NLOS Error Mitigation in Hybrid RSS-TOA-Based Localization Through Semi-Definite Relaxation

摘要

This letter focuses on the accurate target localization in non-line of sight (NLOS) environment through effective hybridization of time of arrival (TOA) and received signal strength (RSS) measurements. A non-linear weighted least square (NLWLS) problem is formulated from general approximation of hybrid maximum-likelihood (ML) estimation problem using balancing parameter i.e., mean NLOS effect. The NLWLS problem is then solved using semi-definite relaxation (SDR) technique which employs a novel convex hull constraint. The proposed method requires neither prior NLOS statistics nor NLOS path identification. The numerical simulations validate that the proposed method significantly outperforms the existing methods and achieves the performance comparatively closer to that of Cramer-Rao lower bound (CRLB).