LSTM-Autoencoder Deep Learning Technique for PAPR Reduction in Visible Light Communication

Abstract

Visible light communication (VLC) is a relatively new wireless communication technology that allows for high data rate transfer. Because of its capability to enable high-speed transmission and eliminate inter-symbol interference, orthogonal frequency division multiplexing (OFDM) is widely employed in VLC. Peak to average power ratio (PAPR) is an issue that impacts the effectiveness of OFDM systems, particularly in VLC systems, because the signal is distorted by the nonlinearity of light-emitting diodes (LEDs). The proposed method Long Short Term Memory-Autoencoder (LSTM-AE) uses an autoencoder as well as an LSTM to learn a compact representation of an input, allowing the model to handle variable length input sequences as well as predict or produce variable length output sequences. This study compares the suggested model with various PAPR reduction strategies to demonstrate that it offers a superior improvement in PAPR reduction of the transmitted signal while maintaining BER. Also, this model provides a flexible compromisation between PAPR and BER.

A Survey of PAPR Techniques Based on Machine Learning

“…Motivated by the above cited advantages, several works in the literature explored different ML techniques [ 134 , 135 , 141 , 142 , 143 , 144 , 145 , 146 , 147 , 148 , 149 , 150 , 151 , 152 , 153 , 154 , 155 ] to reduce the PAPR. Therefore, to improve the understanding of readers not particularly familiar with ML, we next present a brief explanation of the ML techniques considered by the surveyed papers.…”

“…As previously mentioned, ML is a key to solving the PAPR issue in OFDM systems. Therefore, several works in the literature have studied the application of different ML techniques to reduce the PAPR [ 134 , 135 , 141 , 142 , 143 , 144 , 145 , 146 , 147 , 148 , 149 , 150 , 151 , 152 , 153 , 154 , 155 ]. Therefore, as the main goal of this paper is to show the role of ML in PAPR reduction, we next present the reviewed papers, which are organized following the considered ML area, and describe their main contributions, features, and limitations.…”

“…After understanding the basic operation of an ANN-based PAPR reduction solution, we can discuss in more detail the main approaches in the literature [ 135 , 141 , 142 , 143 , 144 , 145 , 146 ]. More specifically, in [ 141 ], an innovative approach is proposed to mitigate the PAPR in a Visible Light Communication (VLC) Direct Current-biased Optical OFDM (DCO-OFDM) system, employing a deep autoencoder.…”

“…It highlights the capability of this method to accurately predict optimal PAPR targets. Moreover, the authors of [ 146 ] emphasized the LSTM-Autoencoder (LSTM-AE) model for its ability to handle variable input sequential data and its flexibility in finding a compromise between the PAPR and BER through hyperparameter tuning. This paper aims to tackle the PAPR issue in OFDM systems specifically within VLC.…”

Emerging Technologies for 6G Communication Networks: Machine Learning Approaches

“…Throughput improvement [92] Spectral efficiency [129] Computational reduction [130] Reliable communication [132,138] Energy efficient communication [138] RNN Multi-layer feed-forward NNs Trained using the backpropagation method which considers input, weights, and memory for each output layer Energy efficient communication [103] Reliable communication [112,116,120] Throughput improvement [120,141] RL ML algorithms that allow machines to continuously learn from their experience data sets to automatically make the most accurate decisions…”