A Wi-Fi Indoor Positioning Method Based on an Integration of EMDT and WKNN

• DOI: 10.3390/s22145411
• Publication Date: 2022-07-21T07:34:40Z
• Authors: Rong Zhou; Fengying Meng; Jing Zhou; Jing Teng
• Keywords: Signal strength; Smoothing

In indoor positioning, signal fluctuation is one of the main factors affecting positioning accuracy. To solve this problem, a new method based on an integration empirical mode decomposition threshold smoothing (EMDT) and improved weighted K nearest neighbor (WKNN), named EMDT-WKNN, proposed in paper. First, nonlinear non-stationary received strength indication (RSSI) sequences are constructed. Secondly, intrinsic functions (IMF) selection criteria energy analysis coefficients proposed. Thirdly, EMDT employed to smooth RSSI fluctuation. Finally, further avoid influence accuracy, deviated matching points removed, more precise combined weights constructed by combining geometric distance Euclidean fingerprints method-WKNN. The experimental results show that, underground parking dataset, accuracy EMDT-WKNN can reach 1.73 m 75th percentile error, which 27.6% better than 2.39 original method.