Hybrid Adaptive Precoder and Post-Distorter for Massive-MIMO VLC

Abstract

Performance of massive multi-input multi-output (MIMO) visible light communication (VLC) link is limited by ill-conditioned massive-MIMO channel, and nonlinearity of light emitting diodes (LEDs), which degrades the bit error rate (BER) performance. In this letter, to circumvent the above limitations, we propose a hybrid solution by using a singular value decomposition based adaptive precoder and a kernel minimum symbol error rate based post-distorter. Simulations indicate that the proposed algorithm outperforms the existing algorithms in terms of BER and computational complexity. Further, for the first time, the effect of DC-bias fluctuations on BER and dictionary-size is quantified analytically, and verified through numerical simulations. © 1997-2012 IEEE.