Multiuser Massive MIMO-OFDM for Visible Light Communication Systems

Abstract

Visible light communication (VLC) has emerged as a viable supplement to existing radio frequency (RF) communication systems. The limited modulation bandwidth of sources is one of the significant challenges in the multiuser (MU) VLC systems, which prohibits transmission at high data rates for each user. Since an array of LEDs is used to illuminate a room, a MU VLC system utilizing massive multiple-input multiple-output (mMIMO) orthogonal frequency-division multiplexing (OFDM) is proposed for achieving high data rates in this paper. Since the distances of the multiple transmitter-receiver links are different, their temporal delays are also different, resulting in complex channel gain and phase differences when transformed to the frequency domain in the OFDM technique. Complex channel vectors associated with different users are mutually spatially orthogonal as the number of transmitters increases. Therefore, inter-user interference (IUI) can be eliminated with simple linear signal processing, and more users can simultaneously communicate in the same time-frequency resource. In this paper, three linear precoding methods, including the maximum ratio transmission (MRT), the minimum mean square error (MMSE), and the zero-forcing (ZF), are investigated for VLC systems based on proposed MU-mMIMO-OFDM in intensity-modulation and direct-detection link. We evaluate and compare the performance of the aforementioned precoding methods in terms of the achievable spectral efficiency and total downlink optical power for different scenarios. Moreover, we derive a closed-form expression for the lower bound on the average achievable sum-rate, which we confirm the accuracy of the derived expression with the numerical simulation. The results demonstrate that the performance of the proposed method in this paper for all precoder techniques, i.e., ZF, MMSE, and MRT, is better than the previous works since the complex channel in the frequency domain is used in this paper. Furthermore, the ZF and MMSE methods are better than MRT when the ratio of the number of LEDs to the number of users is large. At the same time, when this ratio is relatively small, the MRT precoding technique outperforms the ZF and MMSE techniques.