Optimal correlative coding for discrete-time OFDM systems

Frequency-domain correlative coding (CC) has been widely considered to be effective in intercarrier interference (ICI) mitigation in orthogonal frequency-division multiplexing (OFDM) systems due to its potential in the considerable enhancement of the carrier-to-interference power ratio (CIR). Existing work has focused on searching for the optimal coding coefficients in minimizing the ICI power. However, the fundamental analysis is largely based on the continuous-time signal model (CTSM) and the assumption of an infinite number of subcarriers. In practice, the OFDM signal always involves a finite bandwidth and is sampled for digital processing purposes. Attentive to the given needs of practical systems, we derive the optimal coding coefficients in the maximization of the CIR or the minimization of the ICI power based on the discrete-time signal model (DTSM) with a finite number of subcarriers. Results show that the optimal coding coefficients for the DTSM differ from the existing coefficients developed for the CTSM and that our performance analysis leads to a much improved coding gain prediction for practical systems.