Research On Target Localization And Tracking Algorithm Based On Wireless Sensor Networks

As a new interdisciplinary subject,Wireless sensor network integrates many technologies,such as microelectronics,wireless communication,data transmission and so on.It has wide application prospect in various fields.In recent years,due to the characteristics of small size,low cost and low energy consumption,WSN has a prominent position in the areas of public security,intelligent life,military defense and other fields.Target location and tracking is one of the most important applications of WSN.Because the movement of the target is nonlinear,it is very difficult to achieve better tracking.At present,the WSN target tracking algorithm has many shortcomings,such as low tracking precision,high complexity and poor real-time performance.How to achieve accurate,fast and stable target tracking has become the orientation of the majority of researchers.First of all,the target localization algorithm is used as the breakthrough point and three kinds of target location algorithm based on RSSI ranging are introduced.Meanwhile,it is pointed out that the evaluation index of the performance of the algorithm are accuracy and complexity.The simulation of static target location and moving target tracking shows that the RSSI-TL algorithm has good accuracy,real-time and stable tracking process.The accuracy of RSSI-WC algorithm is second only to RSSI-TL algorithm.In addition,the RSSI-MLE algorithm has a large error and poor accuracy.Secondly,four kinds of nonlinear filtering target tracking algorithms are discussed.The tracking performance of MLKF,EKF,UKF and PF algorithm is compared and analyzed through the simulation of linear uniform motion target and nonlinear moving target.In the process of linear uniform motion target tracking,the tracking accuracy of PF is the highest,while the EKF is the worst.Among them,the MLKF algorithm has the shortest tracking time,but the PF tracking is the longest.With the increase of the number of nodes,the tracking accuracy of MLKF,PF and UKF algorithm is gradually improved,while the tracking accuracy of EKF algorithm is firstly decreased and then increased.In the nonlinear variable motion in the process of target tracking,tracking accuracy and real-time performance of the MLKF algorithm is better than other algorithms.However,tracking accuracy and real-time performance of PF algorithm is the worst.At the same time,the number of sensor nodes has little effect on the tracking accuracy of MLKF and PF algorithm.Thirdly,the CV,CA,CT three mathematical target motion model are analyzed and IMMPF maneuvering target tracking algorithm is proposed.The tracking performance of the MLKF,PF and IMMPF algorithm is analyzed and compared by the simulation of the CV-CT target motion model.The simulation results show that the tracking accuracy of the IMMPF algorithm,which has better real-time performance,is significantly higher than that of the MLKF and PF algorithm.With the increase of the total number of samples,the tracking accuracy of PF and IMMPF algorithm is improved,but the tracking time is also increased.Then,an improved algorithm is proposed to solve the problem of particle degeneracy in the tracking process of IMMPF algorithm.It uses the MLKF algorithm to filter in the linear module and uses the improved residual resampling of PF algorithm to filter in the nonlinear module.At last,the predicted value of the target is output by weighted average.The tracking performance of MLKF,RSSI-TL,IMMPF and improved algorithm is compared and analyzed by the simulation of CA-CV and CA-CT-CV model.The validity of the improved algorithm is verified.The results show that the tracking accuracy of the improved algorithm is the highest,while the MLKF is the worst,Among them,the RSSI-TL algorithm has the shortest tracking time.At the same time,the number of sensor nodes has little effect on the tracking accuracy of the improved algorithm.Finally,Summarizing the content of the article and pointing out that it is necessary to do further research on the realization of accurate,fast and stable target tracking under the condition of bad environment and complex target motion.