NLOS mitigation in TOA-based localization using semidefinite programming

In this work, time-of-arrival (TOA)-based wireless sensor localization in non-line-of-sight (NLOS) environments is investigated. In such environments, the accuracy of localization techniques is significantly degraded. While previous work often assumes some knowledge of the NLOS environment, we assume that the estimator knows neither which connections are NLOS nor the distribution of the NLOS errors. It is shown that the maximum likelihood estimator using only LOS connections provides a lower bound on the estimation accuracy. Furthermore, a novel NLOS mitigation technique based on semidefinite programming (SDP) is proposed. The proposed SDP technique estimates the source location jointly with the NLOS biases. The performance of the proposed estimator is compared with the aforementioned lower bound and with previous algorithms through computer simulations. Simulation results show that the proposed SDP estimator outperforms the other algorithms substantially, especially in severe NLOS environments.