WSN (Wireless Sensor Network) energy efficient coverage method and device for improving bacterial foraging algorithm

Abstract

Aiming at the problems of unbalanced energy and low coverage efficiency of the existing WSN, the invention provides a WSN energy efficient coverage method and device for improving a bacterial foraging algorithm (EABFO). According to the method, an adaptive variation method is adopted during position updating, before a predicted peak value is reached, the variation probability is increased along with the increase of the number of iterations, the randomness of bacteria is increased, global large-range search is realized, omission of an optimal solution is avoided, local optimum is tended, after the predicted peak value is reached, the variation probability is reduced, and local detail search is realized; time and space waste is reduced, and algorithm efficiency is improved. In addition, an elite breeding strategy is innovatively provided, the optimal fitness value is reserved, reverse growth of the fitness value is avoided, and the whole flora is evolved in a better direction. Compared with a genetic algorithm, the method has the advantages that the service life of the sensor matrix is prolonged by 200%, the method is suitable for energy-limited scenes such as large-scale static monitoring and dynamic target tracking, and the life cycle of the WSN is effectively prolonged.

Description

Technical Field

[0001] The present invention relates to an improved Elite Adaptive Bacterial Foraging Algorithm (EABFO) for efficient coverage of wireless sensor networks (WSNs) and an apparatus thereof. This invention pertains to the field of wireless sensor network (WSN) technology, and more particularly to an improved Elite Adaptive Bacterial Foraging Algorithm (EABFO) for efficient coverage of WSNs, belonging to the field of wireless sensor network energy management technology. Background Technology

[0002] Wireless sensor networks (WSNs) consist of a large number of sensor nodes deployed within a monitored area. These nodes form a self-organizing network through wireless communication, collaboratively sensing, collecting, and processing information within the monitored area, and then transmitting it to users. In WSNs, energy is a critical factor limiting network performance and lifespan. Sensor nodes are typically powered by batteries with limited energy, and recharging is difficult in many practical applications. Therefore, achieving efficient energy coverage, ensuring relatively uniform energy consumption across all nodes, and preventing some nodes from prematurely running out of energy are crucial for extending the overall lifespan of the WSN.

[0003] Traditional energy-efficient coverage algorithms for WSNs have many shortcomings. In large-scale network environments, traditional algorithms have high computational complexity, slow convergence speed, and cannot quickly find the optimal energy-efficient coverage scheme, making them unsuitable for applications with high real-time requirements.

[0004] Bacterial Foraging Algorithm (BFO), a swarm intelligence-based optimization algorithm, simulates the foraging behavior of bacteria in the environment and possesses certain global search capabilities and parallelism. However, traditional BFO ​​suffers from problems such as being prone to getting trapped in local optima and slow convergence speed when applied to energy-balanced scheduling of Wild-Type Networks (WSNs). For example, during the search process, individual bacteria may over-rely on local information, preventing them from finding the globally optimal solution and thus affecting the quality of energy-efficient coverage schemes. Therefore, improving BFO to better suit the characteristics of WSNs and making it more effective for energy-efficient coverage has become a key focus of current research.

[0005] In view of the above problems, this invention proposes an energy-efficient coverage method for WSN based on an improved bacterial foraging algorithm, aiming to effectively solve the problems of uneven energy consumption and poor algorithm performance in energy-efficient coverage of WSN. Summary of the Invention

[0006] The purpose of this invention is to provide a technical method to optimize the problems of uneven energy distribution and low coverage efficiency in existing WSN technologies. Specifically, it provides a WSN energy-efficient coverage method and apparatus based on a bacterial foraging optimization algorithm. This method optimizes the sensor's on / off combination scheme through an improved bacterial foraging optimization algorithm, thereby improving energy efficiency and the overall lifespan of the sensor matrix.

[0007] Technical Solution 1 is as follows: A high-efficiency energy coverage method for WSN based on an improved bacterial foraging algorithm, the main steps of which are as follows:

[0008] Step 1: Initialize the Network and Bacterial Population. The network and bacterial population are initialized using a location-based encoding method, with each bacterial individual representing a WSN target coverage scheme. The number of sensor nodes and monitoring targets in the WSN are determined, and the initial node scheduling matrix for individuals is initialized within the monitoring area based on coverage relationships. The population `pop` is initialized, where each individual corresponds to a node scheduling scheme matrix E.

[0009] Step 2: Calculate the fitness value of each individual and construct the fitness function, using the maximum number of consecutive coverage rounds as the metric. The monitoring effectiveness matrix ET is obtained by multiplying the node scheduling matrix E by the coverage relationship matrix T. The representations of matrices E, T, and ET are shown below: in, This means that in the i-th round, the m-th monitored target is monitored by at least one of the n sensors, and This means that the m-th monitored target in the i-th round was not monitored by any of the n sensors;

[0010] The fitness value, *fit*, is calculated by counting the maximum number of consecutive rows in the ET matrix where every element is greater than 0. The formula is as follows: fit = row_zero(ET) - 1

[0011] Where row_zero(ET) represents the row number in matrix ET where the first zero element appears, which is the number of rounds in which the sensor network cannot complete full coverage, and the number of rounds minus one is the working life of the wireless sensor network.

[0012] Step 3: Design and calculate the adaptive operator, defining the adaptive mutation probability α_bfo, which is related to the current iteration number gen, the estimated peak iteration number peak_gen, and the maximum iteration number MAXGEN, as shown in the following formula:

[0013] Where θ is a random number in the interval [0,1], and μ is a random number in the interval [0.1,0.5].

[0014] Before reaching the expected peak, as the number of iterations increases, the mutation probability α_bfo increases, which increases the randomness of the bacteria, enabling a global large-scale search, avoiding missing the optimal solution, and tending towards a local optimum. After reaching the expected peak, the mutation probability decreases, enabling a local detailed search, reducing time and space waste, and improving algorithm efficiency.

[0015] Step 4: Mutation-based chemotaxis. During bacterial chemotaxis, individuals update their position based on their own location and the globally optimal location, combined with a mutation mechanism, according to the mutation-based chemotaxis formula:

[0016] For a given matrix of individuals E∈{0,1} M×N Where M is the number of time slots, N is the number of sensors, the mutation probability α_bfo, and the number of working rounds D for each sensor, the formula for the mutated individual matrix E' is as follows:

[0017] Where θ is a random number in the interval [0,1], and M j ∈{0,1} M×1 The mutation mask vector of sensor j is defined as:

[0018] S j ={iIL i,j =1} is the set of time slots S' currently active for sensor j. j It is the new set of activated time slots after mutation, consisting of a portion of randomly changed time slots and randomly retained time slots. O represents element-wise multiplication.

[0019] Step 5: Elite reproduction operation. Sort by fitness value, eliminate the latter half of the bacteria, and allow the first half to divide and replicate. The formula is as follows:

[0020] Where P represents the new population after reproduction, α represents the elite retention rate, K represents the population size, [αK] represents the number of elites after rounding down, j is the index of the individual ranked j in fitness value, and U represents copying the elite set. Next, restore the population size to K.

[0021] Reproduction not only preserves elite individuals but also eliminates inferior ones, driving the entire microbial community towards a better state. The elite strategy also preserves the optimal fitness value in each iteration, preventing fitness values ​​from growing in the opposite direction.

[0022] Step 6: Eliminate diffusion operation. To eliminate the diffusion probability, Ped randomly resets the bacterial position, using the following formula: P'=PO(1-Δ)+X OΔ

[0023] Where P' is the new population after eliminating diffusion, P is the current population, X is the individual matrix generated according to the initialization function, and Δ is the mutation mask tensor.

[0024] δ is a random number in the interval [0,1].

[0025] Technical solution two is as follows:

[0026] The WSN energy-efficient coverage device based on the elite adaptive bacterial foraging optimization algorithm is used to execute the method described in technical solution one, and includes a processor, a memory, a data preprocessing module, a communication module, and an energy monitoring module.

[0027] The processor performs calculations on the improved bacterial foraging algorithm, including population generation, coverage optimization, fitness calculation, location updates, population updates, handling local optima, and strategy selection.

[0028] The memory stores node information, coverage schemes, calculation results, intermediate variables and parameters, providing data support.

[0029] The data preprocessing module reduces noise in the raw data (e.g., using Kalman filtering) and extracts key energy scheduling features, improving the efficiency and reliability of the algorithm.

[0030] Communication module: Enables data communication between sensor nodes and devices, and between devices and base stations. It adopts low-power wireless communication protocols (such as IEEE 802.15.4), and has adaptive power adjustment (dynamically matching node energy) and multi-channel communication capabilities to help optimize target coverage schemes.

[0031] Energy monitoring module: Monitors the energy consumption of sensor nodes in real time, calculates the remaining energy and feeds it back to the processor. It integrates an energy early warning function, which triggers an early warning when the node energy is lower than a preset threshold, providing real-time basis for adjusting the scheduling strategy.

[0032] The beneficial effects of this invention are as follows:

[0033] This invention employs an improved bacterial foraging algorithm for energy-efficient coverage of WSNs. In optimizing the energy-efficient coverage scheme, the algorithm continuously adjusts the scheme by simulating bacterial foraging behavior, effectively improving network coverage. Simultaneously, the fitness function uses the maximum number of consecutive coverage rounds as a metric, making the energy-efficient coverage result more reasonable, thereby reducing node energy consumption and improving energy efficiency. Regarding algorithm performance, the adaptive operator and improved bacterial foraging operation enhance the algorithm's optimization ability. Multiple migrations and local optimizations in the chemotaxis operation allow for fine-tuning of the energy-efficient coverage scheme; the elite retention strategy in the reproduction operation ensures the inheritance of excellent schemes; and the elimination diffusion operation randomly perturbs individuals when the algorithm may get stuck in a local optimum, expanding the population search range and effectively improving the algorithm's stability. Compared with existing technologies, this improved bacterial foraging algorithm for WSN energy-efficient coverage method and device exhibits superior performance in terms of network coverage and energy efficiency, faster algorithm convergence speed, and the ability to quickly find a better energy-efficient coverage scheme, demonstrating significant advantages and application value in the field of WSN energy management technology. Attached Figure Description

[0034] Figure 1 This diagram illustrates the overall process of EABFO in achieving efficient energy coverage of WSN.

[0035] Figure 2 : Schematic diagram of EABFO's WSN energy-efficient coverage device.

[0036] Figure 3 This is a schematic diagram comparing the fitness simulations of EABFO and the Genetic Algorithm (GA).

[0037] Figure 4 : This is a summary figure for a WSN energy-efficient coverage method and device based on EABFO. Detailed Implementation

[0038] To make the technical solutions and advantages of the embodiments of the present invention clearer, the present invention will be further described in detail below with reference to specific embodiments. Obviously, the described embodiments are only a part of the embodiments of the present invention, and not an exhaustive list of all embodiments. It should be noted that, unless otherwise specified, the embodiments and features in the embodiments of the present invention can be combined with each other.

[0039] Example 1: A method and apparatus for energy-efficient coverage scheduling of WSNs based on an improved bacterial foraging algorithm, specifically including the following steps:

[0040] Step 1: Initialize the Network and Bacterial Population. The network and bacterial population are initialized using a location-based encoding method, with each bacterial individual representing a WSN target coverage scheme. The number of sensor nodes and monitoring targets in the WSN are determined. Within the monitoring area, the initial node scheduling matrix for each individual is initialized based on the coverage relationship. The population `pop` is initialized, where each individual corresponds to a node scheduling scheme matrix E, initialized as follows:

[0041] First, an all-zero matrix E with N×10 rows and N columns is initialized, where N is the number of sensors. Then, each sensor (each row) is processed by selecting D unique integers from 1 to N×10, and then setting the value at the selected time slot position to 1, indicating that the sensor is activated in these time slots.

[0042] Step 2: Calculate the fitness value of each individual and construct the fitness function, using the maximum number of consecutive coverage rounds as the metric. The monitoring effectiveness matrix ET is obtained by multiplying the node scheduling matrix E by the coverage relationship matrix T. The representations of matrices E, T, and ET are shown below: in, This means that in the i-th round, the m-th monitored target is monitored by at least one of the n sensors, and This means that the m-th monitored target in the i-th round was not monitored by any of the n sensors;

[0043] The fitness value, *fit*, is calculated by counting the maximum number of consecutive rows in the ET matrix where every element is greater than 0. The formula is as follows: fit = row_zero(ET) - 1

[0044] Where row_zero(ET) represents the row number in matrix ET where the first zero element appears, which is the number of rounds in which the sensor network cannot complete full coverage, and the number of rounds minus one is the working life of the wireless sensor network.

[0045] Step 3: Design and calculate the adaptive operator, defining the adaptive mutation probability α_bfo, which is related to the current iteration number gen, the estimated peak iteration number peak_gen, and the maximum iteration number MAXGEN, as shown in the following formula:

[0046] Where θ is a random number in the interval [0,1], and μ is a random number in the interval [0.1,0.5].

[0047] Before reaching the expected peak, as the number of iterations increases, the mutation probability α_bfo increases, which increases the randomness of the bacteria, enabling a global large-scale search, avoiding missing the optimal solution, and tending towards a local optimum. After reaching the expected peak, the mutation probability decreases, enabling a local detailed search, reducing time and space waste, and improving algorithm efficiency.

[0048] Step 4: Mutation-based chemotaxis. During bacterial chemotaxis, individuals update their position based on their own location and the globally optimal location, combined with a mutation mechanism, according to the mutation-based chemotaxis formula:

[0049] For a given matrix of individuals E∈{0,1} M×N Where M is the number of time slots, N is the number of sensors, the mutation probability α_bfo, and the number of working rounds D for each sensor, the formula for the mutated individual matrix E' is as follows:

[0050] Where θ is a random number in the interval [0,1], and M j ∈{0,1} M×1 The mutation mask vector of sensor j is defined as:

[0051] S j ={i|L i,j =1} is the set of time slots S' currently active for sensor j. j It is the new set of activated time slots after mutation, consisting of a portion of randomly changed time slots and randomly retained time slots. ⊙ represents element-wise multiplication.

[0052] Step 5: Elite reproduction operation. Sort by fitness value, eliminate the latter half of the bacteria, and allow the first half to divide and replicate. The formula is as follows:

[0053] Where P represents the new population after reproduction, α represents the elite retention rate, K represents the population size, [αK] represents the number of elites after rounding down, and j is the index of the individual ranked j in fitness. This indicates that the elite collection will be copied. Next, restore the population size to K.

[0054] Reproduction not only preserves elite individuals but also eliminates inferior ones, driving the entire microbial community towards a better state. The elite strategy also preserves the optimal fitness value in each iteration, preventing fitness values ​​from growing in the opposite direction.

[0055] Step 6: Eliminate diffusion operation. To eliminate the diffusion probability, Ped randomly resets the bacterial position, using the following formula: P'=P⊙(1-Δ)+X⊙Δ

[0056] Where P' is the new population after eliminating diffusion, P is the current population, X is the individual matrix generated according to the initialization function, and Δ is the mutation mask tensor.

[0057] δ is a random number in the interval [0,1].

[0058] Specifically, in this implementation case, the WSN energy balance scheduling method based on the improved bacterial foraging algorithm has a detection range of a square area with a side length of 400, a number of monitored targets of 100, a number of sensors of 120, a sensor monitoring radius of 300, and the monitored targets and sensor node positions are randomly distributed. The bacterial population size is 80, the number of iterations is 500, and the initial mutation probability of the adaptive operator is 0.2.

[0059] refer to Figure 3 The comparative algorithm for the WSN energy balancing scheduling method based on the Improved Bacterial Foraging Algorithm (EABFO) is the Genetic Algorithm (GA). Both algorithms have a population size of 80, 120 sensors, 500 iterations, and a monitoring range of a square with sides of 400m. EABFO's adaptive adjustment follows a peak curve, while GA follows a smooth parabolic model combined with a roulette wheel selection method. Simulation results show that EABFO plateaus around 380 iterations, achieving approximately 120 rounds of optimal continuous full coverage. GA, however, fails to plateau until around 500 iterations, with only about 40 rounds achieving optimal continuous full coverage. The EABFO-based WSN energy balancing scheduling method outperforms GA in terms of convergence speed and optimal fitness. Compared to GA, the EABFO-based method exhibits better convergence performance, lower energy consumption, and longer lifespan.

[0060] Implementation Case 2: A WSN energy-efficient coverage device based on an improved bacterial foraging optimization algorithm, used to execute the method described in Technical Solution 1, including a processor, memory, data preprocessing module, communication module, and energy monitoring module: The processor performs calculations on the improved bacterial foraging algorithm, including population generation, coverage optimization, fitness calculation, location updates, population updates, handling local optima, and strategy selection.

[0061] The memory stores node information, coverage schemes, calculation results, intermediate variables and parameters, providing data support.

[0062] The data preprocessing module reduces noise in the raw data (e.g., using Kalman filtering) and extracts key energy scheduling features, improving the efficiency and reliability of the algorithm.

[0063] Communication module: Enables data communication between sensor nodes and devices, and between devices and base stations. It adopts low-power wireless communication protocols (such as IEEE 802.15.4), and has adaptive power adjustment (dynamically matching node energy) and multi-channel communication capabilities to help optimize target coverage schemes.

[0064] Energy monitoring module: Monitors the energy consumption of sensor nodes in real time, calculates the remaining energy and feeds it back to the processor. It integrates an energy early warning function, which triggers an early warning when the node energy is lower than a preset threshold, providing real-time basis for adjusting the scheduling strategy.

[0065] Although the invention has been described with reference to a limited number of embodiments, those skilled in the art will understand from the foregoing description that other embodiments are conceivable within the scope of the invention described herein. Furthermore, it should be noted that the language used in this specification has been chosen primarily for readability and instructional purposes, and not for the purpose of interpreting or limiting the subject matter of the invention. Therefore, many modifications and variations will be apparent to those skilled in the art without departing from the scope and spirit of the appended claims. The disclosure of the invention is illustrative and not restrictive, and the scope of the invention is defined by the appended claims.

Claims

1. An improved WSN energy-efficient target coverage method for bacterial foraging algorithms, characterized in that, include, S1: Initialize the network and bacterial population; S2: Construct the fitness function and calculate the fitness value; S3: Design adaptive operators; S4: Mutant chemotaxis operation; S5: Elite Breeding Operation; S6: Eliminate diffusion operation.

2. The method for initializing bacterial populations and constructing an energy-efficient target coverage model for WSN according to claim 1, characterized in that, S1: Initialize the network and bacterial population: Adopt an encoding method based on the working status of sensor nodes. Each individual represents a WSN target coverage scheme. Determine the number of sensor nodes and the number of monitored targets in the WSN. In the monitoring area, initialize the initial node scheduling matrix of the individual according to the coverage relationship.

3. The WSN energy-efficient target coverage method for improving bacterial foraging algorithm according to claim 2, characterized in that, The monitoring effectiveness matrix ET is obtained by multiplying the node scheduling matrix E with the coverage relationship matrix T, where the matrices E, T and ET are represented as follows: in, This means that in the i-th round, the m-th monitored target is monitored by at least one of the n sensors, and This means that in the i-th round, the m-th monitored target was not monitored by any of the n sensors.

4. The WSN energy-efficient target coverage method for improving bacterial foraging algorithm according to claim 1, characterized in that, In S2, a fitness function is constructed, where `fit` is measured by the maximum number of consecutive coverage rounds, and is determined through: fit = row_zero(ET) - 1 The value is obtained, where row_zero(ET) represents the row number in matrix ET where the first zero element appears, i.e., the number of rounds in which the sensor network cannot complete full coverage, and the number of rounds minus one is the working life of the wireless sensor network.

5. The WSN energy-efficient target coverage method for an improved bacterial foraging algorithm according to claim 1, characterized in that, In S3, an adaptive operator is designed. The mutation rate α_bfo is related to the current iteration number gen, the estimated peak iteration number peak_gen, and the maximum iteration number MAXGEN, as shown in the following formula: Where θ is a random number in the interval [0,1] and μ is a random number in the interval [0.1,0.5].

6. The WSN energy-efficient target coverage method for an improved bacterial foraging algorithm according to claim 1, characterized in that, In the mutation chemotaxis operation in S4, individual bacteria update their positions based on their own position and the global optimal position, combined with the mutation mechanism. The formula for the individual matrix E' after mutation according to the mutation chemotaxis formula is as follows: Where for a given individual matrix L∈{0,1} M×N Where M is the number of time slots, N is the number of sensors, and D is the number of working wheels for each sensor, θ is a random number in the interval [0,1], and M j ∈{0,1} M×1 The mutation mask vector of sensor j is defined as: S j ={i|L i,j =1} is the set of time slots S' currently active for sensor j. j It is the new set of activated time slots after mutation, consisting of a portion of randomly changed time slots and randomly retained time slots. ⊙ represents element-wise multiplication.

7. The WSN energy-efficient target coverage method for an improved bacterial foraging algorithm according to claim 1, characterized in that, In S5, the elite reproduction operation sorts bacteria by fitness value, eliminates the latter half of the bacteria, and allows the former half to divide and replicate. The formula is as follows: Where Q' represents the new population after reproduction, α is the elite retention rate, K is the population size, [αK] represents the number of elites after rounding down, and j is the index of the individual ranked j in fitness. This indicates that the elite collection will be copied. Next, restore the population size to K.

8. The WSN energy-efficient target coverage method for an improved bacterial foraging algorithm according to claim 1, characterized in that, In S6, the diffusion elimination operation randomly resets the bacterial position using the elimination diffusion probability Ped, as shown in the following formula: P'=P⊙(1-Δ)+X⊙Δ Where P' represents the new population after eliminating diffusion, P represents the current population, X represents the new individual matrix generated according to the initialization function, which replaces the original individual matrix E, and Δ represents the mutation mask tensor. δ is a random number in the interval [0,1].

9. A WSN energy-efficient target coverage device for implementing the improved bacterial foraging algorithm of claims 1-8, characterized in that, Includes a processor, memory, data preprocessing module, communication module, and energy monitoring module: The processor performs calculations on the improved bacterial foraging algorithm, including population generation, coverage optimization, fitness calculation, location updates, population updates, handling local optima, and strategy selection. The memory stores node information, coverage schemes, calculation results, intermediate variables and parameters, providing data support. The data preprocessing module reduces noise in the raw data (e.g., using Kalman filtering) and extracts key energy scheduling features, improving the efficiency and reliability of the algorithm. Communication module: Enables data communication between sensor nodes and devices, and between devices and base stations. It adopts low-power wireless communication protocols (such as IEEE 802.15.4), and has adaptive power adjustment (dynamically matching node energy) and multi-channel communication capabilities to help optimize target coverage schemes. Energy monitoring module: Monitors the energy consumption of sensor nodes in real time, calculates the remaining energy and feeds it back to the processor. It integrates an energy early warning function, which triggers an early warning when the node energy is lower than a preset threshold, providing real-time basis for adjusting the scheduling strategy.

Images