Attack-resistant wireless localization using an inclusive disjunction model

Wireless positioning is vulnerable to malicious attacks due to the nature of its open medium. This study proposes an attack-resistant fingerprinting localization algorithm based on a probabilistic inclusive disjunction model. This model allows an attacked observation to play a less significant role during the localization process, thus achieving more robust location estimations under security threats. This study included experiments conducted in an actual Wi-Fi network. Experimental results demonstrated that this approach apparently achieves more robust estimation than cluster-based, median-based, and sensor-selection methods under various attacks on RSS.