MUSIC Algortihm for RSSI-Based DoA Estimation on Standard IEEE 802.11/802.15.x Systems

Existing global positioning systems (GPS) applications are normally denied in indoor environments where, in spite of this limitation, a number of interesting applications are evolving into commercial products. At present, for indoor positioning, there are few cost-effective alternatives to GPS. One promising approach is based on the received signal strength indicator (RSSI) estimation, which is available in every IEEE standard compliant wireless transceiver, and includes useful information about the mutual positions between two position reference nodes. Extrapolating distance evaluation from RSSI is not reliable without the adoption of an accurate channel model [1],[2]. By using an anchor node/router with an array of directive antennas, it is possible to implement a spatial subdivision aimed at achieving an estimation of the direction of arrival (DoA) by evaluating the ratios between the single anchor’s antennas, and thus independently of the channel characteristics. This approach makes use of already existing spectral-based DoA localization algorithms ([3],[4]), but all of them assume a real-time analysis of both the amplitude and the phase of the received signals. In this article, we demonstrate analytically the capability of the MUSIC algorithm [3] to elaborate only the RSSI readings. We also compare the DoA localization performance using the classical information set (i.e., amplitude and phase) versus the use of the RSSI set (obtained simply by reading transceiver standard registers). In conclusion, we introduce some fundamental array design principles for optimal MUSIC RSSI implementation, with a real implementation tracking test.