Joint Spatial and Spectral Localization of OFDM Sources with Noncoherent Arrays

Localizing users in an orthogonal frequency-division multiple access (OFDMA) system without the knowledge of their subcarrier assignments poses a unique challenge. This problem arises in non-cooperative situations where the localizing algorithm does not have access to the OFDMA multiple access layer. In this paper we propose a clustering-based algorithm for joint spatial and spectral localization of OFDMA users with noncoherent arrays. The proposed method consists of three step. First, each array estimates the direction-of-arrival (DOA) of a potential source in every subcarrier. Next, the potential source's position is estimated via multiangulation of these per-subcarrier DOAs. Finally a clustering step identifies the source occupying each subcarrier and combines subcarrier-wise position estimates to form the aggregate estimate. Simulation results are included to illustrate the performance of the proposed algorithm.