Clustered OFDM With Adaptive Antenna Arrays for Interference Suppression

Abstract

We investigate clustered orthogonal frequency-division multiplexing (OFDM) with adaptive antenna arrays for interference suppression. To calculate weights for interference suppression, instantaneous correlations of the received signals and channel responses corresponding to the desired signals have to be estimated. However, due to smaller size of each cluster for clustered OFDM than for classical OFDM, the discrete Fourier transform (DFT)-based estimator has large leakage and results in severe performance degradation. Therefore, a polynomial-based parameter estimator is proposed to combat the severe leakage of the DFT based estimator. We study the impact of the polynomial order and window size on the estimation error. An approximately optimal window size for the polynomial-based estimator is obtained and an adaptive algorithm for the optimal window size is developed. With the adaptive algorithm, the polynomial-based estimator has no leakage and does not require channel statistics. Simulation results show that the developed algorithm improves the performance significantly.