Deep Learning for Improving Performance of OOK Modulation Over FSO Turbulent Channels

Abstract

Free space optical (FSO) communication technology has become increasingly advanced with capabilities of high speed, high capacity, and low power consumption. However, despite the great potential of FSO, its performance is limited in a turbulent atmosphere. Atmospheric turbulence causes scintillation in the FSO propagated signals, leading to an increase in the bit error rate (BER) performance of the recovered signals at the receiver. In this paper, we demonstrate that the use of deep learning (DL) detection methods could overcome these limitations. We present a new detection method of on-off keying (OOK) modulated signals by using different models of DL over different strength FSO turbulent channels, without the need for prior knowledge of the parameters of the channel. The demonstrated DL decoders improve the performance of the FSO turbulent channel and decrease the power consumption. Moreover, the demonstrated DL models also work faster than maximum likelihood (ML) methods with perfect channel estimation decoders, with even slightly better performance because of the turbulence, thus enabling realization of FSO over turbulent atmospheric channels. INDEX TERMS Free space optical communication, deep learning, on-off keying modulation, amplitude shift keying modulation, maximum likelihood, channel state information, fully convolutional neural network, fully connected neural network, additive white gaussian noise, intensity modulation, direct detection, photodetector, bit error rate.

Low complexity deep learning algorithms for compensating atmospheric turbulence in the free space optical communication system

“…After combining the received signal, it is entered in the detector. Our detector is inspired from Ref [38] where a convolutional layer is used in DNN for the sake of dimensionality reduction and speedup. To distinguish it from DNN and CNN, we call it the dense CNN (DCNN).…”

Improved Performance in the Detection of ACO-OFDM Modulated Signals Using Deep Learning Modules

“…In recent years, many researchers have used DL for improving the performance of a system in many fields, including speech recognition, wireless communication, optical wireless communication, etc. In [23], we successfully used DL for improving the performance of OOK modulation over different FSO turbulent channels. The received data are a corrupted version of the data after passing through turbulence, and the output data are the original data bits that we want to detect.…”

“…The received data are a corrupted version of the data after passing through turbulence, and the output data are the original data bits that we want to detect. Here, we want to use the same concept as in [23], to improve the performance of modulated ACO-OFDM signals. We know that this modulation is similar to OOK but is more sophisticated.…”

“…In [22], DL was used as a solution for the imperfect CSI problem in FSO communication systems. Recently, in [23], we used DL for improving the performance of on-off keying (OOK) modulation over different FSO turbulence channels.…”

Free-space optical channel performance under atmospheric losses using orthogonal frequency division multiplexing

“…DL and artificial intelligence (AI) algorithms provide the most economical way to overcome this problem [17], [18], and these algorithms are applied in broad applications like self-driving cars, visual recognition, healthcare [19], [20]. Different deep learning models can be applied to different strengths of FSO turbulent channels to detect OOK modulated signals [21]. A machine-learning-based methodology was presented for improving future optical wireless communication systems from existing fiber-based networks to cognitive networks based on fiber-based learning that provides cognitive capabilities at the physical layer [22].…”