CN116208969B - Wireless sensor network coverage optimization method based on improved marine predator algorithm - Google Patents

Wireless sensor network coverage optimization method based on improved marine predator algorithm Download PDF

Info

Publication number
CN116208969B
CN116208969B CN202310221067.7A CN202310221067A CN116208969B CN 116208969 B CN116208969 B CN 116208969B CN 202310221067 A CN202310221067 A CN 202310221067A CN 116208969 B CN116208969 B CN 116208969B
Authority
CN
China
Prior art keywords
predator
algorithm
iter
prey
node
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Active
Application number
CN202310221067.7A
Other languages
Chinese (zh)
Other versions
CN116208969A (en
Inventor
金子龙
蒋嘉乐
潘成胜
孔志翔
钱钧
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Nanjing University of Information Science and Technology
Original Assignee
Nanjing University of Information Science and Technology
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Nanjing University of Information Science and Technology filed Critical Nanjing University of Information Science and Technology
Priority to CN202310221067.7A priority Critical patent/CN116208969B/en
Publication of CN116208969A publication Critical patent/CN116208969A/en
Application granted granted Critical
Publication of CN116208969B publication Critical patent/CN116208969B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Classifications

    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04W—WIRELESS COMMUNICATION NETWORKS
    • H04W16/00—Network planning, e.g. coverage or traffic planning tools; Network deployment, e.g. resource partitioning or cells structures
    • H04W16/18—Network planning tools
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F17/00—Digital computing or data processing equipment or methods, specially adapted for specific functions
    • G06F17/10—Complex mathematical operations
    • G06F17/16—Matrix or vector computation, e.g. matrix-matrix or matrix-vector multiplication, matrix factorization
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/004—Artificial life, i.e. computing arrangements simulating life
    • G06N3/006—Artificial life, i.e. computing arrangements simulating life based on simulated virtual individual or collective life forms, e.g. social simulations or particle swarm optimisation [PSO]
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L41/00—Arrangements for maintenance, administration or management of data switching networks, e.g. of packet switching networks
    • H04L41/14—Network analysis or design
    • H04L41/142—Network analysis or design using statistical or mathematical methods
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04W—WIRELESS COMMUNICATION NETWORKS
    • H04W16/00—Network planning, e.g. coverage or traffic planning tools; Network deployment, e.g. resource partitioning or cells structures
    • H04W16/22—Traffic simulation tools or models
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04W—WIRELESS COMMUNICATION NETWORKS
    • H04W24/00—Supervisory, monitoring or testing arrangements
    • H04W24/02—Arrangements for optimising operational condition
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04W—WIRELESS COMMUNICATION NETWORKS
    • H04W84/00—Network topologies
    • H04W84/18—Self-organising networks, e.g. ad-hoc networks or sensor networks
    • Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
    • Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
    • Y02D—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN INFORMATION AND COMMUNICATION TECHNOLOGIES [ICT], I.E. INFORMATION AND COMMUNICATION TECHNOLOGIES AIMING AT THE REDUCTION OF THEIR OWN ENERGY USE
    • Y02D30/00—Reducing energy consumption in communication networks
    • Y02D30/70—Reducing energy consumption in communication networks in wireless communication networks

Landscapes

  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Mathematical Physics (AREA)
  • Computer Networks & Wireless Communication (AREA)
  • Signal Processing (AREA)
  • Theoretical Computer Science (AREA)
  • Data Mining & Analysis (AREA)
  • Mathematical Analysis (AREA)
  • Mathematical Optimization (AREA)
  • Pure & Applied Mathematics (AREA)
  • Computing Systems (AREA)
  • General Engineering & Computer Science (AREA)
  • Computational Mathematics (AREA)
  • Algebra (AREA)
  • Software Systems (AREA)
  • Computational Linguistics (AREA)
  • Health & Medical Sciences (AREA)
  • Evolutionary Computation (AREA)
  • General Health & Medical Sciences (AREA)
  • Molecular Biology (AREA)
  • Probability & Statistics with Applications (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Biophysics (AREA)
  • Biomedical Technology (AREA)
  • Artificial Intelligence (AREA)
  • Databases & Information Systems (AREA)
  • Management, Administration, Business Operations System, And Electronic Commerce (AREA)

Abstract

The invention discloses a wireless sensor network coverage optimization method based on an improved marine predator algorithm, which aims at a target network monitoring area in a fixed range, takes the number of sensor nodes, the deployment positions of the sensor nodes and the uniform sensing radius and sensing radius of the sensor nodes in the target network monitoring area as inputs, and takes the network coverage rate of the target network monitoring area as output to construct a sensor node joint sensing model based on Boolean sensing; and (2) taking the current two-dimensional deployment position of the sensor node and the network coverage rate of a target network monitoring area as inputs, and optimizing the parameters to be optimized of the sensor node joint perception model obtained in the step (S1) by utilizing a marine predator algorithm and combining with Tent chaos to obtain an optimized maximum network coverage rate and an optimal node position deployment scheme corresponding to the optimized maximum network coverage rate. The improved marine predator algorithm has better optimization effect, more uniform network node distribution and less coverage blind area and node coverage redundancy.

Description

Wireless sensor network coverage optimization method based on improved marine predator algorithm
Technical Field
The invention relates to the field of wireless communication, in particular to a wireless sensor network coverage optimization method based on an improved marine predator algorithm.
Background
With the development of new technologies such as artificial intelligence, big data, cloud computing and 5G, the Internet of things has become a research hotspot of new generation information technologies. As one of core technologies for supporting the internet of things, WSNs (Wireless Sensor Networks ) play an indispensable role in the fields of military defense, industrial and agricultural control, smart cities, biomedicine, and the like. However, the problem of coverage optimization in the WSN has the problems of uneven distribution of initial node positions, blind coverage or redundancy, and the like, and self-adaptive redeployment adjustment is needed to repair coverage gaps, so that the coverage rate of the network is improved, and the method has important research significance for prolonging the service life of the network and improving the reliability of the network.
The current research methods related to the problem of WSN coverage optimization in the field can be generally divided into a traditional method and an intelligent optimization algorithm. Conventional approaches include force-based techniques, computational geometry-based techniques, and mesh-based techniques. Although the force-based technology can reduce the moving distance, the force-based technology is not suitable for the network with strong topological time-varying property; the technique based on computational geometry reduces coverage blind areas by determining the center movement of the polygon, but fails to consider the convergence problem of the algorithm; grid-based techniques are suitable for network coverage problems for mobile nodes, but the consumption of network communications is significant. The intelligent optimization algorithm is an approximate optimization algorithm which refers to the bionics idea and is widely used for solving high-dimensional optimization problems and various complex engineering problems, and the application of the intelligent optimization algorithm to WSN coverage optimization problems has become a hot spot in recent years. Wang Zhendong et al propose an enhanced sparrow search algorithm based on cauchy variation improvement that can achieve higher network coverage while maintaining high convergence speed; li Shouyu et al propose an improved balance optimizer algorithm based on reverse learning and dynamic sine and cosine factors, which improves node coverage optimization compared with other improved algorithms, and can effectively reduce deployment cost; song Tingting et al propose an improved whale algorithm based on the Lewy flight that can greatly improve network coverage compared to the original algorithm. However, the intelligent optimization algorithm has a slower convergence speed, and the overall optimization effect is still to be improved.
Disclosure of Invention
The technical problems to be solved by the invention are as follows: in a network monitoring area, in order to meet the monitoring requirement, a mode of randomly scattering sensor nodes is generally adopted, so that the conditions of uneven distribution of initial positions of the nodes, coverage dead areas or redundancy and the like occur in partial areas easily, and the problem of massive resource waste is caused due to low network coverage rate. The invention aims at: the wireless sensor network coverage optimization method based on the improved marine predator algorithm is provided, and the problems in the prior art are solved.
In order to achieve the above purpose, the present invention provides the following technical solutions: the wireless sensor network coverage optimization method based on the improved marine predator algorithm comprises the following steps:
step S1, aiming at a target network monitoring area in a fixed range, taking the number of sensor nodes, the deployment positions of the sensor nodes and the uniform sensing radius and sensing radius of the sensor nodes in the target network monitoring area as inputs, and taking the network coverage rate of the target network monitoring area as output, and constructing a sensor node joint sensing model based on Boolean sensing; the construction of the sensor node joint perception model based on Boolean perception comprises the following substeps:
step S2, the network coverage rate of the current two-dimensional deployment position of the sensor node and the target network monitoring area is taken as input, the parameters to be optimized of the sensor node joint perception model obtained in the step S1 are optimized by utilizing a marine predator algorithm and combining with Tent chaos, the optimized maximum network coverage rate and the optimal node position deployment scheme corresponding to the maximum network coverage rate are obtained, and in the step S1, the sensor node joint perception model based on Boolean perception is constructed, and the method comprises the following substeps:
s101, based on the target network monitoring area being s=l 1 ×l 2 Is divided into l in a discrete way 1 ×l 2 Each monitoring point grid is assembled intoWherein the monitoring point m j The position coordinates of (2) are { x } j ,y j }，i∈(1,l 1 ×l 2 )，The geometric center point of each monitoring point is a coverage optimization target position;
s102, based on sensor node Z= { Z 1 ,z 2 ,…,z i ,...,z n -wherein node z i The position coordinates of (2) are { x } i ,y i I epsilon (1, n), unifying the perception radius of each node as R s The communication radius is R c The method comprises the steps of carrying out a first treatment on the surface of the And R is c ＝2R s ；
S103, defining a sensor node z by adopting a Boolean perception model i With monitoring point m j Is defined by the following formula:
s104, determining that the monitoring point grid is covered: monitoring point m j Node z i When the distance of the monitoring point grid is not greater than the node perceived radius, the monitoring point grid is considered to be covered as follows
S105, calculating monitoring points m of all sensor nodes in the target network monitoring area j Joint perceptual probability C of (2) p The following formula:
s106, according to joint perception probability C p Calculating the total coverage rate R of the target monitoring area cov The following formula:
further, the step S2 includes the following sub-steps:
s201, setting parameters to be optimized of a sensor node joint perception model as population individuals, wherein predators and preys are all search individuals, creating initial preys in a population initialization mode, and constructing elites by the predators; wherein Prey is a matrix of n multiplied by d, n is the population number, and d is the population dimension, namely the two-dimensional coordinate number of the sensor node; based on prey individuals, using the network coverage rate of the node joint perception model as an fitness function, reserving population individuals corresponding to the minimum fitness function value, and copying the population individuals n times to form a predator matrix Elite; prey individuals were population initialized as follows:
X＝X min +R(X max -X min ) (5)
wherein X= [ X ] 1 ,X 2 ,…,X i ,…,X N ]，i∈[0,N]，X i ＝[X i,1 ,X i,2 ,…,X i,j ,…,X i,D ]，j∈[0,D]，X i,j Representing the position of population individuals i in the j-th dimension, wherein N is the population scale, and D is the search dimension; r is a random number and R.epsilon.0, 1]，X max and Xmin The upper and lower bounds of the search space, respectively;
the predator matrix Elite is defined as follows:
prey matrix is defined as follows:
s202, updating a Prey matrix based on preset iteration times, and iteratively updating and finally outputting an predator matrix Elite based on various groups of individuals in the Prey matrix Prey: according to the moving speed of predators and prey, the iteration cycle of the marine predator algorithm can be divided into three phases, wherein Iter is the current iteration number, and Max_Iter is the maximum iteration number:
(1) The first stage: the prey moves faster than the predator, and when Iter < Max_Iter/3, the updated predator position changes by:
wherein ,for the population of individuals, move step length, +.>Is based on a vector which is not normally distributed and contains random numbers to represent Brownian motion; r is a random number vector and R.epsilon.0, 1]，/>For predator position,/->P is a constant for the prey location;
(2) And a second stage: the predator and the prey move at the same speed, the prey moves through the Lewy flight, the predator moves through the Brownian motion, and when Max_Iter/3 is less than or equal to Iter is less than or equal to 2Max_Iter/3, the position changes as follows:
wherein ,is a random number vector based on the Lewye distribution; CF= (1-Iter/Max_Iter) (2Iter/Max_Iter) For controlling the movement step of predators;
(3) And a third stage: predators move faster than prey, and move through Laiyyvern flights, when Iter > 2Max_Iter/3, the position changes as follows:
step S203, adopting jump to avoid local optimal stagnation, and adopting marine predator algorithm jump update to predators, wherein the following formula is as follows:
wherein FADs represent the probability that the effect of the FADs of the vortex formation or fish gathering device affects the optimization process, r is a random number and rE [0,1 ]]，Representing a binary vector 0 or 1,/and-> and />Two random prey positions are shown.
Further, in the aforementioned step S201, the following formula is included to initialize the hunt population by using the sequence generated by Tent chaos:
wherein ,Zt Is interval [0,1 ]]T epsilon [0, max_Iter)]The method comprises the steps of carrying out a first treatment on the surface of the Alpha is between [0,1 ]]Then mapping the chaotic sequence to a solution space according to the following formula to obtain the initial position of the prey population:
X i,j ＝X min,j +Z t ×(X max,j -X min,j ) (14)
further, in the aforementioned step S203, an adaptive jump control factor ω is introduced to control the algorithm jump step size, as follows:
ω＝|2 [-4*Iter/(3*Max_Iter)] ×cos(πt)| (15)
by introducing the self-adaptive jump step length factor, the algorithm has a larger step length factor in the early stage of iteration, so that the global searching capability of the algorithm can be improved, and the smaller step length factor in the later stage of iteration can further enhance the local searching capability, thereby being beneficial to better jumping out of local optimization of the algorithm.
Further, the wireless sensor network coverage optimization method based on the improved marine predator algorithm further comprises the step of introducing a double subgroup position update strategy before each iteration of the algorithm is finished, and specifically comprises the following steps:
step A: before each algorithm iteration is finished, sequencing the population in sequence from low to high according to the fitness value, and naming the subgroup in the first half of sequencing as a leader subgroup; the subgroup in the second half of the rank is named follower subgroup;
and (B) step (B): the t distribution mutation operator taking the current iteration number Iter as a degree of freedom parameter carries out disturbance update on the position of the leader subgroup, and the position of the leader subgroup is updated as follows:
X′ i ＝X i +X i ·t(Iter),0≤i≤N/2 (17)
wherein ,X′i A new location for the ith leader individual;
further, the wireless sensor network coverage optimization method based on the improved marine predator algorithm adopts a dynamic learning strategy to update the positions of the leader subgroups, and the method comprises the following steps:
X′ i ＝X best -X mean -R 1 (X min +R 2 (X max -X min )),N/2≤i≤N (18)
X i ' New location, X, of the ith follower individual best X is the global optimal individual position mean R is the average of all individual positions globally 1 and R2 Is [0,1]A random value uniformly distributed thereon.
Further, in the aforementioned step S201, when the predator position is updated in the first stage, p=0.5.
Further, in the aforementioned step S203, fads=0.2 when the predator is updated by using the marine predator algorithm in the third stage.
Further, the α is 0.7.
Compared with the prior art, the invention has the following beneficial effects:
in the aspect of simulation experiments, the improved marine predator algorithm is firstly tested by 4 groups of reference optimization functions, the superiority of improvement measures is verified, and the algorithm is superior to other 4 common intelligent optimization algorithms in aspects of convergence accuracy, speed and the like; then the improved marine predator algorithm is applied to wireless sensor coverage optimization, the average coverage rate in three network coverage scenes of 20m multiplied by 20m, 50m multiplied by 50m and 100m multiplied by 100m can reach 97.03 percent, 97.49 percent and 99.79 percent respectively, compared with the original algorithm and other 6 improved algorithms, the improved marine predator algorithm has better optimization effect, more uniform network node distribution and less coverage blind area and node coverage redundancy phenomenon; finally, the coverage condition of the improved marine predator algorithm after the initial node deployment quantity is reduced is tested, and the improved marine predator algorithm can save 8%, 12.5% and 20% of node deployment cost under three network coverage scenes under the condition of keeping the network coverage similar to other optimization algorithms.
Drawings
Fig. 1 is a flow chart of a wireless sensor network coverage optimization method based on an improved marine predator algorithm of the present invention.
Fig. 2 is a schematic diagram of a model of the present invention.
Fig. 3 is a diagram of adaptive jump step size factor proposed by the present invention.
Fig. 4 shows t distribution diagrams under different degrees of freedom parameters according to the present invention.
FIG. 5 is a graph comparing convergence curves of reference function tests of other algorithms according to the present invention; in the figure, (a) is F 1 A comparison graph of the convergence curve of the function test, (b) is F 2 A comparison graph of the convergence curve of the function test, wherein (c) is F 3 A comparison graph of the convergence curve of the function test, wherein (d) is F 4 And (5) comparing the function test convergence curve with a graph.
FIG. 6 is a graph of node distribution in a 20m monitored coverage scene in accordance with the present invention; in the figure, (a) is a random shed node profile, (b) is an MPA node profile, (c) is an EMPA node profile, and (d) is an EMPA (n=22) node profile.
FIG. 7 is a graph showing node distribution in a 50m monitored coverage scene in accordance with the present invention; in the figure, (a) is a random shed node profile, (b) is an MPA node profile, (c) is an EMPA node profile, and (d) is an EMPA (n=35) node profile.
FIG. 8 is a graph showing node distribution in a 100m monitoring coverage scenario according to the present invention; in the figure, (a) is a random shed node profile, (b) is an MPA node profile, (c) is an EMPA node profile, and (d) is an EMPA (n=40) node profile.
Detailed Description
For a better understanding of the technical content of the present invention, specific examples are set forth below, along with the accompanying drawings.
Aspects of the invention are described herein with reference to the drawings, in which there are shown many illustrative embodiments. The embodiments of the present invention are not limited to the embodiments described in the drawings. It is to be understood that this invention is capable of being carried out by any of the various concepts and embodiments described above and as such described in detail below, since the disclosed concepts and embodiments are not limited to any implementation. Additionally, some aspects of the disclosure may be used alone or in any suitable combination with other aspects of the disclosure.
According to the invention, MPA (Marine Predators Algorithm, marine predator algorithm) is improved, a wireless sensor network coverage optimization method based on EMPA (Enhanced Marine Predators Algorithm, improved marine predator algorithm) is constructed, aiming at a target monitoring area, the number of sensor nodes, the deployment positions of the sensor nodes and the uniform sensing radius and sensing radius of the sensor nodes in the target network monitoring area are taken as inputs, and the network coverage rate of the target network monitoring area is taken as output, so that a sensor node joint sensing model based on Boolean sensing is constructed; and then, taking the two-dimensional deployment position of the current sensor node and the network coverage rate of a target network monitoring area as inputs, and optimizing the parameters to be optimized of the sensor node joint perception model obtained in the step S1 by utilizing a marine predator algorithm and combining with Tent chaos to obtain the optimized maximum network coverage rate and the optimal node position deployment scheme corresponding to the optimized maximum network coverage rate.
As shown in fig. 1, specifically: (1) and (5) setting parameters. The number of sensor nodes, the range of the monitored area, the perceived radius, the communication radius, the maximum number of iterations, and the population of the improved marine predator algorithm are entered. (2) An objective function is determined. And adjusting the fitness function into the non-coverage rate of the WSN monitoring area, randomly initializing the positions of the sensor nodes of the group, and calculating the coverage rate of the initial area according to the joint perception model. (3) And initializing a population. And determining an initial population according to the introduced Tent chaotic map, and accelerating the early convergence speed of the algorithm. (4) An initial prey matrix and elite matrix are determined. And calculating a fitness value of the hunting matrix, recording the optimal position, and calculating an elite matrix. (5) And (5) updating the prey. Adopting different updating modes in different iteration periods, and adopting Brownian motion to perform global search on preiteration prey; the population is divided into two parts in the iterative metaphase, wherein the prey makes Lewy flight, the algorithm is responsible for developing in the search space, the predator makes Brownian motion, and the algorithm is responsible for exploring in the search space; and the predator adopts the Levin motion strategy to perform local search in the later iteration stage. (6) Solving the vortex formation and FADs effect. And updating all population positions according to the jump step length control factors, and jumping out of the local optimum. (7) A double subgroup location update strategy is introduced. The population is divided into a leader subgroup and a follower subgroup according to the low-to-high arrangement of fitness values, and the population positions are updated based on adaptive t distribution variation and dynamic learning respectively. (8) Iter=Iter+1, and if Iter is equal to or less than Max_Iter, the process returns to step (5). (9) And outputting the optimal node deployment scheme. And solving the coverage rate of each group of network node deployment schemes according to the joint perception coverage formula, and outputting an optimal node deployment scheme corresponding to the current maximum network coverage rate.
As shown in fig. 2, according to the model schematic diagram of the present invention, constructing a sensor node joint perception model based on boolean perception includes:
(1) Coverage area modeling
Monitoring area based on target network as s=l 1 ×l 2 Is divided into l in a discrete way 1 ×l 2 Each monitoring point grid is assembled intoWherein the monitoring point m j The position coordinates of (2) are { x } j ,y j }，i∈(1,l 1 ×l 2 ) The geometric center point of each monitoring point is the coverage optimization target position;
(2) Sensor node modeling
Based on sensor node z= { Z 1 ,z 2 ,…,z i ,…,z n -wherein node z i The position coordinates of (2) are { x } i ,y i I epsilon (1, n), unifying the perception radius of each node as R s The communication radius is R c The method comprises the steps of carrying out a first treatment on the surface of the And R is c ＝2R s To ensure connectivity of the wireless sensor network.
(3) Boolean perception model
Defining a sensor node z by using a boolean awareness model i With monitoring point m j Is defined by the following formula:
(4) Coverage calculation
Determining that the monitoring point grid is covered: monitoring point m j Node z i When the distance of the monitoring point grid is not greater than the node perceived radius, the monitoring point grid is considered to be covered as follows
Calculating the monitoring point m of all sensor node pairs in the monitoring area of the target network j Joint perceptual probability C of (2) p The following formula:
according to joint perception probability C p Calculating the total coverage rate R of the target monitoring area cov The following formula:
the invention takes the two-dimensional deployment position of the current sensor node and the network coverage rate of the target network monitoring area as inputs, optimizes the parameters to be optimized of the sensor node joint perception model by utilizing the marine predator algorithm and combining with Tent chaos, and obtains the optimized maximum network coverage rate and the optimal node position deployment scheme corresponding to the maximum network coverage rate specifically as follows:
s201, setting parameters to be optimized of a sensor node joint perception model as population individuals, wherein predators and preys are all search individuals, creating initial preys in a population initialization mode, and constructing elites by the predators; wherein Prey is a matrix of n multiplied by d, n is the population number, and d is the population dimension, namely the two-dimensional coordinate number of the sensor node; based on prey individuals, using the network coverage rate of the node joint perception model as an fitness function, reserving population individuals corresponding to the minimum fitness function value, and copying the population individuals n times to form a predator matrix Elite; prey individuals were population initialized as follows:
X＝X min +R(X max -X min ) (5)
wherein X= [ X ] 1 ,X 2 ,…,X i ,…,X N ]，i∈[0,N]，X i ＝[X i,1 ,X i,2 ,…,X i,j ,…,X i,D ]，j∈[0,D]，X i,j Representing the position of population individuals i in the j-th dimension, wherein N is the population scale, and D is the search dimension; r is a random number and R.epsilon.0, 1]，X max and Xmin The upper and lower bounds of the search space, respectively;
the predator matrix Elite is defined as follows:
prey matrix is defined as follows:
s202, updating a Prey matrix based on preset iteration times, and iteratively updating and finally outputting an predator matrix Elite based on various groups of individuals in the Prey matrix Prey: according to the moving speed of predators and prey, the iteration cycle of the marine predator algorithm can be divided into three phases, wherein Iter is the current iteration number, and Max_Iter is the maximum iteration number:
(1) The first stage: the prey moves faster than the predator, and when Iter < Max_Iter/3, the updated predator position changes by:
wherein ,for moving step length of population individual, R B Is based on a vector which is not normally distributed and contains random numbers to represent Brownian motion; r is a random number vector and R.epsilon.0, 1]，/>For predator position,/->For the prey location, P is a constant and can take a value of 0.5.
(2) And a second stage: the predators and the prey move at the same speed, the marine predator optimizing process is converted from exploration to development, and half of the marine predator optimizing process is used for exploration and is generally used for development, and the predator is mainly responsible for exploration and development. The prey moves through the Lewy flight, the predator moves through the Brownian motion, and when Max_Iter/3 is less than or equal to Iter and less than or equal to 2Max_Iter/3, the position changes as follows:
wherein ,is a random number vector based on the Lewye distribution; CF= (1-Iter/Max_Iter) (2Iter/Max_Iter) For controlling the movement step of predators;
(3) And a third stage: predators move faster than prey, and move through Laiyyvern flights, when Iter > 2Max_Iter/3, the position changes as follows:
fig. 3 is an adaptive jump step size factor graph. The figure shows the change in the adaptive jump step size factor, which decreases non-linearly from 1 to 0.4 as the number of iterations increases. The larger step factor exists in the early stage of algorithm iteration, so that the global searching capability of the algorithm can be improved, and the smaller step factor in the later stage of algorithm iteration can further enhance the local searching capability, so that the algorithm can better jump out of local optimum. Therefore, the jump step length of the algorithm is controlled by introducing the self-adaptive cosine weight as a jump step length control factor, so that the balance algorithm is explored and developed, and the capability of the algorithm to jump out of local optimum is improved.
Jump is adopted to avoid local optimal stagnation, marine predator algorithm jump update is adopted for predators, and an adaptive jump control factor omega is introduced, wherein the following formula is as follows:
ω＝|2 [-4*Iter/(3*Max_Iter)] ×cos(πt)| (12)
wherein FADs represent the probability that the effect of the FADs of the vortex forming or fish gathering device affects the optimization process, and can take a value of 0.2; r is a random number and r.epsilon.0, 1]，Representing a binary vector 0 or 1,/and-> and />Two random prey positions are shown.
The sequence generated by Tent chaos is used for initializing the hunt population, and the following formula is shown:
wherein ,Zt Is interval [0,1 ]]T epsilon [0, max_Iter)]The method comprises the steps of carrying out a first treatment on the surface of the Alpha is between [0,1 ]]Is 0.7 here.
Then mapping the chaotic sequence to a solution space according to the following formula to obtain the initial position of the prey population:
X i,j ＝X min,j +Z t ×(X max,j -X min,j ) (15)
fig. 4 is a t-profile for different degrees of freedom parameters.
The improved algorithm, after considering the influence of environmental factors such as fish gathering devices (FADs), sorts the population in order from low to high according to fitness values before each algorithm iteration ends, and names the subgroup of the first half of the sorting as the leader subgroup; the subgroup in the second half of the rank is named follower subgroup. The leader subgroup has better optimizing capability, but is easy to fall into local optimum in the later iteration period. The method adopts the t distribution mutation operator taking the current iteration number (Iter) as the degree of freedom parameter to update the position of the leader subgroup, integrates the advantages of Gaussian distribution and Cauchy distribution, and improves the global development capacity of the algorithm in the early stage of iteration and the local exploration capacity of the algorithm in the later stage of iteration.
The location of the leader subgroup is updated as follows:
X′ i ＝X i +X i ·t(Iter),0≤i≤N/2 (16)
wherein ,X′i The self-adaptive disturbance is performed near the individual for the new position of the ith leader individual by taking the ith leader individual as a reference, so that not only is the excellent information of the leader per se saved, but also the population diversity can be kept, and the leader jumps out of the local optimum. In the early iteration stage, the t distribution is similar to the Cauchy distribution, the step size adopted by the position variation is larger, the algorithm has good global exploration capacity, and the global searching characteristic of the first stage of EMPA is considered; in the middle iteration stage, t distribution is between cauchy distribution and normal distribution, and the step size adopted by the position variation of a leader subgroup is relatively large, so that the global search and the local development of the second stage of EMPA can be simultaneously considered; in the later iteration stage, the t distribution is similar to the standard normal distribution, the step size adopted by the position variation of the leader subgroup is smaller, and the local development characteristic of the third stage of EMPA can be considered. The follower subgroup has excellent optimizing capability, but is easy to fall into local optimum in the later iteration stage, and the position of the leader subgroup is updated based on a dynamic learning strategy.
The optimization capability of the follower subgroup is enhanced by adding the update of global optimum and average individuals and setting the upper and lower boundaries of the dynamic boundary, the iterative optimization time of an algorithm is reduced, and the following formula is adopted:
X′ i ＝X best -X mean -R 1 (X min +R 2 (X max -X min )),N/2≤i≤N (17)
X i ' New location, X, of the ith follower individual best X is the global optimal individual position mean R is the average of all individual positions globally 1 and R2 Is [0,1]A random value uniformly distributed thereon.
FIG. 5 is a graph comparing the convergence curves of the reference function tests of the algorithms. To verify the optimal performance and effectiveness of the improved marine predator algorithm in network coverage problems,4 representative intelligent optimization algorithms are selected to perform a benchmark function optimization performance test; at the same time select 2 single peaks (F 1 、F 2 ) And 2 multimodal (F3, F4) benchmark test functions are subjected to numerical simulation, the specific function expression, dimension, definition domain and theoretical optimal values of the benchmark test functions are shown in a benchmark test function table in table 1, comparison (30 rounds) of benchmark function optimization results in table 2 gives average values and ranks of 6 optimization algorithms on 4 benchmark test functions after 30 rounds of simulation experiments, and it can be seen that the optimization results of the EMPA algorithm on 4 benchmark functions are all the best.
TABLE 1
TABLE 2
In terms of the unimodal benchmark function, as can be seen from the two graphs (a) and (b) in fig. 5, the MPA algorithm can maintain a higher convergence rate in the middle before iteration, but falls into local optimum in the later stage; the EMPA can hardly fall into local optimum in the whole iteration process, the convergence speed and the precision are higher than those of other algorithms, especially in the 900 th iteration of F1 function test, the magnitude is improved by 100-200 orders of magnitude compared with other algorithms, and finally, the theoretical optimum value 0 of the F1 and F2 unimodal functions can be reached in the 923 rd iteration and 1187 th iteration respectively. The improved strategy can effectively improve the convergence rate of the algorithm, and has excellent unimodal optimizing capability. For the multimodal reference test function with a large number of local optimal values, as can be seen from fig. 5 (c), the convergence accuracy of the EMPA algorithm on the F3 function is not much different from that of other algorithms, only a few orders of magnitude are improved, but the convergence speed of the algorithm in the early stage of iteration is obviously higher than that of other algorithms, and the effectiveness of introducing the Tent chaotic map is verified. In fig. 5, (d) shows the comparison of the F4 function test convergence curves, and it can be seen that the convergence accuracy of the EMPA only at the 300 th iteration can almost reach the theoretical optimum 12569.5, and both the convergence speed and the convergence accuracy are far higher than those of the MPA algorithm, which proves that the improvement measures can better balance the local and global searches, and further enhance the optimizing capability of the algorithm.
Fig. 6 is a graph of node distribution in a 20m x 20m monitored coverage scene. The three graphs (a), (b) and (c) in fig. 6 respectively describe node distribution diagrams of the monitoring area after random scattering, MPA algorithm and EMPA algorithm optimization in a 20m×20m scene. In fig. 6, (a) it can be seen that a large area coverage blind area occurs in the monitoring area by randomly scattering the nodes, wherein the large area coverage blind area occurs in the middle part, the lower left and the lower right of the area, and a large amount of node coverage redundancy phenomenon exists in the left and upper right corner areas; as can be seen from fig. 6 (b), the coverage dead zone in the monitoring area deployed by the general MPA algorithm is smaller, but a serious node coverage redundancy phenomenon occurs in the lower right area; in fig. 6, (c) optimizing node deployment using the EMPA algorithm is more uniform than node deployment using the ordinary MPA algorithm, with very little coverage dead zone area. In addition, in fig. 6 (d), the number of network nodes is reduced to 22 by using the EMPA algorithm to perform network coverage optimization, so that coverage dead zones and coverage redundancy phenomena in a target area are less, coverage rate can reach 95.32%, and coverage effect is far more than that of MPA algorithm deployment. Under the scene, compared with other optimization algorithms, the EMPA algorithm can save the deployment cost of 2 nodes and ensure the coverage rate of similar networks.
Fig. 7 is a graph of node distribution in a 50m×50m monitored overlay scene. As can be seen from the two graphs (a) and (b) in fig. 7, by adopting the common MPA algorithm to optimize node deployment, compared with random node deployment, the coverage area of the coverage dead zone is small, and the coverage rate of the node network is greatly improved, but more serious node coverage redundancy still exists in the monitoring area at the lower left. As can be seen from fig. 7 (c), deployment of nodes in this scenario using the EMPA algorithm achieves substantially full network coverage, with only slight coverage redundancy on the left side of the monitored area. Based on this, fig. 7 (d) shows the final node distribution diagram of the EMPA algorithm for coverage optimization using only 35 nodes in this scenario, and it can be seen that the coverage redundancy phenomenon of the target monitoring area is less, and the coverage rate can reach 94.60% higher than that of the MPA algorithm. Therefore, the EMPA algorithm can save the deployment cost of 5 nodes compared with other optimization algorithms under the scene, and can achieve coverage effects similar to other improvement algorithms.
Fig. 8 is a node distribution diagram under a 100m×100m monitoring coverage scene. As can be seen from the two diagrams (b) and (c) in fig. 8, the optimized node deployment using the EMPA algorithm is more uniform than the node deployment using the ordinary MPA algorithm, and almost no coverage dead zone exists. In fig. 8 (d), the number of network nodes is reduced to 40 by using the EMPA algorithm to perform network coverage optimization, so that it can be seen that there are no serious coverage dead zones and redundancy phenomena in the monitored area in the scene, and the coverage rate of the final nodes can reach 98.61%, so that the network full coverage is basically achieved. Compared with other algorithms, the EMPA algorithm can ensure similar network coverage rate under the condition of saving one fifth of node deployment cost.
While the invention has been described in terms of preferred embodiments, it is not intended to be limiting. Those skilled in the art will appreciate that various modifications and adaptations can be made without departing from the spirit and scope of the present invention. Accordingly, the scope of the invention is defined by the appended claims.

Claims (7)

1. The wireless sensor network coverage optimization method based on the improved marine predator algorithm is characterized by comprising the following steps of:
step S1, aiming at a target network monitoring area in a fixed range, taking the number of sensor nodes, the deployment positions of the sensor nodes and the uniform sensing radius and sensing radius of the sensor nodes in the target network monitoring area as inputs, and taking the network coverage rate of the target network monitoring area as output, and constructing a sensor node joint sensing model based on Boolean sensing; the construction of the sensor node joint perception model based on Boolean perception comprises the following substeps:
s101, based on target networkThe area of the monitoring network is s=l 1 ×l 2 Is divided into l in a discrete way 1 ×l 2 Each monitoring point grid is assembled intoWherein the monitoring point m j The position coordinates of (2) are { x } j ,y j }，i∈(1,l 1 ×l 2 ) The geometric center point of each monitoring point is the coverage optimization target position;
s102, based on sensor node Z= { Z 1 ,z 2 ,...,z i ,...,z n -wherein node z i The position coordinates of (2) are { x } i ,y i I epsilon (1, n), unifying the perception radius of each node as R s The communication radius is R c The method comprises the steps of carrying out a first treatment on the surface of the And R is c ＝2R s ；
S103, defining a sensor node z by adopting a Boolean perception model i With monitoring point m j Is defined by the following formula:
s104, determining that the monitoring point grid is covered: monitoring point m j Node z i When the distance of the monitoring point grid is not greater than the node perceived radius, the monitoring point grid is considered to be covered as follows
S105, calculating monitoring points m of all sensor nodes in the target network monitoring area j Joint perceptual probability C of (2) p The following formula:
s106, according to the joint perception probabilityRate C p Calculating the total coverage rate R of the target network monitoring area cov The following formula:
step S2, optimizing parameters to be optimized of the sensor node joint perception model obtained in the step S1 by using a current two-dimensional deployment position of the sensor node and network coverage rate of a target network monitoring area as inputs and combining a Tent chaos by using a marine predator algorithm to obtain an optimized maximum network coverage rate and an optimal node position deployment scheme corresponding to the optimized maximum network coverage rate;
s201, setting parameters to be optimized of a sensor node joint perception model as population individuals, wherein predators and preys are all search individuals, creating initial preys in a population initialization mode, and constructing elites by the predators; wherein Prey is a matrix of n multiplied by d, n is the population number, and d is the population dimension, namely the two-dimensional coordinate number of the sensor node; based on prey individuals, using the network coverage rate of the node joint perception model as an fitness function, reserving population individuals corresponding to the minimum fitness function value, and copying the population individuals n times to form a predator matrix Elite; prey individuals were population initialized as follows:
X＝X min +R(X max -X min ) (5)
wherein X= [ X ] 1 ,X 2 ,…,X i ,…,X N ]，i∈[0,N]，X i ＝[X i,1 ,X i,2 ,…,X i,j ,…,X i,D ]，j∈[0,D]，X i,j Representing the position of population individuals i in the j-th dimension, wherein N is the population scale, and D is the search dimension; r is a random number and R.epsilon.0, 1]，X max and Xmin The upper and lower bounds of the search space, respectively;
the predator matrix Elite is defined as follows:
prey matrix is defined as follows:
s202, updating a Prey matrix based on preset iteration times, and iteratively updating and finally outputting an predator matrix Elite based on various groups of individuals in the Prey matrix Prey: according to the moving speed of predators and prey, the iteration cycle of the marine predator algorithm can be divided into three phases, wherein Iter is the current iteration number, and Max_Iter is the maximum iteration number:
(1) The first stage: the prey moves faster than the predator, and when Iter < Max_Iter/3, the updated predator position changes by:
wherein ,for the population of individuals, move step length, +.>Is based on a vector which is not normally distributed and contains random numbers to represent Brownian motion; r is a random number vector and R.epsilon.0, 1]，/>For predator position,/->P is a constant for the prey location;
(2) And a second stage: the predator and the prey move at the same speed, the prey moves through the Lewy flight, the predator moves through the Brownian motion, and when Max_Iter/3 is less than or equal to Iter is less than or equal to 2Max_Iter/3, the position changes as follows:
wherein ,is a random number vector based on the Lewye distribution; CF= (1-Iter/Max_Iter) (2Iter/Max_Iter) For controlling the movement step of predators;
(3) And a third stage: predators move faster than prey, and move through Laiyyvern flights, when Iter > 2Max_Iter/3, the position changes as follows:
step S203, adopting jump to avoid local optimal stagnation, and adopting marine predator algorithm jump update to predators, wherein the following formula is as follows:
wherein FADs represent the probability that the effect of the FADs of the vortex formation or fish gathering device affects the optimization process, r is a random number and rE [0,1 ]]，Representing a binary vector 0 or 1,/and-> and />Representing two random prey locations;
an adaptive jump step factor omega is introduced in the jump phase as follows:
ω＝|2 [-4*Iter/(3*Max_Iter)] ×cos(πt)| (13)
2. the method of claim 1, wherein in step S201, the method comprises initializing a prey population using a sequence generated by Tent chaos, wherein the following formula is:
wherein ,Zt Is interval [0,1 ]]T epsilon [0, max_Iter)]The method comprises the steps of carrying out a first treatment on the surface of the Alpha is between [0,1 ]]Then mapping the chaotic sequence to a de-null according to the following equationObtaining an initial position of the prey population:
X i,j ＝X min,j +Z t ×(X max,j -X min,j ) (16)。
3. the method for optimizing coverage of a wireless sensor network based on an improved marine predator algorithm according to claim 1, further comprising introducing a double subgroup position update strategy before the end of each iteration of the algorithm, comprising the steps of:
step A: before each algorithm iteration is finished, sequencing the population in sequence from low to high according to the fitness value, and naming the subgroup in the first half of sequencing as a leader subgroup; the subgroup in the second half of the rank is named follower subgroup;
and (B) step (B): the t distribution mutation operator taking the current iteration number Iter as a degree of freedom parameter carries out disturbance update on the position of the leader subgroup, and the position of the leader subgroup is updated as follows:
X′ i ＝X i +X i ·t(Iter),0≤i≤N/2 (17)
wherein ,X′i A new location for the ith leader individual.
4. A wireless sensor network coverage optimization method based on an improved marine predator algorithm according to claim 3, characterized in that the leader subgroup is updated with a location based on a dynamic learning strategy, as follows:
X′ i ＝X best -X mean -R 1 (X min +R 2 (X max -X min )),N/2≤i≤N (18)
X′ i new location, X, for the ith follower individual best X is the global optimal individual position mean R is the average of all individual positions globally 1 and R2 Is [0,1]A random value uniformly distributed thereon.
5. The method of optimizing coverage of a wireless sensor network based on an improved marine predator algorithm according to claim 1, wherein in step S201, p=0.5 when the predator position is updated in the first stage.
6. The method for optimizing coverage of a wireless sensor network based on an improved marine predator algorithm according to claim 1, wherein fads=0.2 when the predator is updated with a marine predator algorithm jump in the third stage in step S203.
7. The method for optimizing coverage of a wireless sensor network based on an improved marine predator algorithm according to claim 2, wherein α is 0.7.
CN202310221067.7A 2023-03-09 2023-03-09 Wireless sensor network coverage optimization method based on improved marine predator algorithm Active CN116208969B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202310221067.7A CN116208969B (en) 2023-03-09 2023-03-09 Wireless sensor network coverage optimization method based on improved marine predator algorithm

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202310221067.7A CN116208969B (en) 2023-03-09 2023-03-09 Wireless sensor network coverage optimization method based on improved marine predator algorithm

Publications (2)

Publication Number Publication Date
CN116208969A CN116208969A (en) 2023-06-02
CN116208969B true CN116208969B (en) 2023-11-03

Family

ID=86509341

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202310221067.7A Active CN116208969B (en) 2023-03-09 2023-03-09 Wireless sensor network coverage optimization method based on improved marine predator algorithm

Country Status (1)

Country Link
CN (1) CN116208969B (en)

Families Citing this family (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN116916475B (en) * 2023-08-10 2024-05-07 华东交通大学 Wireless sensor network deployment method based on multi-strategy improved badger algorithm
CN117062092B (en) * 2023-10-11 2023-12-08 成都大学 Wireless sensor network deployment method

Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN110062389A (en) * 2019-04-19 2019-07-26 江西理工大学 Sensor network nodes Optimization deployment method based on improved differential evolution algorithm
CN112333810A (en) * 2020-11-03 2021-02-05 温州大学 TMPA algorithm-based hierarchical wireless sensor network topology optimization method
CN115243273A (en) * 2022-09-23 2022-10-25 昆明理工大学 Wireless sensor network coverage optimization method, device, equipment and medium

Family Cites Families (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN109447359B (en) * 2018-11-06 2021-04-16 成都信息工程大学 Data acquisition point deployment method and system

Patent Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN110062389A (en) * 2019-04-19 2019-07-26 江西理工大学 Sensor network nodes Optimization deployment method based on improved differential evolution algorithm
CN112333810A (en) * 2020-11-03 2021-02-05 温州大学 TMPA algorithm-based hierarchical wireless sensor network topology optimization method
CN115243273A (en) * 2022-09-23 2022-10-25 昆明理工大学 Wireless sensor network coverage optimization method, device, equipment and medium

Non-Patent Citations (2)

* Cited by examiner, † Cited by third party
Title
Improved Marine Predator Algorithm for Wireless Sensor Network Coverage Optimization Problem;Qing He 等;Sustainability;摘要，第2-3节 *
基于改进MPA 的分层无线传感器网络优化部署;陈龙 等;传感技术学报;第1-2节 *

Also Published As

Publication number Publication date
CN116208969A (en) 2023-06-02

Similar Documents

Publication Publication Date Title
CN116208969B (en) Wireless sensor network coverage optimization method based on improved marine predator algorithm
CN110996287B (en) Network node selection method, system and storage medium based on whale optimization algorithm
CN103744428B (en) A kind of unmanned surface vehicle paths planning method based on neighborhood intelligent water drop algorithm
CN105975342B (en) Based on the cloud computing method for scheduling task and system for improving cuckoo searching algorithm
Jun et al. Multi-objective mobile robot path planning based on improved genetic algorithm
Udgata et al. Sensor deployment in irregular terrain using artificial bee colony algorithm
CN102843743B (en) To look for food based on flora the wireless sense network Tiered routing method of optimized algorithm
CN107229287A (en) A kind of unmanned plane global path planning method based on Genetic Ant algorithm
Wang et al. AFSAOCP: a novel artificial fish swarm optimization algorithm aided by ocean current power
CN111432368A (en) Ranging and positioning method suitable for sparse anchor node WSN
CN113905389A (en) Wireless sensor network coverage method based on particle swarm optimization imperial butterfly algorithm
CN114489061A (en) Whale optimization algorithm-based smooth path planning method
Pan et al. The phasmatodea population evolution algorithm and its application in 5g heterogeneous network downlink power allocation problem
CN116528252A (en) Wireless sensor network coverage optimization method based on improved white bone top chicken algorithm
CN115866621A (en) Wireless sensor network coverage method based on whale algorithm
Al Aghbari et al. An adaptive coverage aware data gathering scheme using kd-tree and aco for wsns with mobile sink
CN116866951A (en) Sensor network coverage optimization method and system based on improved artificial buzzer algorithm
CN114980134B (en) Mobile multi-target dynamic coverage method based on virtual force
Wu et al. Consensus based distributive task allocation for multi-AUV in searching and detecting
Kumar et al. Spider monkey optimization algorithm in data science: a quantifiable objective study
Lu et al. Enhancing sensor duty cycle in environmental wireless sensor networks using Quantum Evolutionary Golden Jackal Optimization Algorithm
CN115169754A (en) Energy scheduling method and device, electronic equipment and storage medium
Yu et al. Optimization of WSN localization algorithm based on improved multi-strategy seagull algorithm
CN111062515B (en) Distribution network distributed power supply configuration method
Zhao et al. Opposition-Based Artificial Bee Colony Using Different Learning Models.

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
GR01 Patent grant
GR01 Patent grant