A Hybrid Algorithm of Source Localization Based on Hyperbolic Technique in WSN

Electromagnetic (EM) source localization has become a vital issue lately to improve civilian safety, increase the military security and mitigate the disaster effect. The localization is embedded in battle-field surveillance, traffic alert, emergency call 911 (E911), resource allocation and mitigation of disaster effect. The location of the emitter can be determined by bestowing its transmitted signal measured at an array of spatially separated receivers. Several methods have been developed for estimating the EM source location. These methods include Time of Arrival (TOA), Time Difference of Arrival (TDOA), Frequency Difference of arrival (FDOA), Angle of Arrival (AOA) and Received Signal Strength (RSS) of the transmitted emitter signal. Comparing all localization techniques, TDOA and FDOA localization techniques (hyperbolic) are one of the simplest and most cost effective. The sensors situated on an axis in two-dimensional scenario measuring the TDOA and FDOA of the emitting signal from a moving source can estimate its position and velocity from the intersection point of hyperbola, which is created from TDOA and FDOA non-linear equations set. However, the hyperbolas may not be intersected at a single point due to the non-linear localization equations set and measurement noise in wireless sensor network (WSN). It is therefore important to estimate a source position that minimizes its deviations from the actual position. In this paper, a hybrid method combined with maximum likelihood (ML) and genetic algorithm (GA) are proposed to determine the instantaneous position of the moving source by estimating the position and velocity based on hyperbolic techniques (TDOA and FDOA). Firstly ML is applied in the position and velocity localization data. Additionally, GA is implemented to acquire the globally best solution of localization parameters from non-linear equations set of ML solution. The results obtained confirmed that the proposed method achieved the theoretical lower bound for near to far-field with same and different velocity and different baseline of sensors in low to high Gaussian noise level. In this study, explicit solutions are provided by the proposed methods that are not achievable through the established methods in all cases.