<![CDATA[The indoor positioning method based on Wi-Fi fingerprinting has the advantages of simple acquisition and low cost. However, during signal collection, the presence of significant noise in the environment can cause fluctuations in signal strength measurements due to environmental variations. Additionally, a large number of fingerprints usually need to be collected to achieve high positioning accuracy. To address these issues, this paper proposes a positioning method based on a robust principal component analysis algorithm (RPCA) in a sparse fingerprint environment. Firstly, considering the outlier noise present in the collected signals, purification is performed based on signal measurement weights, and the refined fingerprints are stored in the fingerprint database. Secondly, given the high cost of collecting fingerprints, this paper generates some virtual fingerprints near reference points based on a transmission loss model, all of which are stored in an offline fingerprint database. Finally, adaptive K-value fingerprint matching is used to obtain the final results. The results show that the proposed algorithm can improve positioning accuracy in a sparse fingerprint environment.]]>
Copyrights © 2025
