TOA Based Non-cooperative And Cooperative Source Localization In Mixed LOS/NLOS Environments

Source localization has been widely used in the fields of target tracking,aviation navigation,and emergency rescue.A crucial performance indicator for source localization is the localization accuracy problem,Source localization accuracy is mainly affected by noise,node position errors,non-line-of-sight(NLOS)errors,and other factors,where the effect of NLOS errors caused by blocked signals is particularly significant.In the existing research,the methods for processing NLOS errors are mainly divided into two categories: one method is the state of known paths and statistical information of NLOS errors,line-of-sight(LOS)and NLOS measurement information is processed separately.Such methods can achieve high-precision localization,however,these methods require a lot of testing to obtain this information,which is difficult to achieve in a time-varying urban environment;the other methods are completely unknown of the path state and NLOS errors statistics,these methods do not require prior information,however,the existing methods are difficult to achieve high-precision localization at the same time in sparse and dense NLOS environments.In view of the above difficulties,the main study results of this paper are as follows:(1)In this paper,the problem of non-cooperative localization based on time of arrival(TOA)under mixed LOS and NLOS environments is addressed.In this paper,a new robust weighted least squares(RWLS)problem is established by introducing a balancing parameter related to NLOS errors,the RWLS problem takes unknown target positions and NLOS balancing parameter as estimated variables.By using S-Lemma,the RWLS problem is transformed into a non-convex optimization problem,and then it is relaxed into a convex semidefinite programming(SDP)problem.The proposed method does not require the statistical information and path state of NLOS errors.Simulation results show that the method can achieve high localization accuracy in both sparse and dense NLOS environments.(2)For the problem of cooperative localization of multiple fixed targets in mixed LOS/NLOS environments,under the cases of the NLOS errors statistics and path state information are completely unknown,this paper proposes a cooperative RWLS problem in mixed LOS/NLOS environments to locate multiple fixed targets.This paper introduces some balancing parameters to alleviate the problem of the large NLOS errors upper bounds that may exist in existing robust methods,and establishes a non-convex RWLS problem to jointly estimate the position of the targets and balancing parameters.Finally,a semidefinite relaxation(SDR)method is used to obtain a semidefinite and second-order cone programming(SD/SCOP)problem to approximate the non-convex problem.Simulation results and real data show that this method performs better than the existing cooperative localization methods with multiple fixed targets.(3)For the problem of cooperative localization of multiple moving targets in mixed LOS/NLOS environments,where the velocity is assumed to be fixed within a short time period.Under the cases of the NLOS errors statistics and path state information are completely unknown,this paper proposes a cooperative RWLS problem in mixed LOS/NLOS environments to locate multiple moving targets.In this problem,the idea of multiple fixed targets localization is extended to multiple moving targets localization.This paper introduces some balancing parameters to alleviate the problem of the large NLOS upper bounds that may exist in existing robust methods,and establishes a non-convex RWLS problem to jointly estimate target positions,target velocities,and balancing parameters.Finally,a SDR method is used to obtain a SD/SCOP problem to approximate the non-convex problem.Simulation results show that this method performs better than the existing cooperative localization methods with multiple moving targets.