Efficient cooperative localization method with node selection based on position error bound

Abstract

Cooperative Localization (CL) has become a promising solution that circumvents the needs for high-density anchor deployment, and offers additional localization accuracy in Wireless Sensor Networks (WSNs). However, it suffers from high computational complexity and increased network traffic problem. To alleviate this problem, we propose an iterative greedy node selection approach based on Position Error Bound (PEB) for Angle-of-arrival (AOA) based CL. We introduce a Boolean vector to formulate the node selection optimization problem that maximizes the Decrease of PEB (DPEB) from agents’ cooperation. Since the problem is NP-hard and difficult to solve, the iterative greedy algorithm is proposed to approach the sub-optimal solutions. Then, in order to improve the accuracy of localization phase, we derive the efficient closed-form solution of AOA CL problem and propose an iterative constrained Least Squares (LS) method. Moreover, the theoretical analysis shows that the localization estimator is very close to the Cramer Rao Lower Bound (CRLB) performance. Finally, simulation results show that the proposed node selection approach can greatly decrease the number of agents participating in the CL without significant degradation in localization accuracy. Besides, the proposed CL method outperforms the existing ones in terms of localization accuracy.