LAUSR

In MIMO-OFDM systems, knowledge of the precoding matrix enables effective resource allocation for high rates and low BER. However, feeding back precoders for all subcarriers introduces large overheads. Past work has shown that the orthonormal structure of the precoder enables effective parameterization and quantization using Givens rotations and Householder transformations. Also, the use of temporal and frequency correlations can further reduce the feedback requirements. We present a framework that utilizes adaptive feedback with joint time-frequency prediction for precoder construction at the transmitter. Simulations reveal that the proposed method achieves significantly lower BERs for various channel profiles compared to other approaches.