Study On Robust NLOS Location Algorithms Based On TOA

Target localization plays an increasingly important role in military and civilian applications such as military operations,emergency rescue and response,seismic prospecting,aeronautical navigation,and most particularly location-based services.With the deepening of the research on target localization,some difficult problems are exposed.One of the most challenging problems is the non-line-of-sight(NLOS)propagation of the signal,and how to suppress the NLOS errors' impact to the localization accuracy in the severe NLOS environment becomes a hot and difficult problem to be addressed.Although there are many algorithms at present,quite a lot of literatures depend on assumptions such as known the NLOS status or the distribution of the NLOS error.However,the existing methods are usually assumed that all or part of the statistics of the NLOS errors is known,or known the NLOS status.However,it is impractical to obtain the statistics of NLOS errors in a complex and changeable wireless environment,and it could not avoid the occurrence of decision error when judging the path state.In addition,the existing methods are difficult to meet at the same time between lower computational complexity and robust localization performance.In view of these shortcomings,the main research results are as follows:(1)In this paper,we propose a robust second-order cone relaxation method for noncooperative positioning in NLOS conditions.Our method does not require any statistics of the NLOS errors,only need to know its upper bound,which is easier to obtain in practice.In addition,the method is not sensitive to the detection of the NLOS status,in another words,when the detection error occurs,the positioning performance is still robust.The simulation results show that the proposed method performs better than other algorithms in severe NLOS scenarios,and it has lower complexity and shorter average solution time.(2)In order to solve the problem of cooperative localization in NLOS conditions,this paper formed it as a robust least squares problem.Since the problem is non-convex,it can be solved by relaxing the problem into a mixed semi-definite and second-order cone programming problems.The method also does not require any statistics of the NLOS errors,only need to know its upper bound.The cooperative localization not only utilizes the connection between the unknown target and the anchor node,but also utilizes the connection between unknown targets to improve the localization performance in terms of accuracy and range.Simulation results show that the proposed algorithm is not sensitive to the NLOS error and performs well in severe NLOS environments.