A Self-Localization Method Using Propagation of Existence Probability

In the location dependent information services, not only estimation of the coordinates of a user (coordinate estimation), but also existence check of a user whether he/she exists in a certain area or not (area estimation) are required. However, no existing research on location estimation method using wireless communication treats both coordinate estimation and area estimation simultaneously with high accuracy. Moreover, the estimation accuracy is low if a user exists at a area where less nodes of wireless communication has its coordinates. In this paper, we propose a novel self-localization method using propagation of geographical existence probability distribution based on RSSI. Proposed method can deal with coordinate and area estimation simultaneously. It also improves the accuracy of estimation by using propagation of existence probability distribution instead of propagating the coordinate estimation result. The simulation result shows that the proposed method can raise recall to 74% keeping precision as 97% for an area estimation.