Wireless Emitter Location Estimation Based on Linear and Nonlinear Algorithms using TDOA Technique

Abstract

Low-power devices such as cell phones, and wireless routers are commonly used to control Improvised Explosive Devices (IEDs) and as the communication nodes for the sake of command and control. Quickly locating the source of these signals is ambitious, specifically in a metropolitan environment where buildings and towers may cause intervention. This presents a geolocation system that compounds the attributes of several proven geolocation and error mitigation methods to locate an emitter of interest in an urban environment. The proposed geolocation system uses a Time Difference of Arrival (TDOA) approach to estimate the position of the emitter of interest. Using multiple sensors at known locations, TDOA estimates are achieved by the cross-correlation of the signal received at all the sensors. A Weighted Least Squares (WLS) solution, Linear least Square (LLS) method and maximum likelihood (ML) estimation is used to estimate the emitter's location. If the variance of this location estimate is too high, a sensor is detected and identified as possessing a Non-Line of Sight (NLOS) path from the emitter. This poorly located sensor is then removed from the geolocation system and a new position estimate is computed with the remaining sensor TDOA information. The performance of the TDOA system is determined through modeling and simulations. Test results confirm the feasibility of identifying a NLOS sensor, thereby improving the geolocation system's accurateness in a metropolitan environment.