in Proc. of Military Communications Conference, pp. 1-6, Atlantic City, NJ, Oct. 2005.
In this paper, we propose a receiver architecture that can demodulate multiple data streams with different frequency offsets and propagation delays in multiple input multiple output orthogonal frequency division multiplexing (MIMO-OFDM) ad hoc networks. The receiver performs joint synchronization and interference cancellation to separate these interfering data streams and achieve synchronization for each stream. We propose a novel training-based minimizing mean square error (MMSE) algorithm to jointly obtain the frequency offsets and the equalization delays. The effects of frequency offsets and propagation delays are compensated with these estimated parameters. We also propose a direct training-based equalizer design algorithm to obtain the equalizer coefficients. This algorithm involves solving an optimization problem that considers symbol error rate. Simulation shows good uncoded bit error rate performance for our proposed algorithms.
This paper is available in