An Indoor Positioning Algorithm Based on Particle Filter and Neighbor-Guided Particle Optimization for Wireless Sensor Networks

In order to achieve accurate and robust indoor position estimation in a wireless sensor network using particle filtering (PF)-based positioning algorithms, particle impoverishment, a problem that aroused from the traditional replicating-and-replacing resampling operation, must be well addressed. The loss of particle diversity not only degrades positioning accuracy but can also result in filtering divergence. To address this problem, this article proposes an adaptive neighbor-guided particle optimization strategy to substitute the traditional resampling operation. The strategy optimizes the distribution of the posterior particles through three steps: neighbor radius calculation, true neighbor identification, and neighbor-guided attraction. The proposed strategy is then integrated into the PF framework to form a novel positioning algorithm referred to as adaptive neighbor-guided particle optimization-based PF algorithm (ANPOPF). The test results show that the integration of the proposed particle optimization strategy considerably enhances the robustness of the PF algorithm, mitigating the effects of particle impoverishment. With the aid of the strategy, the ANPOPF algorithm achieves higher positioning accuracy compared to several existing positioning algorithms. Moreover, the ANPOPF algorithm owns an affordable computation load for most real-time applications.