Data-aided distorted constellation estimation and demodulation for 60 GHz mmWave WLAN

Abstract

The nonlinear distortion caused by power amplifier (PA) nonlinearity distorts the constellation and degrades the bit error rate (BER) performance of 60 GHz millimeter wave (mmWave) wireless local area network (WLAN) easily, due to the huge bandwidth and high equivalent isotropic radiated power (EIRP), especially for quadrature amplitude modulation (QAM) signals. This paper aims at data-aided (DA) distorted constellation (DC) estimation and demodulation to resist PA nonlinearity. A maximum likelihood (ML) estimation algorithm is proposed, and the corresponding Cramer-Rao lower bounds (CRLBs) are computed. Furthermore, we implement the demodulation based on the DC rather than the standard constellation (SC). Simulation results indicate that the means of the estimated parameters are unbiased and the mean square errors (MSEs) approach the CRLBs. Meanwhile, the proposed DC demodulation with PA nonlinearity achieves BER performance close to the ideal PA case, and evidently outperforms the traditional SC demodulation especially in the high-code-rate or uncoded case.