Improving source localization in NLOS conditions via ranging contraction

In this paper a novel distance-based source localization algorithm is proposed that is effective in minimizing the error due to biased measurements. In particular, we show how to exploit the knowledge of the feasibility region, constructed via trilateration, to contract the measured distances such that the cost-function of the LS formulation becomes convex and the global optimum is closer to the true location. The proposed ranging contraction source-localization algorithm is shown via simulations to overperform existing alternatives, such as the unconstrained and the constrained LS approach. The results also show that the localization error performance of the proposed technique remains close to the theoretical position error bound.