Research On Multiple Target Localization Of Wireless Sensor Network Based On Compressive Sensing

With the development of wireless communication technology,more and more people pay attention to location-based services.The wireless sensor network localization method based on the received signal strength is widely used in the industrial field due to its simple principle and low cost.The compressive sensing theory proposed in recent years is applied in the positioning field,which reduces the processing of redundant data and improves the positioning effect.On this basis,this paper makes an in-depth study on the multiple target localization based on compressed sensing under the two conditions of known target number and unknown target number.Firstly,in the case that the number of targets is known,a regularized increasing support set subspace pursuit algorithm is proposed for target positioning.This algorithm retains the idea of iterative backtracking in Subspace Pursuit(SP)algorithm and incorporates regularization method to obtain the real support set with a higher probability,which further improves the reconstruction probability and accuracy.Secondly,in the case of unknown target number,this paper conducts research from three directions:For dictionary mismatch scenario multi-objective location problem,put forward the Greedy Matching Pursuit(GMP)back multiresolution analysis algorithm based on polar decomposition,the algorithm would observe the dictionary very decomposition,fusion using multi-resolution analysis to locate,the experimental results show that the algorithm is compared with the requirements of GMP,the algorithm can effectively enhance the catch ability of the,has higher positioning accuracy.Under the scenario of positioning accuracy is not high,is given based on the reconstruction of power field compression perception multi-target localization algorithm,set for sparse matrix Discrete Cosine Transform(DCT)matrix,perception matrix column units,from the reconstruction of the superposition of power in the maximum present in the field of find the target location and gradually divest the power of each of the target field to target count and positioning,the experimental results show that the overall count performance is better than GMP algorithm,and has certain stability and incomparable response speed,at the same time,the operation is simple,the algorithm reduces the amount of calculation,also reduced the information transmission frequency of the sensor,prolong the service life of the sensor.Under the integration of intelligent algorithm,binary salp swarm algorithm used to compress the perception of the multiple objective positioning,the continuous salp swarm algorithm of discrete in binary space,keep salp group of the essential characteristic of rapid changes of coordination and foraging,solves the continuous salp swarm algorithm reconstruction discrete signal low accuracy and search problems for a longer time,has realized the multiple objective positioning,the simulation results show that compared with the traditional algorithm,this algorithm although time cost slightly more,but have more superior resistance to noise,under the condition of same counting has better performance and positioning performance.