Indoor high-precision three-dimensional positioning algorithm based on visible light communication and fingerprinting using K-means and random forest

Recently, visible light communication (VLC) has been widely used in indoor positioning. Considering that conventional VLC-based positioning (VLP) algorithms are susceptible to interference and cannot adapt well to the complex indoor location environment, we propose a high-precision indoor three-dimensional positioning algorithm based on VLC and fingerprinting using fusion of K-means and random forest algorithms. Unlike the trilateration techniques based on received signal strength, the mentioned system does not depend on the model parameters and is robust to external interference. Besides, the proposed system introduces the concept of VLP kernel to compress the fingerprint database and shorten the training time. And we have validated the feasibility of our proposed VLC-based positioning system using fingerprinting in the real indoor VLP environment. Both simulation and experiment results verify that our proposed algorithm delivers satisfactory performance in terms of real-time ability and positioning accuracy, both of which are crucial for the performance of an indoor positioning system. Therefore, our proposed positioning system with strong interference immunity is a promising candidate for future indoor positioning applications.