm is used to optimize the coverage model of the soil moisture monitoring wireless sensor network, including:

[0072] Step 1: Set initialization parameters to initialize the sensor node positions in the soil moisture monitoring wireless sensor network coverage model;

[0073] Step 2: Initialize the population position through the Logistic chaos operator and calculate the individual fitness value of the population;

[0074] Step 3: Use the group linear decrease method to update the population size, and update the position through the behaviors of rolling dung beetles, breeding dung beetles, foraging dung beetles and stealing dung beetles in the dung beetle optimization algorithm to obtain the optimal individual and the best position;

[0075] Step 4: Use an adaptive quasi-reverse learning strategy to replace individuals, update the global optimal fitness and the global optimal position, until the maximum number of iterations is reached, and output the global optimal fitness. Otherwise, return to step 1.

[0076] Specifically, the initialization parameters in step 1 include: population size (pop), the number of individuals of rolling dung beetles, breeding dung beetles, foraging dung beetles and stealing dung beetles, the maximum number of iterations (n max ), dimension (D), and upper and lower limits (hq and uq);

[0077] In step 2, the Logistic chaos operator is used to initialize the population position, which can generate a random sequence and use it to adjust the value of the parameter w. The formula is as follows:

[0078]

[0079] Where n represents the total number of dung beetle populations, w avg is the average fitness of the population, w b is the fitness of individual b, and w is used to limit σ 2 The size of σ 2 An indicator for evaluating whether a dung beetle population converges to a local optimum prematurely.

[0080] In order to prevent the algorithm from falling into the local optimum during the iteration process and make the initial node distribution as uniform as possible, the Logistic chaotic mapping strategy is added as follows:

[0081] x n+1 = r·x n (1-x n)(4)

[0082] For SMMWSNs, x n ∈[0,1], and r represents a positive integer related to the reproduction rate and mortality rate. The chaotic sequence is mapped to the value space of the optimization variable, thereby utilizing the chaotic characteristics to determine the initial node position.

[0083] This embodiment introduces the Logistic chaos operator, which not only optimizes the positioning of sensor nodes in the soil moisture monitoring network, but also improves the convergence speed and global search capability of the algorithm. This improvement is particularly critical in agricultural applications, because the precise positioning of sensor nodes plays a key role in the quality and reliability of real-time soil moisture data. By preventing the algorithm from falling into local optimality, the Logistic chaos operator can ensure more comprehensive coverage of agricultural fields, thereby supporting more accurate irrigation and soil management strategies, which is crucial for optimizing water resource utilization and increasing crop yields.

[0084] In step 3, the population linear decrease method is used to update the population pop:

[0085]

[0086] Where pop' represents the updated population size, pop max and pop min represent the maximum and minimum numbers respectively, N is the total number of iterations, and P represents the total number of dung beetle populations, that is, the number of sensor nodes.

[0087] In step 3, the position is updated through the behaviors of the rolling dung beetle, breeding dung beetle, foraging dung beetle and stealing dung beetle in the dung beetle optimization algorithm, including:

[0088] When the dung beetle performs the rolling behavior, the position update when there is no obstacle in the forward direction is as follows:

[0089]

[0090] Where n is the current iteration number, indicating the position of the jth dung beetle at the nth iteration; k is the deflection coefficient, which is in the range of (0,0.2]; B is a fixed value in the range of (0,1); C is a natural coefficient, whose value may be -1 or 1, where 1 indicates no deviation and -1 indicates deviation from the initial path. m represents the global worst position, and Δx is used to simulate the intensity variation of light.

[0091] The position updates of dung beetles as they forage, breed, and steal are as follows:

[0092]

[0093] Among them, the improved variable spiral search strategy is: β = e qn ×cos(2πq), q is defined as a random number uniformly distributed in the range of 0 to 1; at the same time, h is affected by the cosine function and changes with the number of iterations; x* is the current local optimal value under the breeding state of the dung beetle, x n is the current local optimal value of the dung beetle in the foraging state, x q is the current local optimal value of the dung beetle in the stealing state, while hq* and uq* represent the upper and lower limits of the breeding area respectively. q and uq q represent the upper and lower limits of the foraging area respectively; at the same time, g is a random vector of size 1×D that conforms to the normal distribution, and D represents the dimension of the optimization challenge.

[0094] In step 4, an adaptive quasi-reverse learning strategy is used to replace the worst individual and update the global optimal fitness to:

[0095]

[0096] Among them, x j Q represents the position of the jth dung beetle in the Qth dimension, hq is used to guide the movement direction of the dung beetle in the qth dimension to help the individual move towards a better solution, uq is used to control the displacement amplitude of the dung beetle in the qth dimension to adapt to the search needs at different stages, and x j represents the position of the jth dung beetle, rand represents a random number used to introduce randomness, and its value range is [0,1].

[0097] Evaluate the adaptability. If the new position is better than the original position, replace the new position. At this time, apply the adaptive quasi-reverse learning strategy according to the above formula, generate a reverse individual in the population, and calculate the fitness value of the reverse individual. If the fitness of the reverse individual exceeds that of the original individual, replace the worst individual and update the global optimal fitness. Therefore, the dimension-by-dimensional mutation strategy is used to update the global optimal position in step 4, specifically:

[0098]

[0099] Among them, x q (i) represents the i-th dimension of the optimal individual, and the cauchy theorem is used to calculate using random allocation. q new (i) represents the updated value of the global best position in the i-th dimension, x random (i) represents a random position in the i-th dimension.

[0100] Determine whether the maximum number of iterations n has been reachedmax , if the maximum number of iterations is reached, the best fit value and its corresponding sensor distribution are output; if the maximum number of iterations is still not reached, return to initialization step 1.

[0101] This embodiment combines the variable spiral search strategy with the dimension-by-dimension mutation strategy and proposes a new multi-strategy method, thereby improving the network coverage efficiency and robustness and ultimately achieving maximum resource utilization.

[0102] Experimental analysis:

[0103] Through a large number of simulation experiments, the coverage optimization effects of five algorithms, CADBO, SO, SSA, CWOA and MRFO, on SMMWSNs under different conditions such as network coverage radius limitation, number of sensor nodes and spacing are systematically compared:

[0104] The experiments were conducted in a controlled computing environment using an Intel(R) Core(TM) i7-8565U CPU with a base frequency of 1.80 GHz and a processing speed of 1.99 GHz, running on a Windows 10 operating system with a 64-bit architecture.

[0105] This setup was chosen to reflect the balance between computing power and actual resource availability, simulating the actual conditions of SMMWSNs deployment. The experiments were implemented using Matlab R2023b software, which provides a powerful platform for modeling, simulation, and analysis of complex algorithms such as CADBO.

[0106] It is assumed that SMMWSNs nodes are deployed in 80m×80m farmland monitoring areas, 160m×160m farmland monitoring areas and 240m×240m farmland monitoring areas respectively. 30 to 40 sensors are randomly deployed and the node sensing radius is 10 to 20m. The experiment uses five deployment methods, including CADBO, SO, SSA, CWOA and MRFO, for comparative experiments.

[0107] Specifically, refer to Figure 2 ,The SMMWSNs nodes are deployed in a 80m×80m farmland monitoring area, 30 sensors are randomly deployed, the node sensing radius is 10m, and the communication radius is 20m. ,The experiment uses five deployment methods, CADBO, SO, SSA, CWOA and MRFO, for comparison.

[0108] from Figure 2It can be clearly seen from (a), (b), (c), (d), (e), and (f) that the other four improved comparison algorithms have poor optimization effects and are prone to coverage holes. In contrast, the CADBO algorithm has a more uniform node distribution and less node redundancy after optimization, which makes the wireless network coverage wider. This shows that CADBO performs better in terms of node coverage of SMMWSNs and has advantages in balance and coverage performance.

[0109] Reference Figure 3 ,The SMMWSNs nodes are deployed in a 160m×160m farmland monitoring area, 40 sensors are randomly deployed, the node sensing radius is 15m, and the communication radius is 25m. ,The experiment uses five deployment methods, CADBO, SO, SSA, CWOA and MRFO, for comparison.

[0110] Compared with the four algorithms SO, SSA, CWOA and MRFO, CADBO improves the average coverage by 10.56%, 7.38%, 5.30% and 12.44% in 50 coverage optimizations respectively. Figure 3 From the node distribution shown in (a), (b), (c), (d), (e), and (f), it can be seen that the regional node coverage blind area after SO, SSA, CWOA, and MRFO optimization is large, while the SMMWSNs node distribution after CADBO optimization is more uniform, which significantly improves the distribution of other nodes in a large area and effectively reduces the optimization cost of the wireless sensor network.

[0111] Reference Figure 4 ,The SMMWSNs nodes are deployed in a 240m×240m farmland monitoring area, 45 sensors are randomly deployed, the node sensing radius is 20m, and the communication radius is 30m. ,The experiment uses five deployment methods, CADBO, SO, SSA, CWOA and MRFO, for comparison.

[0112] from Figure 4 From the node distributions in (a), (b), (c), (d), (e), and (f), it can be seen that the CADBO algorithm is significantly better than other algorithms in terms of node coverage. The SO and SSA algorithms show significant overlap and redundancy in their optimized node distributions, indicating excessive duplication between nodes. Meanwhile, the CWOA and MRFO algorithms show progressive coverage blind spots in their optimized distributions, indicating insufficient node coverage in some areas. In contrast, the CADBO algorithm shows greater uniformity in its optimized node distribution, effectively reducing the overlap between nodes.

[0113] In the simulation experiment, the present invention maintains the consistency of the parameters of several comparison algorithms. In order to improve the credibility of the results, each comparison algorithm is independently simulated 200 times, and then the average coverage rate is used for comparative analysis. This method more accurately evaluates the performance of various algorithms in SMMWSNs node coverage, ensuring the reliability and persuasiveness of the research results. Table 1 shows the average coverage rates achieved by different algorithms in three farmland monitoring areas of different sizes: 80m×80m, 160m×160m, and 240m×240m.

[0114] Table 1

[0115]

[0116] Simulation experimental results show that CADBO outperforms the other four algorithms and its coverage is improved by 8.44%, 14.17%, 3.95% and 9.17% respectively, demonstrating its excellent effectiveness in improving the coverage of SMMWSNs.

[0117] refer to Figure 5 , the performance of CADBO, SO, SSA, CWOA and MRFO in terms of convergence speed, sensitivity to local optimum and computational complexity in optimizing the network coverage of SMMWSNs are compared. Specifically, although CWOA and SSA show fast convergence in the early stage, both algorithms tend to fall into the local optimum prematurely at about the 300th generation, which hinders their ability to obtain the optimal solution. On the other hand, SO converges much slower, requiring about 600 generations to approach convergence, and minimal improvements are observed after the 400th generation. This slow convergence highlights its inefficiency in effectively exploring the search space. Although MRFO converges faster, it also has the problem of falling into the local optimum, which limits its overall effectiveness. The running time of the five algorithms in different regions is shown in Table 2, and the time complexity calculation of the five algorithms in different regions is shown in Table 3.

[0118] Table 2

[0119]

[0120] Table 3

[0121]

[0122] At the same time, in the SMMWSNs (wireless soil moisture monitoring network) environment, the convergence speed and time complexity of the five algorithms CADBO, SO, SSA, CWOA and MRFO were compared and analyzed. CADBO always showed the best convergence speed and the lowest time complexity in all three farmland monitoring areas, showing its ability to jump out of local optimality, enhance computational efficiency and reduce resource consumption in large-scale and complex SMMWSNs scenarios. At the same time, it shows that CADBO has better scalability and faster response time, making it highly applicable in various application scenarios.

[0123] The embodiments described above are only descriptions of the preferred embodiments of the present invention and are not intended to limit the scope of the present invention. Without departing from the design spirit of the present invention, various modifications and improvements made to the technical solutions of the present invention by ordinary technicians in this field should all fall within the protection scope determined by the claims of the present invention.

Claims

1. A soil moisture monitoring wireless sensor network coverage optimization method based on chaotic adaptive dung beetle optimization algorithm, characterized in that: include: Construct a wireless sensor network coverage model for soil moisture monitoring in the monitoring area; The improved dung beetle optimization algorithm is used to optimize the coverage of the soil moisture monitoring wireless sensor network coverage model to obtain the optimized coverage result, wherein the improved dung beetle optimization algorithm is obtained by improving the Logistic chaos operator and the adaptive quasi-reverse learning strategy.

2. The soil moisture monitoring wireless sensor network coverage optimization method based on chaotic adaptive dung beetle optimization algorithm according to claim 1 is characterized in that: Constructing the soil moisture monitoring wireless sensor network coverage model of the monitoring area includes: Establishing a two-dimensional plane model of the monitoring area; A number of monitoring target points and sensor nodes are randomly set in the two-dimensional plane model to construct a soil moisture monitoring wireless sensor network coverage model with full coverage connection.

3. The soil moisture monitoring wireless sensor network coverage optimization method based on chaotic adaptive dung beetle optimization algorithm according to claim 2 is characterized in that: The full coverage connection of the soil moisture monitoring wireless sensor network coverage model is: Among them, f i represents the fitness function for optimizing the coverage of wireless sensor networks for soil moisture monitoring, S i,j represents the sensor node p i For monitoring target point q j The coverage of (x i ,y i ) represents the sensor node p i The position of (x j ,y j ) represents the monitoring target point q j The position of s represents the sensor node p i Perception radius.

4. The soil moisture monitoring wireless sensor network coverage optimization method based on chaotic adaptive dung beetle optimization algorithm according to claim 1 is characterized in that: The improved dung beetle optimization algorithm is used to optimize the coverage of the soil moisture monitoring wireless sensor network coverage model, including: Step 1: Set initialization parameters to initialize the sensor node positions in the soil moisture monitoring wireless sensor network coverage model; Step 2: Initialize the population position through the Logistic chaos operator and calculate the individual fitness value of the population; Step 3: Use the group linear decrease method to update the population size, and update the position through the behaviors of rolling dung beetles, breeding dung beetles, foraging dung beetles and stealing dung beetles in the dung beetle optimization algorithm to obtain the optimal individual and the best position; Step 4: Use an adaptive quasi-reverse learning strategy to replace individuals, update the global optimal fitness and the global optimal position, until the maximum number of iterations is reached, and output the global optimal fitness. Otherwise, return to step 1.

5. The soil moisture monitoring wireless sensor network coverage optimization method based on chaotic adaptive dung beetle optimization algorithm according to claim 4 is characterized in that: The initialization parameters in step 1 include: population size, individual numbers of rolling dung beetles, breeding dung beetles, foraging dung beetles and stealing dung beetles, maximum number of iterations, dimension, and upper and lower limits.

6. The soil moisture monitoring wireless sensor network coverage optimization method based on chaotic adaptive dung beetle optimization algorithm according to claim 4 is characterized in that: The method of initializing the population position by the Logistic chaos operator in step 2 is: Among them, w avg is the average fitness of the population, w b is the fitness of individual b, and w is used to limit σ 2 The size of σ 2 An indicator for evaluating whether a dung beetle population converges to a local optimum prematurely.

7. The soil moisture monitoring wireless sensor network coverage optimization method based on chaotic adaptive dung beetle optimization algorithm according to claim 4 is characterized in that: The method of updating the population size using the population linear decrease method in step 3 is: Among them, pop' represents the updated population size, pop max and pop min They represent the maximum and minimum population sizes respectively, N is the total number of iterations, and P represents the total number of dung beetle populations, that is, the number of sensor nodes.

8. The soil moisture monitoring wireless sensor network coverage optimization method based on chaotic adaptive dung beetle optimization algorithm according to claim 4 is characterized in that: The method for updating the position in step 3 through the behaviors of the rolling dung beetle, the breeding dung beetle, the foraging dung beetle and the stealing dung beetle in the dung beetle optimization algorithm is: Update the position of the rolling dung beetle behavior: Among them, n represents the current iteration number, x j n represents the position of the jth dung beetle at the nth iteration; k is the deflection coefficient; B represents a fixed value in the range of (0,1); C is a natural coefficient, X m represents the global worst position, and Δx is used to simulate the intensity change of light; Update the position of the behavior of breeding dung beetles: Update the location of the foraging dung beetle behavior: Update the location of the stealing dung beetle: Where β and h are the parameters of the improved variable spiral search strategy, β = e qn ×cos(2πq), q is a random number uniformly distributed in the range (0,1), h is affected by the cosine function and changes with the number of iterations; n max is the maximum number of iterations, x* is the current local optimal value under the dung beetle breeding state, and x n is the current local optimal value of the dung beetle in the foraging state, x q is the current local optimal value of the dung beetle in the stealing state, hq* and uq* represent the upper and lower limits of the breeding area respectively, hq q and uq q They represent the upper and lower limits of the foraging area respectively, g is a random vector of size 1×D that conforms to the normal distribution, and D represents the dimension of the optimization challenge.

9. The soil moisture monitoring wireless sensor network coverage optimization method based on chaotic adaptive dung beetle optimization algorithm according to claim 4 is characterized in that: In step 4, an adaptive quasi-reverse learning strategy is used to replace individuals, and the method for updating the global optimal fitness is: Among them, x j Q represents the position of the jth dung beetle in the Qth dimension, hq is used to guide the movement direction of the dung beetle in the qth dimension to help the individual move towards a better solution, uq is used to control the displacement amplitude of the dung beetle in the qth dimension to adapt to the search needs at different stages, and x j represents the position of the jth dung beetle, rand represents a random number used to introduce randomness, and its value range is [0,1].

10. The soil moisture monitoring wireless sensor network coverage optimization method based on chaotic adaptive dung beetle optimization algorithm according to claim 4 is characterized in that: In step 4, the global optimal position is updated by using a dimension-by-dimension mutation strategy, which is: Among them, x q new (i) represents the updated value of the global best position in the i-th dimension, x q (i) represents the i-th dimension of the optimal individual, n max is the maximum number of iterations, n is the current number of iterations, x random (i) represents a random position in the i-th dimension, and cauchy represents the Cauchy distribution, which is used to generate a random variable that obeys the Cauchy distribution.

Images