Memory-controlled deep LSTM neural network post-equalizer used in high-speed PAM VLC system

Abstract

Linear and nonlinear impairments severely limit the transmission performance of high-speed visible light communication systems. Neural network-based equalizers have been applied to optical communication systems, which enables significantly improved system performance, such as transmission data rate and distance. In this paper, a memory-controlled deep long short-term memory (LSTM) neural network post-equalizer is proposed to mitigate both linear and nonlinear impairments in pulse amplitude modulation (PAM) based visible light communication (VLC) systems. Both 1.15-Gbps PAM4 and 0.9Gbps PAM8 VLC systems are successfully demonstrated, based on a single red-LED with bit error ratio (BER) below the hard decision forward error correction (HD-FEC) limit of 3.8 x 10 −3 . Compared with the traditional finite impulse response (FIR) based equalizer, the Q factor performance is improved by 1.2dB and the transmission distance is increased by one-third in the same experimental hardware setups. Compared with traditional nonlinear hybrid Volterra equalizers, the significant complexity and system performance advantages of using a LSTMbased equalizer is demonstrated. To the best of our knowledge, this is the first demonstration of using deep LSTM in VLC systems. IntroductionVisible light communications (VLC) based on light emitting diodes (LEDs) has become an attractive and promising technology due to its cost effectiveness, immunity to electromagnetic interference, license-free and high security [1]. In recent years, transmission rate of visible light systems has been increasing with the use of` high-order modulation, such as orthogonal frequency division multiplexing (OFDM) [2], carrier-less amplitude and phase modulation (CAP) [3] and pulse amplitude modulation (PAM) [4].Equalizer is one of the most critical parts of the VLC systems. As a typical communication system, the transmitted signal of VLC systems is distorted in amplitude and delay causing the inevitable inter-symbol interference (ISI). On the other hand, nonlinear distortion has gradually become a new bottleneck in high-speed transmission systems due to nonlinear V-I response of LED source [5] and other origins that have not been well-modeled such as non-linear distortion arising from the transmitter driving circuits and the electrical amplifier (EA). Recently, Adaptive finite impulse response (FIR) filter based linear equalizers in VLC systems have been widely studied, such as scalar modified constant multi-modulus algorithm (S-MCMMA) blind equalization algorithm [6], data-aided recursive least square (RLS) [7] and least mean squares (LMS) [8]. In order to compensate for both linear and nonlinear effects, hybrid equalization schemes have been shown to be effective approaches to improve the performance of VLC systems, such as adaptive FIR linear equalizer with

Space Division Multiple Access With Distributed User Grouping for Multi-User MIMO-VLC Systems

“…Note that it is reasonable to assume a 20-MHz modulation bandwidth in the simulations, since various FDE schemes can be applied to extend the limited 3-dB modulation bandwidth of COTS white LEDs [11]- [13]. Moreover, by regulating the input signal within the dynamic range of the LED and applying various LED nonlinearity compensation techniques such as pre-or post-distortion and post-equalization [40]- [44], it is also reasonable to neglect the nonlinear effects of the LED in the simulations.…”

Data Rate Enhancement in Optical Camera Communications Using an Artificial Neural Network Equaliser

“…Note, the boundaries have a high dependency on the number of neurons and the hidden layers, which are analogous to the human brain, where the synaptic weight is changed based on the training sequence. Different ANN approaches can be deployed for equalisation including the single-layer [33] and the multi-layer perceptron (MLP) [31]. In [26] it was shown that, MLP offers superior performance in mitigating ISI in optical wireless systems and hence, has been adopted in this work.…”

Convolutional neural network-based signal demodulation method for NOMA-PON

“…In recent years, machine learning especially neural network (NN) has become very popular and been introduced in optical communications for monitoring the optical signal to noise ratio (OSNR) [14], modulation format recognition [14], nonlinearity mitigation [15] and CE [16][17], etc. In [16], the NN-based equalizer was used to mitigate both linear and nonlinear distortion in a 100 Gb/s PON system.…”

“…In [16], the NN-based equalizer was used to mitigate both linear and nonlinear distortion in a 100 Gb/s PON system. In [17], a memory controlled deep long short-term memory (LSTM) NN-based post-equalizer was proposed to mitigate the transmission impairment in pulse amplitude modulation VLC systems. In [18], a LSTM network was proposed to detect the channel characteristics automatically for a typical NOMA system.…”