An Indoor Probabilistic Localization Method Using Prior Information

Abstract

In this paper, we propose a new probabilistic method for determining the position of an unknown node in an indoor environment. Our analysis shows that using a small subset of sensors reduces the error in comparison to larger sets. The best subset of sensors is determined by matching the power received by all of the sensors and comparing it to prior measurements. We present experimental measurements made that show the efficacy of this approach and compare this method to previously published techniques. Our analysis shows that the new method, Prior Measurement Comparison (PMC), yields greater estimation accuracy resulting in lower error.