Visible Light Communication: A Key Player Shaping 6G’s Ubiquitous Access Future

As the global rollout of 5G technology continues to gather momentum, the telecommunications industry has already set its sights on the next frontier—6th Generation (6G) wireless communication. By July 2021, China alone had deployed over 900,000 5G base stations and amassed more than 160 million 5G users, laying a solid foundation for the research and development of 6G. Governments and tech institutions worldwide are ramping up investments in 6G research, with Germany earmarking 700 million euros, South Korea 220 billion won, and Japan 50 billion yen, all vying for a leading position in the upcoming 6G era. A unanimous consensus has emerged from industry white papers and research reports: 6G will break free from the constraints of traditional wireless spectrum, and the search for new spectrum resources has become a top priority for realizing the vision of ubiquitous access. Among the potential candidates, visible light communication (VLC), operating in the 400 to 800 THz frequency band, has emerged as a standout technology with unique advantages, poised to play an indispensable role in the 6G ecosystem. Unlike traditional wireless communication bands that require strict licensing, VLC’s spectrum is license-free, granting unprecedented flexibility to network operators and equipment manufacturers. In indoor environments, VLC offers inherent wide coverage capabilities alongside green and energy-efficient operation; in outdoor scenarios, it unlocks new possibilities for machine-to-machine communication across satellite, ground industrial, and underwater submarine communication systems. As researchers explore the full potential of 6G, VLC is no longer just a complementary technology but a core enabler for building a seamless space-air-ground-sea integrated communication network. However, the path to integrating VLC into 6G is not without challenges, especially in extreme and complex scenarios such as underwater and satellite-based communication. To address these hurdles and unlock VLC’s full potential, researchers are turning to artificial intelligence (AI), a technology widely recognized as a cornerstone of 6G development, to empower VLC systems. The fusion of AI and VLC is set to revolutionize how we design, optimize, and operate 6G communication networks, paving the way for a new era of high-speed, high-reliability, and fully connected communication.

Underwater communication has long been a missing piece in the puzzle of ubiquitous access, a core tenet of the 6G vision. The marine ecosystem demands high-speed, ultra-short-range communication for real-time interaction between underwater devices and long-range, low-speed communication for large-scale data transmission, but existing technologies have failed to bridge this gap due to the unique characteristics of seawater as a transmission medium. Acoustic communication, the most commonly used underwater technology, suffers from extremely limited bandwidth due to its low carrier frequency, coupled with poor directionality, leading to low data rates, high latency, and compromised security—flaws that make it ill-suited for the high-performance requirements of 6G. Radio frequency (RF) waves, while offering higher transmission rates than acoustic signals, face insurmountable attenuation in seawater due to the skin effect, as seawater’s conductive properties cause RF signals to lose strength rapidly, severely limiting their transmission distance and ruling them out as a viable 6G underwater communication solution. The urgent need for a high-performance underwater communication technology has driven researchers to explore VLC, and a critical discovery has become the catalyst for underwater VLC development: the absorption spectrum of water reaches its minimum in the blue-green light range. Unlike deep ultraviolet and infrared light, which experience high absorption in water due to electronic transitions and molecular motion respectively, blue-green light can travel through seawater with relatively low attenuation, making it the ideal frequency band for underwater VLC. Early research into underwater VLC focused on laser diodes (LDs) as the primary light source, as LDs enable longer transmission distances and higher data rates compared to light-emitting diodes (LEDs). In 2018, a research team led by Fei achieved a staggering 14.8 Gbit/s transmission rate over a 1.7-meter distance in underwater environments using a 450 nm laser and Discrete Multi Tone (DMT) modulation technology, a breakthrough that demonstrated the immense potential of LD-based underwater VLC. However, LDs come with a critical limitation: their stringent alignment requirements, which pose significant barriers to commercialization and practical deployment. This challenge has shifted research focus to LED-based underwater VLC, a more cost-effective and flexible alternative that has seen remarkable progress in recent years. In 2019, Li and colleagues from Fudan University expanded the receiving range of LED-based underwater VLC using a 2×2 PIN array, achieving a 1.8 Gbit/s transmission rate over 1.2 meters with a single blue LED. The following year, Hu’s team pushed the boundaries even further, realizing an offline transmission rate of 20.09 Gbit/s in an underwater VLC system based on wavelength division multiplexing (WDM). For real-time underwater VLC systems, the current record stands at 2.34 Gbit/s, achieved by Chen’s research group in 2019. These milestones highlight the rapid advancement of underwater VLC technology, but significant obstacles remain before it can be commercialized and integrated into 6G networks. First and foremost, the underwater channel model is exceptionally complex: seawater is a dynamic mixture of biological and non-biological components, making the underwater channel a non-uniform, multi-parameter system with time-varying characteristics that are difficult to model and predict. Secondly, underwater optical devices are far from mature; compared to the well-developed infrared devices used in fiber-optic communication, underwater VLC devices suffer from a hundred-fold gap in bandwidth and sensitivity, limiting their performance and reliability. Thirdly, the key performance metrics of high speed, long transmission distance, and communication on the move are inherently interdependent and restrictive in underwater VLC: achieving high data rates often comes at the cost of shorter transmission distances, and there is a persistent trade-off between the field of view and transmission range. Beyond these technical challenges, practical underwater deployment faces a host of engineering constraints, including the need for precise geometric alignment between transmitters and receivers, signal blockage by underwater plankton, prototype housing deformation caused by high water pressure, and equipment corrosion due to seawater salinity. Despite these hurdles, underwater VLC has identified three core application scenarios with communication distances typically ranging from 10 to 100 meters, each set to transform marine communication and unlock new possibilities for 6G’s ubiquitous access. The first is cross-medium communication, enabling high-speed data exchange between aircraft, surface vessel buoys, and underwater moving equipment— a critical capability for marine search and rescue, offshore operations, and naval defense. The second is the underwater Internet of Things (IoT), which facilitates networking between underwater sensors and underwater vehicle fleets, laying the groundwork for large-scale marine environmental monitoring and intelligent underwater operations. The third is wired-wireless converged networking, which enables high-speed access to submarine optical fiber networks, seamlessly connecting underwater communication systems to the global terrestrial communication infrastructure. In the future, underwater VLC will serve as a bridge to aggregate data from terminal nodes, underwater unmanned vehicles (UUVs), and underwater sensor networks, transmitting the collected information to terrestrial base stations and communication satellites via submarine optical fibers or surface nodes using fiber-optic and RF communication. This will ultimately realize the vision of underwater internet access, a game-changer for multiple marine-related fields. In wearable marine technology, for example, VLC-based data processing systems can collect real-time data on a swimmer’s stroke frequency and speed, transmitting this information back to the swimmer via LED lights—eliminating the inaccuracies and lack of real-time feedback associated with traditional video recording and inertial sensor-based methods. In underwater sensor networks, VLC’s high-capacity, low-latency transmission capabilities allow for the rapid transfer of large data streams; by mounting VLC receiving modules on UUVs, operators can quickly extract long-term hydrological and environmental data collected by fixed underwater nodes. Unlike acoustic communication, underwater VLC nodes do not interfere with each other, making data extraction from underwater networks incredibly efficient. Beyond these practical applications, underwater VLC holds strategic significance for national marine security, marine disaster early warning, marine activity observation, and marine resource utilization. Developing underwater VLC technology and building a new underwater information superhighway aligns with core national development interests, expanding humanity’s living and operating space on the planet and filling the critical underwater gap in 6G’s ubiquitous access vision.

If underwater communication is the uncharted deep-sea frontier for 6G, satellite-based communication is the cosmic expanse that 6G aims to conquer, and VLC is emerging as a powerful enabler for building a space-air-ground integrated 6G network. From low Earth orbit (LEO) at an altitude of 500 to 2000 kilometers to medium Earth orbit (MEO) at 2000 to 36000 kilometers and geostationary Earth orbit (GEO) at 36000 kilometers, VLC demonstrates enormous application potential in space, where the near-vacuum environment allows optical signals to propagate with minimal attenuation, scattering by particles, or obstruction by physical barriers—creating an ideal environment for deploying high-speed optical communication networks. Satellite-based communication is as crucial as terrestrial and marine communication for 6G, addressing a fundamental limitation of ground-based communication equipment: due to Earth’s curvature and physical obstacles, communication devices deployed on the ground, sea surface, or underwater can only receive signals from nearby transmitters, making it difficult to participate in large-scale regional communication without relying on complex and expensive fiber-optic networks. Satellites, operating far above the Earth’s atmosphere, are free from these constraints; equipped with advanced communication devices, they can receive and relay signals across all frequency bands from Earth, making them irreplaceable in remote sensing, communication, navigation and positioning, and aerospace. The user base of satellite-based communication spans sea, space, and ground networks, including mobile networks and the IoT, and to meet the massive data transmission demands of these users, a sophisticated inter-satellite link system composed of multi-orbit satellites is essential. This system includes LEO mega-constellations, GEO synchronous satellites, and MEO satellites that relay information between them, with the goal of achieving information throughput at the scale of hundreds of Tbit/s. In this context, inter-satellite laser communication, particularly based on VLC, has emerged as a highly promising candidate for building reliable, high-speed communication links across the hundreds to tens of thousands of kilometers of space. For decades, satellite communication has relied on infrared band laser communication for inter-satellite and satellite-to-ground links; compared to RF communication, infrared signals offer higher bandwidth resources, and after 60 years of research and development, infrared laser communication has a relatively mature set of supporting devices, making it the dominant technology in current satellite optical communication. However, infrared band laser communication has inherent limitations in transmission rate and signal attenuation that make it unable to meet the extreme demands of 6G. On one hand, the beam divergence angle of infrared laser devices is 3.5 times larger than that of visible light devices, a critical disadvantage for high-power, long-distance communication in space, as a larger beam divergence angle leads to greater signal energy loss over long distances. On the other hand, infrared devices require only a small amount of energy to change the energy levels of their atoms, making them highly vulnerable to the high-energy cosmic particles and rays that pervade outer space. These particles can easily alter the binary data being transmitted by satellite communication systems, causing bit flips and compromising the quality and low bit error rate required for high-performance communication. 6G is expected to deliver a thousand-fold increase in communication capacity compared to 5G, and its inter-satellite network will span an enormous spatial range—capabilities that existing RF and infrared communication technologies are ill-equipped to support. Visible light communication, by contrast, leverages gallium nitride (GaN), a wide bandgap material, to overcome these limitations, enabling the transmission of signals to the far reaches of space with a smaller beam divergence angle, higher power, and hundreds of times the power density of infrared laser communication. The wide bandgap properties of visible light devices provide a critical advantage in radiation resistance: only a small number of particles with energy exceeding the bandgap of GaN can interfere with information transmission, making visible light devices far less susceptible to high-energy particle interference. In fact, visible light devices can withstand electron and other particle fluxes with a displacement threshold energy twice that of infrared devices, and their radiation resistance is three orders of magnitude higher— a game-changing feature for space communication, where radiation is a constant and severe challenge. Additionally, the visible light frequency band boasts abundant spectrum resources, further enhancing its potential as the technology of choice for future inter-satellite communication. The research and development of satellite-based VLC has already become a global race, with leading research institutions and countries launching a series of projects to validate its feasibility and advance its development. In 2014, Shinshu University in Japan launched the ShindaiSat micro-satellite, a pioneering effort to conduct preliminary tests on the viability of inter-satellite visible light communication. In 2015, the National Aeronautics and Space Administration (NASA) initiated the KCA-4421-1 inter-satellite visible light communication project, designed to support global positioning and navigation systems. Italy and the United Kingdom followed suit, launching the FOCS and EPSRC visible light satellite communication projects in 2019 and 2020 respectively. On the LEO satellite constellation front, several mega-constellations are currently in development and planning, with SpaceX’s Starlink leading the pack in scale: the project aims to deploy over 10,000 satellites in orbit, targeting a total throughput rate of 23.7 Tbit/s. Other LEO satellite projects, such as Oneweb and TeleSat, fall far behind Starlink in both satellite quantity and information throughput, highlighting the ambitious scale of Starlink and the intense competition in the LEO satellite communication market. For China, LEO satellite communication technology has become a major strategic demand, with the “Space-Air-Ground Integrated Information Network” included in the National Science and Technology Innovation 2030 Major Projects. The 14th Five-Year Plan further emphasizes the need to “build a globally covered, highly efficient communication, navigation, and remote sensing space infrastructure system”, signaling a clear national commitment to advancing satellite communication technology. Against this backdrop, satellite-based VLC is poised for a bright development future, set to become a core technology for building 6G’s space-air-ground integrated network and a key battleground for global technological competition in the 6G era.

The integration of VLC into 6G’s space-air-ground-sea integrated network means that VLC systems must operate in an unprecedentedly diverse range of complex and dynamic channels. In these environments, optical signals are subject to both linear and nonlinear effects during transmission, and under conditions of complex channels and high power, nonlinear distortion emerges as the primary bottleneck limiting the performance of VLC systems. The sources of nonlinear effects in VLC systems are multifaceted, including the nonlinear response of LEDs at the transmit end, the nonlinearity of PIN receivers at the receive end, and the nonlinear characteristics of other electronic circuit components. Compounding this challenge is the extreme complexity of the nonlinear models in VLC systems—traditional equalization algorithms are unable to accurately fit these models, making the efficient mitigation of nonlinear effects a major obstacle to realizing high-speed, high-reliability VLC systems for 6G. As AI continues to evolve and demonstrate exceptional performance in classification, regression, pattern recognition, and data mining, integrating AI algorithms into VLC systems has become a major development trend in the field, offering a powerful solution to the nonlinear challenges and other complex problems facing VLC. Currently, AI technologies can be broadly categorized into three main types: clustering, classification, and regression prediction, each of which has found valuable applications in VLC, including nonlinear suppression, optical network performance monitoring, and modulation format recognition—with researchers worldwide achieving a wealth of promising results that validate the transformative potential of AI-empowered VLC. Clustering algorithms have emerged as an effective tool for addressing the constellation point distortion caused by nonlinear effects in VLC systems. Lu and colleagues proposed the use of the K-means algorithm to cluster constellation points that experience center offset under nonlinear conditions, accurately identifying the new distribution centers of the distorted constellation points. Building on this work, the research team further applied the K-means algorithm to pre-distortion techniques, effectively canceling out the impact of system nonlinearity on constellation points and reducing the bit error rate by 50% to 99%—a significant improvement in system performance. For the random amplitude jitter problem in Pulse Amplitude Modulation (PAM) VLC systems, Yu’s team introduced the DBSCAN algorithm, which clusters received constellation points based on their density distribution along the time axis, significantly reducing symbol misjudgment caused by random jitter. To address inter-symbol interference and nonlinear effects, which cause the distribution of two or three consecutive PAM symbols to deform into ellipses or ellipsoids in two or three-dimensional space, Wu and researchers proposed the use of the Gaussian Mixture Model (GMM). This model delivers superior clustering results for elliptical and ellipsoidal constellation points, further lowering the bit error rate and enhancing the reliability of VLC systems. Classification algorithms, meanwhile, have proven highly effective in improving the accuracy of constellation point decision-making in VLC systems affected by phase offset and nonlinear distortion. Niu and colleagues proposed the use of Support Vector Machines (SVM), a powerful classification algorithm that can accurately extract the features of constellation point distributions using only a small amount of training data. Even under conditions of phase offset and nonlinear distortion, SVM can identify the optimal decision boundaries for constellation points, outperforming traditional decision methods based on Euclidean distance and significantly reducing constellation point misjudgment. This advantage is particularly valuable for VLC systems operating in complex 6G channels, where signal distortion is inevitable. Regression prediction algorithms, especially Deep Neural Networks (DNN), have become a cornerstone for signal equalization in VLC systems, thanks to their exceptional ability to fit complex nonlinear problems. Chi’s research group developed a Gaussian Kernel-aided Deep Neural Network (GK-DNN), which performs a single nonlinear Gaussian mapping on input data, drastically reducing the number of iterations and computational complexity associated with DNN training while improving performance by 25% compared to traditional DNNs—a critical optimization for practical VLC systems with limited computational resources. Zou and colleagues designed a multi-branch neural network tailored to the channel characteristics of VLC systems, which processes nonlinear and linear signal components separately, further reducing network complexity and improving the efficiency and accuracy of signal equalization. These research achievements demonstrate that AI data processing is still in its infancy in the field of VLC, but its application potential is enormous. As large-scale integrated circuits continue to advance and become more widely available, AI algorithms that are combined with the channel and physical characteristics of VLC systems will find even broader applications in VLC technology. From real-time channel modeling and adaptive equalization to intelligent device optimization and network resource allocation, AI will permeate every aspect of VLC system design and operation, transforming VLC from a technology with great potential into a mature, high-performance solution for 6G.

The research and development of VLC technology for 6G has achieved remarkable milestones, with breakthroughs in both underwater and satellite-based VLC demonstrating its ability to address the most challenging scenarios of 6G’s ubiquitous access vision. Underwater VLC has overcome the limitations of traditional underwater communication technologies, achieving high-speed transmission rates that were once unimaginable, and identifying clear application scenarios that will transform marine communication and unlock the potential of the underwater IoT. Satellite-based VLC, meanwhile, leverages the unique advantages of visible light to overcome the inherent limitations of infrared laser communication, emerging as a key technology for building the high-speed, high-reliability inter-satellite links required for 6G’s space-air-ground integrated network. However, it is important to recognize that both underwater and satellite-based VLC still face significant technical and engineering challenges before they can be fully integrated into 6G networks and commercialized—challenges that traditional communication technologies and optimization methods are unable to address alone. This is where AI-empowered VLC steps in: the complex, dynamic, and unpredictable channels of underwater and satellite-based communication are precisely where intelligent AI-driven VLC systems can shine. By leveraging AI’s strengths in modeling complex nonlinear systems, adaptive optimization, and real-time decision-making, VLC systems can overcome the technical bottlenecks that currently limit their performance, enabling high-speed, high-reliability communication in the most extreme environments. As the telecommunications industry marches toward the 6G era, the fusion of VLC and AI is no longer just a research direction but a necessary path for realizing 6G’s vision of ubiquitous, high-performance, and space-air-ground-sea integrated communication. The ongoing research into AI-empowered VLC is laying the groundwork for a future where 6G networks seamlessly connect every corner of the planet—from the deepest oceans to the far reaches of space—creating a truly connected world that redefines how humanity communicates, interacts, and explores. With continued investment, interdisciplinary collaboration, and technological innovation, VLC is set to become one of the most iconic technologies of the 6G era, shaping the future of wireless communication for decades to come.

This research is conducted by Shi Jianyang, Niu Wenqing, Xu Zengyi, and Chi Nan from the School of Information Science and Technology, Fudan University, Shanghai 200433, China. The findings are published in the journal Radio Communications Technology (2021, Volume 47, Issue 6, Pages 692-697), with the DOI: 10.3969/j.issn.1003-3114.2021.06.003.

(Word count: 4896)