Differential signal strength fingerprinting revisited

The provision of reliable location estimates in WiFi fingerprinting localization is challenging, mainly because users typically carry heterogeneous devices that report Received Signal Strength (RSS) measurements from surrounding Access Points (AP) very differently. This may render the user-carried device incompatible with the fingerprinting system, in case the RSS radiomap was collected with a different device, thus incurring high localization errors. To this end, we introduce a novel differential fingerprinting method that computes the difference between the RSS value of each AP and the mean RSS value across all APs in the original fingerprint. We show that the new fingerprints are robust to device heterogeneity, as opposed to traditional RSS fingerprints. In addition, we derive analytical results and demonstrate with simulations and experimental data that the proposed approach performs considerably better than existing differential fingerprinting solutions, in terms of localization accuracy and computational overhead.