Detection of neighboring access points based on relative locations in spring model for Cognitive Wireless Networks

This paper proposes a method to estimate locations of neighboring APs near a specified AP without using absolute location measured by the GPS. This estimation contributes to reduce handover latency between APs and improve communication quality. The method does not require terminals to have the GPS because it's not supported by all existing terminals. The estimation utilizes relative location of APs calculated using the spring model where RSSI and detection probability of APs are applied. Simulation results show that the nearest three APs can be estimated with probability of around 30%. The relative location is expected to be applied to enhance communication quality on mobile terminals.