Efficient Localization based on Information Fusion in Vehicular Ad-Hoc Networks

This paper reinvestigate the cooperative localization problem based on the information fusion of the GPS and the time of arrival (TOA) measurements in vehicular ad-hoc networks. The formulation based on the Maximum-Likelihood method is a non-convex and NP-hard problem. The existed semidefinite relaxation (SDR) method can provide an approximate solution but with quite poor efficiency. Therefore, an improved SDR method with local TOA information fusion is proposed in this paper, which can efficiently decrease the computational complexity. The Monte Carlo simulations are carried out and the results demonstrate that the localization accuracy of the proposed method is nearly the same with that of the SDR method without local fusion, but its computation time is considerably less than that of the previous SDR method.