Research On A Class Of Sensor Target Tracking Problem Based On Swarm Intelligence Combined Algorithm

With the further improvement of information technology,the application of wireless sensor networks with measurement and control properties has become more and more extensive,and the target tracking problem based on wireless sensor networks has gradually become a research hotspot.Target tracking is a persistent task,and the target positioning state at each time step has a huge impact on the effectiveness of tracking.The essence of the tracking problem is to find a reasonable task allocation scheme,if all the sensor nodes are put into the working state during the tracking process,a high-precision tracking result will inevitably be obtained,but in this case,the energy loss of the entire network will also be very serious.How to effectively reduce energy consumption while maintaining tracking accuracy can be embodied in the solution of multi-objective optimization problems,and intelligent optimization algorithms with a lot of experience can perfectly deal with the above problems.In this paper,particle swarm optimization(PSO)and beetle antennae search(BAS)are improved respectively,and a population intelligent combination algorithm is proposed and applied to target tracking problem.The specific contents are as follows:(1)New inertia weights and exponential decreasing step factor strategies are used to improve PSO and BAS,respectively.A new inertia weight in the form of linear differential decreasing fusion with nonlinear decline is proposed to change the original inertia weight in the PSO algorithm to a fixed value,which solves the problem of unclear optimization and slow optimization speed due to the change trend of inertia weight not conforming to the transformation mode of the optimization process.The exponentially decreasing step factor is proposed to solve the problem that the local search and global search cannot be properly coordinated due to the unreasonable setting of the step size factor in the BAS algorithm.(2)Based on different strategies,a swarm intelligence combination algorithm based on PSO and BAS is proposed.Combining improved PSO and improved BAS,a new iterative mode is constructed,and when the improved PSO algorithm finds the local optimum,the improved BAS algorithm continues to search on both sides of the local optimum,which solves the problem of particle diversity decline and optimization accuracy reduction caused by particle range concentration in the later stage of algorithm optimization.By comparing the performance of three advanced optimization algorithms on 10 benchmark functions with different characteristics,the results show the superiority of the proposed algorithm.(3)The model of sensor target tracking problem is established,and the proposed swarm intelligent combination algorithm is used to solve the problem and the results are analyzed.A comprehensive optimization index of mapping tracking accuracy and energy loss was constructed to solve the problem of short operating life or low tracking accuracy of sensor networks caused by unreasonable index construction.The combined algorithm is applied to the target tracking problem,and the optimal subset of the task performed by the sensor is selected,and the experimental results are analyzed and compared to verify the effectiveness of the improved algorithm.