EM-Based Localization of Noncooperative Multicarrier Communication Sources with Noncoherent Subarrays
Multicarrier communication signals possess distinct characteristics different from narrowband signals often targeted by localization algorithms. Each source occupies a unique set of subcarriers and is practically absent from other subcarriers. Moreover, subcarrier allocation for a source may change over time. In noncooperative geolocation applications, knowledge of the set of subcarriers occupied by a source may not be available. This paper proposes iterative localization algorithms which jointly estimate the locations of the sources in both the spatial and spectral domains. It is assumed that the transmitted signals are intercepted by spatially dispersed subarrays which are asynchronous. This eliminates the need for accurate clock synchronization across the geographically distributed subarrays. The proposed method is based on the expectation-maximization (EM) algorithm, and the subcarrier occupancy detection is achieved via the likelihood ratio test within each the EM iteration. The performance of the algorithm is evaluated through simulation for a multi-source localization problem. It is shown that for signal-to-noise ratios of interest, the proposed method achieves a performance close to the algorithm, which is aware of all the subcarrier allocations.
Publication Source (Journal or Book title)
IEEE Transactions on Wireless Communications
Naraghi-Pour, M., & Ikuma, T. (2018). EM-Based Localization of Noncooperative Multicarrier Communication Sources with Noncoherent Subarrays. IEEE Transactions on Wireless Communications, 17 (9), 6149-6159. https://doi.org/10.1109/TWC.2018.2855126