Interpolation Method for Radio Map Establishment Based on RSS Clustering and Propagation Model Optimization

In recent years, Location-Based Services (LBS) have been widely applied in people's life with various localization technologies. Because outdoor localization methods are not suitable for indoor environments, various indoor localization methods have been developed. Among the existing indoor localization methods, Wi-Fi fingerprinting localization has attracted great concerns because of its wide applicability, simple deployment, and comparable performance. This paper proposed an interpolation method for radio map establishment based on RSS clustering and propagation model optimization. Fuzzy C-Means (FCM) clustering algorithm is used to cluster the Received Signal Strength (RSS) samples collected at Reference Points (RPs). In each cluster, propagation model parameters are optimized. Then RSS samples are estimated at select locations for radio map establishment. With the radio map after interpolation, more accurate localization results can be computed using K Nearest Neighbors (KNN) fingerprinting algorithm.