A sequential detection approach to indoor positioning using RSS-based fingerprinting

Using received signal strength (RSS) fingerprints from nearby Wi-Fi access points is a common approach to indoor localization in modern applications, in part due to its relatively low hardware cost and computational complexity. We present a sequential detection approach to RSS-based positioning that employs a Bayesian metric to identify the most likely path, given a time series of RSS measurements, traveled by the indoor user. Offline RSS training data is collected, and the k-nearest neighbors (KNN) algorithm is applied to approximate position likelihoods conditioned on RSS observations. The performance of the proposed approach is evaluated using data collected in an academic building at varying times of day and across several days. Results show that the sequential detection approach achieves strong performance even when only a small series of RSS measurements are available.