Adaptive beamforming based on frequency-to-time pilot transform for OFDM

In this paper, we propose beamforming schemes for the combination of orthogonal frequency division multiplexing (OFDM) and adaptive antenna array. The least mean square (LMS) and sample matrix inversion (SMI) algorithms are considered in the adaptive beamforming. These combinational schemes are characterized as the frequency-to-time pilot transform and pre-FFT combining. In every OFDM block containing both data and pilot symbols, we transform the frequency-domain pilot symbols into the corresponding time-domain components. The simulation results show that the beamforming scheme based on the frequency-to-time pilot transform is very effective in improving the estimation accuracy of antenna weights for LMS beamforming, and very effective in improving both the estimation accuracy and convergence behavior for SMI beamforming.