Range filtration algorithm for wireless sensor networks

摘要

The combination of small scale, low cost, and sensing ability has made wireless sensor network (WSN) a best solution for low complexity indoor localization. The localization precision is highly dependent on the ranging accuracy. The inaccuracy caused by various error sources results in range estimation error (REE). In this paper a range filtration algorithm (RFA) based on the AndersonDarling (AD) goodness-of-fit (GOF) test is proposed. The RFA exploits the distribution of REE to obtain the range estimates with least error. The performance of RFA is assessed by applying it to real time ranging data obtained from experiments using IEEE 802.15.4 compliant devices, covering indoor/outdoor environments with both line-of-sight (LOS) and non-line-of-sight (NLOS) conditions. The outcome demonstrates a significant improvement in the mitigation of the REE using RFA.