Performance of Star 16/64QAM Schemes Using Turbo FDE with Iterative Decision-Directed Channel Estimation for DFT-Precoded OFDMA

This paper presents the average block error rate (BLER) performance of star 16/64QAM schemes using iterative decision-directed channel estimation (IDDCE) associated with the turbo frequency domain equalizer (FDE) for discrete Fourier transform (DFT)-precoded Orthogonal Frequency Division Multiple Access (OFDMA). We show that the turbo FDE with the IDDCE based on the a posteriori log-likelihood ratio (LLR) decreases the required average received signal-to-noise power ratio (SNR) compared to that based on the extrinsic LLR. We also show that the turbo FDE is effective in decreasing the required average received SNR considering the cubic metric (CM) compared to the linear minimum mean-square error based FF-FDE for star 16/64QAM schemes. Moreover, we show that the (8, 8) star 16QAM and (16, 16, 16, 16) star 64QAM schemes decrease the required average received SNR considering the CM at the average BLER of 10 by approximately 0.8 and 0.3 dB compared to the square 16QAM and 64QAM schemes, respectively, with a low turbo coding rate such as R = 1/3 when using the turbo FDE associated with IDDCE. 