The rapid development of emerging technologies has led to a significant increase in data traffic, posing challenges for data centers in terms of storing, transmitting, and processing large volumes of information. Intra- and inter-datacenter connections with high bandwidth, low latency, high reliability, and high energy efficiency are in great demand. Traditional electrical communication is becoming increasingly inadequate, while optical communication has emerged as promising solutions. Visible light laser communication (VLLC) utilizes the spectrum ranging from 380nm to 700nm and has the advantages of offering high transmission rates, cost-effectiveness, and simple system structures. In this paper, enabling technologies including high-bandwidth laser diodes, advanced modulation formats, and artificial intelligence (AI)-based equalization schemes are introduced. A compact 50-channel VLLC interconnect system has been developed, achieving ultra-high-speed short-range VLLC transmission with a record data rate of 534.51Gbps, showcasing the potential of VLLC as a high-capacity and cost-effective optical intra-data center connection solution.

Underwater visible light communication (UVLC) based on light-emitting diodes (LEDs) is considered a potential candidate for underwater wireless data transmission. However, the implementation of high-speed UVLC remains a challenge due to bandwidth limitation and nonlinear effects. In this paper, we investigate the performance of Tomlinson-Harashima precoding (THP) in the bandwidth-limited UVLC system for the first time. Apart from research on linear and Volterra series-based nonlinear THP, neural network (NN)-based THP is first proposed and conducted. At the transmitter, the intersymbol interference (ISI) and nonlinearity can be partially mitigated through a feedback neural network (FBN) without error propagation. A modulo operator is employed to eliminate instability. In addition, for the precoded signals, the power spectral density (PSD) is near-white. Therefore, it is possible to avoid the spectrum-shaping-induced signal-to-noise ratio (SNR) loss problem that is common in pre-emphasis methods. At the receiver, another modulo operator is used for decoding. An adaptive feedforward neural network (FFN) is applied to compensate for the channel state information (CSI) mismatch of the FBN and mitigate the remaining ISI and nonlinear impairments. In the demonstrated UVLC system, the experimental results prove the feasibility of the proposed method. The Q factor is increased by 3.39 dB at a data rate of 2.2 Gbps. 630 MBd CAP-16 transmission under a 7% HD-FEC threshold is realized, which is 90 MBd higher than that when employing linear postequalization only.

In this paper, we propose a novel nonlinear resilient learning post equalizer named TFDNet in UVLC system. Unlike the traditional deep neural network (DNN) based post equalizers which merely consider the time domain, the proposed TFDNet exploits time-frequency image analysis which considers the time and frequency domains simultaneously and transforms the signal into 2D time-frequency image, which is further learned by neural network. Experimental results demonstrate that TFDNet outperforms Volterra and DNN based methods for compensating nonlinear distortions through a 1.2 m underwater channel using 64 quadrature amplitude modulation-carrierless amplitude modulation (64QAM-CAP). Even under severe nonlinear distortions where Volterra and DNN cannot work, TFDNet retains valid bit error rate (BER) below the 7% forward error correction (FEC) limit of 3.8 × 10 ^-3 . The performance of TFDNet verifies the effectiveness of time-frequency image analysis which has been applied to tackle nonlinear distortions in UVLC system for the first time.

Application of light-emitting diodes (LEDs) based visible light communication (VLC) in underwater wireless communication has become a hot topic. However, on the one hand, the performance of underwater VLC (UVLC) system is limited due to power attenuation in water medium. On the other hand, increasing the driving power will induce nonlinearity from the imperfect optoelectronic components. Particularly at high signal amplitude, nonlinearity will become the dominant bottleneck. In this paper, we focus on nonlinearity mitigation method which preprocesses signal at transmitter-side. Multi-bit delta-sigma modulation combined carrier-less amplitude and phase modulation (DSM-CAP) is proposed to obtain quantized signal with lower peak-to-average power ratio (PAPR). Furthermore, as assuming the quantization noise being additive white noise is not appropriate in nonlinearity, modulus pruned weighted look-up table (MP-WLUT) is employed to modify the signal before DSM quantization to mitigate pattern dependent nonlinear distortion. Experimental results indicate that the DSM-CAP has better nonlinearity mitigating performance than normal CAP. At the data rate of about 1.69Gbp/s, utilizing MP-WLUT pre-distorted DSM-CAP an improvement of 1.44 dB Q-factor is realized compared with normal CAP without WLUT, and the size of MP-WLUT can be reduced to 62.5% of the full-size WLUT.

Lookup table (LUT) is an efficient digital predistortion technique for mitigating the pattern dependent distortion in the field of underwater wireless optical communication (UWOC) system. However, the complexity of LUT grows geometrically when applied with high-order QAM modulation formats with multi-pattern length sequences. The complexity reduction of LUT is essential for practical applications. In this paper, we study the patterning effects of LUT predistortion and propose to reduce the LUT complexity by pruning its table entries aided with multi-modulus weighted strategy. The size of LUT has been reduced significantly in experiments, which shows that the complexity of the proposed MMW-RLUT only accounts for 17.6%, 9.8%, 4.9% and 14% of full-size LUT for 64QAM, normal 32QAM, square geometrical shaping (SGS) 32QAM and 16QAM modulated UWOC system, without degradation of performance.