Key Technologies of Precise Target Localization under the Influence of Multipath and Non-line-of-sight
Multipath and non-line-of-sight(NLoS)propagation make it difficult for a localization system to accurately measure the distance between transmitting and receiving nodes,seriously affecting positioning accuracy,even leading to positio-ning failure.This article proposes the unified positioning framework in multipath and NLoS scenarios,analyzes the basic princi-ples of multipath and NLoS and their impact on target positioning,and then reviews the mainstream methods of LoS beacons i-dentification,NLoS paths identification and corresponding localization methods,including:① identification of NLoS beacons,including distance measurement based identification and residual test based identification;② NLoS path recognition in line of sight beacons,including recognition based on statistical features,recognition based on CSI information,recognition based on tra-ditional machine learning,and recognition based on neural networks;③ multipath and NLoS assisted target localization,inclu-ding localization using the multipath path of line of sight beacons,localization using the ranging results of NLoS beacons,and localization using only NLoS beacons;④ blind localization in multipath and NLoS scenarios,including localization based on the directly extracted phase information,localization based on filtering methods.Finally,taking unmanned driving in underground mines as an example,application methods of the target localization framework in multipath and NLoS scenarios are analyzed,and future study directions are discussed.
multipath effectnon-line-of-sightchannel state informationdeep learningprecise positioningunmanned driving
万方数据
维普
