Indoor Fingerprint Positioning Method with Standard Particle Swarm Optimization

This paper introduces the standard particle swarm optimization (SPSO) algorithm to fingerprint positioning. To reduce the computation amount, the estimated position from the weighted K-nearest neighbor (WKNN) algorithm is regarded as the initial point of the SPSO algorithm. The search area of SPSO can adapt to dynamic adjustment according to the locations of the nearest reference points (RPs). The fingerprint for each particle is generated by the inverse distance weighted (IDW) algorithm. SPSO aims to find a point where the fingerprint is closest to the online received signal strength (RSS) measurements, which is the global best position in the entire particle swarm. The experimental results show that the proposed method has a far better positioning effect than WKNN and Horus, with the improvement of positioning accuracy by 40.5% and 33.41%.