Abstract—In this paper, we perform a comparative study of both the achievable performance and the associated computational complexity of two major time-domain prediction strategies proposed for employment in wireless mobile communication systems. Specifically, we investigate the intrinsic design tradeoffs of the so-called stationary robust predictor and the adaptive Recursive-Least-Squares (RLS) predictor. We demonstrate that the RLS predictor outperforms its robust counterpart at a cost of slightly higher computational complexity, and hence, the RLS predictor constitutes a better alternative for employment in wireless transceivers. Index Terms—Channel prediction of time-varying channels, minimum mean square error (MMSE) channel prediction, orthogonal-frequencydivision-multiplexing (OFDM), recursive least squares.
展开▼
