TDOA based approach for accurate target localization based on hybrid genetic algorithm

Abstract

Accurate localization of target based on time difference of arrival (TDOA) measurements is of crucial importance in a large number of different military and civil applications, especially in security systems, radars, sonars etc. This paper focuses on the determining the position of a target from a set of TDOA measurements obtained on several receivers whose positions are known. The considered target localization problem is formulated as the optimization problem, where the corresponding objective function is obtained based on least squares (LS) method. Due to the complexity of the considered problem, the resulting objective function is highly nonlinear and multimodal. Therefore, to solve this complex optimization problem this paper proposes the hybridization of Genetic Algorithm (GA) with well-known Gauss-Newton (GN) method. The performance of considered hybrid algorithm is investigated and compared to well-known conventional optimization algorithms in solving the considered TDOA based localization problem. The simulation results of the proposed optimization method indicate a significant improvement in localization accuracy compared to well-known algorithms.