Anchor-Free Indoor Localization Based on Floor Plan

Abstract

Anchor-free indoor localization technique is very attractive in disaster relief applications because it does not need any anchors. However, the existing anchor-free localization methods have limited performance since they do not consider positioning error from indoor walls and high mobility of nodes. In this paper, we propose an RSSI-based indoor anchor-free localization scheme, which utilizes pre-surveyed data on target area to reduce localization error caused by the walls. First, we propose a heuristic localization scheme based on weighted subgraph matching algorithm, which matches the reported RSSI values from nodes with the pre-surveyed data. Then, we determine expected region of nodes with the reported speed values from nodes so as to realize localization for a node with high mobility. Furthermore, we utilize the recursive location filtering method that estimated locations outside the expected region are filtered out, which ensures robust and fast localization. The simulation results show that, compared with other techniques not utilizing anchors, the proposed scheme significantly improves the localization accuracy and reduces the time to find locations of all nodes.