RSS Estimation Based on Bayesian Learning Mechanism by Vehicular Sensor Networks

Abstract

Received Signal Strength (RSS) estimation for networks in urban transportation systems can be carried out by Vehicular Sensor Networks (VSN). All moving vehicles on roads with signal-sensing applications can act as mobile sensors, collecting RSS along their driving routes and uploading data to data center at the end of every drive. However, there must be inconsistencies among RSS values achieved at same locations owing to different types and brands for vehicles. In this paper, we propose a RSS estimation algorithm to improve the credibility of RSS information at certain locations. This algorithm is based on Bayesian learning mechanism and calibrated by the values gained by high-precision equipment PXI experimental platform. We evaluate the algorithm based on the real-world data collected by our Android application. The results demonstrate the effectiveness and superiority of the method compared with typical algorithms.