Cooperative localization of wireless sensor networks in non-ideal conditions

Abstract

The localization problem plays a vital role in the services provided by wireless sensor network (WSN). Under non-ideal conditions, the transmit power is different and the clock is not synchronized. The measured received signal strength (RSS) and time of arrival (TOA) are subjected to deviation. In this paper, a novel algorithm was proposed to alleviate these impact on localization accuracy and reduce computing consumption of localization. The algorithm called generalized trust region subproblem with improved weights (GTRS-IW) transforms the original non-convex problem into a framework of generalized trust region subproblem and solves it through a bisection process. Additionally, a critical distance was introduced to improve the weights. The proposed GTRS-IW achieves precise localization accuracy with lower computation time and was verified through simulations.