Fast Optimal Discrete Bit-Loading Algorithms for OFDM-Based Systems

Adaptive bit allocation for orthogonal frequency division multiplexing (OFDM)-based systems are critical for avoiding performance degradation due to additional bit allocations when channel conditions are bad. In this paper, we present two novel bit-loading algorithms for OFDM-based systems with target bit rate and fixed energy constraints. The proposed algorithms converge to the same bit allocation as the optimal discrete bit-filling and bit-removal algorithms. The proposed algorithms can approach a near-optimal bit allocation with less computational steps by using a water-filling solution at the initial bit allocation step, and then a multiple-bit loading procedure to satisfy the target bit rate. The proposed algorithms were evaluated by comparing with bit-filling and bit-removal algorithms through numerical analysis. Numerical results demonstrate the computational efficiency, fast convergence and optimality of the proposed algorithms.