Sensor Network Localization via Received Signal Strength Measurements with Directional Antennas

In wireless sensor networks position awareness is necessary to exploit the communication benefits of directional antennas and for sensors to provide meaningful information about their surroundings. In this paper we evaluate the feasibility and quality of self-localization that can be obtained using received signal strength (RSS) measurements from arrays of directional antennas on each sensor node. We compare the performance of an optimal estimator utilizing all RSS data with a suboptimal one using angles of arrival as an intermediate statistic. Further, we compare the performance bounds of our approach with others based on distance measurements potentially obtained through time of arrival or RSS observations. We demonstrate that sub-meter location accuracy is possible using 802.11 radio frequency communication signals and no assumed model for propagation loss. Results from an outdoor field experiment and our proposed printed circuit board quasi-Yagi directional antenna are also presented.