Indoor Visible Light Positioning Using a Novel Fingerprint Regeneration Method Under Uneven Fingerprint Distribution Scenarios

Traditional fingerprint-based visible light positioning (VLP) methods require the collection of received signal strength (RSS) values from a large number of reference locations. Accordingly, various fingerprint regeneration methods were proposed to reduce the number of fingerprints. These methods generally assume a uniform distribution of fingerprints, which limits their applicability because achieving a uniform distribution of fingerprints is not easily attainable in practice. In this paper, we propose a novel fingerprint regeneration method that takes into account uneven fingerprint distribution scenarios. Unlike other existing regeneration methods, the proposed approach does not depend on the accurate signal propagation model, allowing fingerprints to be arbitrarily distributed and can generate regeneration points based on the locations of fingerprint points adaptively. Moreover, we transform the problem of fingerprint weight calculation into an optimization problem of objective function, and employ three intelligent optimization methods to solve it. Simulation results indicate that the proposed method can achieve higher positioning accuracy compared to three existing regeneration methods, and its accuracy is close to the performance of a dense fingerprint database.