Efficient Modulation Classification Based on Complementary Folding Algorithm in UVLC System

In this letter, we jointly consider the problems of blind channel estimation (BCE), channel equalization and data detection in underwater visible light communication (UVLC) systems. Specifically, we propose to solve them with a model-driven deep learning (DL) approach. Inspired by the module-based signal processing methodology and communication theory, we devise a new blind receiver architecture called blind detection network (BDNet), which consists of three modules corresponding to BCE, channel equalization, and symbol demapping, respectively. Different from existing BCE schemes, the BDNet does not directly estimate the channel itself, but instead estimates the inverse channel regardless of the scalar ambiguity issue by learning the latent channel features from the received signal only. The UVLC channel effects can then be partially compensated by the zero-forcing equalization module of BDNet. Finally, the symbol demapping module recovers the transmitted symbols with the aid of the learned decision thresholds. Simulation results show the competitive bit error rate performance of BDNet in comparison to conventional pilot-aided schemes.

In this paper, we propose and evaluate two spatial division transmission (SDT) schemes, including spatial division diversity (SDD) and spatial division multiplexing (SDM), for underwater visible light communication (UVLC) systems. Moreover, three pairwise coding (PWC) schemes, including two one-dimensional PWC (1D-PWC) schemes, i.e., subcarrier PWC (SC-PWC) and spatial channel PWC (SCH-PWC), and one two-dimensional PWC (2D-PWC) scheme are further applied for signal-to-noise ratio (SNR) imbalance mitigation in the UVLC systems using SDD and SDM with orthogonal frequency division multiplexing (OFDM) modulation. The feasibility and superiority of applying SDD and SDM with various PWC schemes in a practical bandlimited two-channel OFDM-based UVLC system have been verified through both numerical simulations and hardware experiments. The obtained results show that the performance of SDD and SDM schemes are largely determined by both the overall SNR imbalance and the system spectral efficiency. Moreover, the experimental results demonstrate the robustness of SDM with 2D-PWC against bubble turbulence. Specifically, SDM with 2D-PWC can obtain bit error rates (BERs) under the 7% forward error correction (FEC) coding limit of 3.8 × 10-3 with a probability higher than 96% for a signal bandwidth of 70 MHz and a spectral efficiency of 8 bits/s/Hz, achieving an overall data rate of 560 Mbits/s.

Underwater visible light communication (UVLC) has recently come to light as a viable wireless carrier for signal transmission in risky, uncharted, and delicate aquatic environments like seas. Despite the potential of UVLC as a green, clean, and safe alternative to conventional communication methods, it is challenged by significant signal attenuation and turbulent channel conditions compared to long-distance terrestrial communication. To address linear and nonlinear impairments in UVLC systems, this paper presents an adaptive fuzzy logic deep-learning equalizer (AFL-DLE) for 64 Quadrature Amplitude Modulation-Component minimal Amplitude Phase shift (QAM-CAP)-modulated UVLC systems. The proposed AFL-DLE is dependent on complex-valued neural networks and constellation partitioning schemes and utilizes the Enhanced Chaotic Sparrow Search Optimization Algorithm (ECSSOA) to improve overall system performance. Experimental outcomes demonstrate that the suggested equalizer achieves significant reductions in bit error rate (55%), distortion rate (45%), computational complexity (48%), and computation cost (75%) while maintaining a high transmission rate (99%). This approach enables the development of high-speed UVLC systems capable of processing data online, thereby advancing state-of-the-art underwater communication.

This paper demonstrates a multi-hop underwater visible light communication (VLC) system using light emitting diodes (LEDs). It is demonstrated that using a multi-hop technique, the overall underwater VLC distance can be increased significantly without adding significant complexity to the physical layer. The performance of the multi-hop system is compared to a single-hop technique utilizing three types of water in the laboratory environment, namely sea, canal, and tap water. The obtained results validate that a multi-hop technique can be a viable solution to overcome the problem of limited underwater communication range.

Ongoing advances in the research and development of visible light communication (VLC) systems has opened up many opportunities for using the concept of light based communication in underwater medium. Water presents a favourable medium for communication of optical wavelengths, particularly in the blue-violet regions of the visible spectrum and parts of ultraviolet wavelength. Prior work in designing underwater VLC systems and exploration of underwater optical wireless channels have largely been limited to theoretical propositions. Experimentation work in this space have not studied the impact of water quality factors on communication, and there exists no clear database of underwater VLC channel studies. Prior work have also largely focused only on horizontal configurations, where the transmitter and receiver are at the same depth underwater. To bridge this gap between existing work and advancement of underwater VLC systems, this paper presents the development of an empirical model for underwater VLC across the vertical configuration and across different water quality factors such as turbidity, salinity and temperature.

In this article, we demonstrate two transfer learning–based dual-branch multilayer perceptron post-equalizers (TL-DBMLPs) in carrierless amplitude and phase (CAP) modulation-based underwater visible light communication (UVLC) system. The transfer learning algorithm could reduce the dependence of artificial neural networks (ANN)–based post-equalizer on big data and extended training cycles. Compared with DBMLP, the TL-DBMLP is more robust to the jitter of the bias current (Ibias) of light-emitting diode (LED), which indicates that TL-DBMLP does not require further training in Ibias varying UVLC system. In terms of voltage peak-to-peak (Vpp) varying VLC system, DBMLP requires a training set with a size of more than 105 and 50 training epochs, which quantitatively prove the effectiveness of DBMLP in reducing reliance on large amount of training epochs. On the counterpart, the TL-DBMLP only requires a training set with a size of less than 2×104 and 10 training epochs, which quantitatively prove the effectiveness of DBMLP in reducing reliance on big data. Finally, we experimentally demonstrate that transfer learning can effectively reduce ANN dependence on extensive size training data and large amount of training epochs, whether in VLC systems with varying Ibias and varying Vpp.

To satisfy the demands of high data rate underwater applications such as image and real-time video transmission, underwater visible light communication (UVLC) has emerged as an alternative to acoustic signaling. One of the major impairments in UVLC systems is turbulence-induced fading as a result of temporal variations in temperature and salinity. Furthermore, unlike the horizontal links modeled with fixed turbulence strength, vertical links experience varying turbulence strength based on the depth-dependent temperature and salinity profiles. In this paper, we consider a multiple-input multiple-output (MIMO) UVLC link over a vertical turbulence channel which is modeled as the concatenation of multiple non-mixing layers. Under the assumption of cascaded log-normal channel model, we derive the outage probability of vertical MIMO UVLC links and quantify the diversity gain in terms of the number of transmitter/receivers.

Underwater optical wireless communication (UWOC) is an interesting area need to be explored. In recent years, underwater visible light communication (UVLC) is expected to act as an alternative candidate in next-generation underwater wireless optical communications for the Internet of Underwater Things. The challenge is that the absorption, scattering, diffraction effect and turbulence of water medium, which can bring large attenuation and nonlinearity penalty. Therefore, it needs to apply lots of advanced modulation formats which including orthogonal frequency division multiplexing (OFDM), discrete Fourier transform spread orthogonal frequency division multiplexing (DFT-S OFDM) and carrier-less amplitude and phase modulation (CAP) in UVLC system. What's more, advanced digital signal processing algorithm such as pre/post equalization, nonlinear compensation and other algorithms based on artificial intelligence are needed. In this paper, we introduce three advanced modulation formats including DFT-S OFDM, OFDM and CAP employing probabilistic shaping for high-speed LED based underwater visible light communication.

ABSTRACTThis research endeavors to investigate the end‐to‐end performance metrics of a submerged AUV within the Single‐Input‐Multi‐Output underwater visible‐light communication (UVLC) system operating in moderate‐to‐strong turbulence channel conditions while accounting for pointing errors. In this study, the closed‐form expressions of analytical work are derived by using the probability density function and cumulative density function with a specific focus on Gamma‐Gamma distribution channel. The analytical framework is used to gain insight into the performance of developed UVLC system within the mixed waters. To validate the accuracy of the analytical approach under the Monte Carlo simulation methodology is used to improve the quality of service. The designed system aims to improve the reliability, practical applicability, and efficiency of the UVLC system, which is the paramount importance in emerging communication technologies.

Machine Learning (ML) technologies have been widely used for blind signal classification problem (Modulation Classification) to automatically identify the modulation schemes of Radio Frequency (RF) signal from complex- valued IQ (In-phase Quadrature) samples. Traditional ML methods usually have two stages where the first stage is to manually extract the features of the IQ symbols by subject matter experts and the second stage is to feed the features to an ML algorithm (e.g., a support vector machine) to develop the classifier. The state- of-art technology is to apply Deep Learning (DL) technologies such as Convolutional Neural Network (CNN) or Recurrent Neural Network (RNN) directly to the complex-value IQ symbols to train a multi-class classifier. In this effort we are focused on the modulation and coding rate classification problems for multi-carrier Orthogonal Frequency-Division Multiplexing (OFDM) signals. The generated OFDM data set consists of over 480k instances across four different modulations and three different coding rates which include BPSK + 1=2, BPSK+3=4, QPSK+1=2, QPSK+3=4, QAM16+1=2, QAM16+3=4, QAM64+2=3 and QAM64+3=4. Deep Learning (DL) models can successfully catch the features of OFDM modulations and identify the modulation scheme from the baseband IQ symbols. Four DL models including Residual Neural Network 34(Resnet34), Resnet18, Squeezenet and Long Short-Term Memory (LSTM) trained over the data set can achieve over 94% overall accuracy for signals across 􀀀10dB to 10dB with step size of 4dB. Among them, Squeezenet and LSTM models have much smaller model sizes which can be easily loaded in resource-limited edge computing platforms. To the best of our knowledge, coding rate classification of OFDM signals has not been studied in previous works. Therefore, three different DL models including a single-stage coding rate classifier, a combination of modulation and coding rate classifier, and a two-stage coding rate classifier are developed to identify the coding rate. However, coding rates can only be identified for low-order modulation schemes BPSK and QPSK but not for high-order modulation schemes QAM16 and QAM64. Further investigation is needed to understand the coding rate classification for high-order QAM signals.

High power green micro-LED is essential for extending the range of free-space and underwater visible light communication systems. We demonstrate a ~520nm micro-LED with 11.5mW output power, achieving a data rate of 3.77Gbps.

This paper presents the design, development and test of an underwater visible light communication system for the deep sea divers. To achieve a practical communication range, frequency modulation technique has been utilized to modulate the voice signal which is eventually transmitted using visible light. Blue and green light emitting diodes have been chosen to design a full duplex system for simultaneous communication. The proposed system achieved a bidirectional communication range of approximately 3.5 m using off-the-shelf components and without using any optical concentrator.

The Internet of Underwater Things (IoUT) has attracted significant attention from researchers due to the fact that seventy percent of the Earth's surface is covered by water. Reliable underwater communication is the enabler of IoUT. Different carriers, such as electromagnetic waves, sound, and light, are used to transmit data through the water. Among these, optical waves are considered promising due to their high data rates and relatively good bandwidth efficiency, as water becomes transparent to light in the visible spectrum (400-700 nm). However, limitations such as link range, path loss, and turbulence lead to low power and, consequently, a low signal-to-noise ratio (SNR) at the receiver. In this article, we present the design of a smart transceiver for bidirectional communication. The system adapts the divergence angle of the optical beam from the transmitter based on the power of the signal received. This paper details the real-time data transmission process, where the transmitting station consists of a light fidelity (Li-Fi) transmitter with a 470 nm blue-light-emitting diode (LED) and a software-defined radio (SDR) for underwater optical communication. The receiving station is equipped with a Li-Fi receiver, which includes a photodetector with a wide field of view and an SDR. Furthermore, we use pulse position modulation (PPM), which demonstrates promising results for real-time transmission. A key innovation of this paper is the integration of the Li-Fi system with the SDR, while the system adapts dynamically using a servo motor and an Arduino microcontroller assembly. The experimental results show that this approach not only increases throughput but also enhances the robustness and efficiency of the system.

In this work, we proposed a Bidirectional Temporal Convolutional network (BiTCN) model to process signal post-equalization in underwater visible light communication channel (UVLC) for the first time. This strategy is more resistant to nonlinearities compared to traditional equalizers like least mean square (LMS) equalizer and avoids the inefficiency issue caused by the sequential processing requirement of Bidirectional Gate Recurrent Unit (BiGRU) networks, through this approach, we successfully conduct a transmission of a 64quadrature amplitude modulation-carrierless amplitude (64QAM-CAP) signal in a 1.2m underwater experiment using a blue laser operating at a wavelength of 450 nm. Remarkably, our system achieves a high data rate of 16.2 Gbps, all while maintaining the bit error rate (BER) below the 7% hard decision forward error correction (HD-FEC) threshold of 3.8E-3. Under optimal Peak-to-Peak Voltages (VPP) and bias current conditions, compared to the post equalization methods of LMS+Volterra and BiGRU, our proposed method achieved speed improvements of 550 and 210 Mbps, respectively.

As the underwater applications require reliable, fast and efficient communications, the underwater visible light communication (VLC) is attracting more and more attention. Due to the complicated channel condition of the underwater environment, accurate channel estimation is still challenging for underwater VLC systems. This paper investigates the underwater VLC channel estimation problem by following two main assumptions. Firstly, the number of the channel paths is assumed to be much smaller than the channel length, which will obtain the sparsity property in the distance domain. Secondly, the attenuation coefficient is assumed to be approximately linear with the frequency, which is true when the modulation bandwidth is much more narrow compared with the carrier frequency. The compressive sensing (CS) framework is formulated based on the two assumptions, whose sensing matrix has a large coherence. In this condition, the Bayesian CS algorithm is adopted instead of the conventional schemes, and the subspace scheme is also investigated for comparison. The simulation results demonstrate that the proposed scheme outperforms the conventional ones in terms of mean square error (MSE), hence better system performance can be expected.