Source-linked AI summary
6G for Vehicle-to-Everything (V2X) Communications: Enabling Technologies, Challenges, and Opportunities
Md. Noor-A-Rahim, Zilong Liu, Haeyoung Lee, M. Omar Khyam, Jianhua He, Dirk Pesch, Klaus Moessner, Walid Saad, H. Vincent Poor
TL;DR
Next-generation V2X needs intelligent communication supporting hyper-fast, ultra-reliable, low-latency massive information exchange. This article surveys 6G-V2X enabling technologies and machine-learning advances, highlighting challenges and opportunities for future research.
Problem
Existing V2X technologies have limited coverage, data rate, QoS guarantees, and bounded channel-access performance in dense, high-mobility environments.
Method
The paper comprehensively surveys forward-looking 6G-V2X technologies across materials, algorithms, architectures, computing, and machine-learning applications.
Results
The article identifies enabling technologies, reviews machine-learning advances, and discusses their strengths, challenges, maturity, and opportunities for 6G vehicular networks.
Takeaways & Limitations
The survey provides academic and industry professionals with insights intended to stimulate innovative research toward next-generation 6G-based V2X.
Takeaways & Limitations
Practical quantum-computing and communication systems remain in their infancy, so their role may emerge only late in 6G development or beyond.
Abstract
from arXiv · showhide
We are on the cusp of a new era of connected autonomous vehicles with unprecedented user experiences, tremendously improved road safety and air quality, highly diverse transportation environments and use cases, as well as a plethora of advanced applications. Realizing this grand vision requires a significantly enhanced vehicle-to-everything (V2X) communication network which should be extremely intelligent and capable of concurrently supporting hyper-fast, ultra-reliable, and low-latency massive information exchange. It is anticipated that the sixth-generation (6G) communication systems will fulfill these requirements of the next-generation V2X. In this article, we outline a series of key enabling technologies from a range of domains, such as new materials, algorithms, and system architectures. Aiming for truly intelligent transportation systems, we envision that machine learning will play an instrumental role for advanced vehicular communication and networking. To this end, we provide an overview on the recent advances of machine learning in 6G vehicular networks. To stimulate future research in this area, we discuss the strength, open challenges, maturity, and enhancing areas of these technologies.
I. INTRODUCTION
The paper motivates 6G-V2X as a foundation for connected autonomous vehicles and surveys its technological evolution, emphasizing broad enabling advances and machine learning.
- V2X includes vehicle-to-vehicle, infrastructure, pedestrian, vulnerable-road-user, and network communications for intelligent transportation systems.
- 6G-supported V2X is envisioned to improve user experience, road safety, air quality, and transportation applications.
- Existing V2X technologies center on DSRC-based and cellular-based vehicular networks, with DSRC facing coverage, data-rate, QoS, and access-delay limitations in dense, mobile settings.
- LTE-eV2X improved reliability, latency, and data rates while retaining backward compatibility with Release 14 LTE-V2X.
- The article takes a forward-looking, inclusive survey approach spanning new materials, algorithms, system structures, and machine learning for future 6G-V2X.
II. OVERVIEW OF 6G-V2X COMMUNICATIONS
The overview presents 6G-V2X as a paradigm shift toward intelligent, diversified networks that integrate emerging communication, computing, and learning capabilities.
- A. Why 6G-V2X?: 6G-V2X is introduced as a response to growing autonomous-vehicle connectivity demands and the capacity limits of existing wireless networks.
- A. Why 6G-V2X?: Machine learning is envisioned to support intelligent and autonomous radios through context awareness, self-aggregation, adaptive coordination, and self-configuration.
B. Key 6G-V2X Technologies
The paper organizes key 6G-V2X enablers into interacting technologies that address communication, computing, security, intelligence, and vehicle-efficiency needs.
- The envisioned 6G-V2X platform is intelligent, autonomous, and user-driven, with technologies classified as revolutionary or evolutionary.
- Figure 3 connects candidate technologies with communication, computing, and security aspects of 6G-V2X.
- Millimeter-wave, visible-light, and terahertz communications can provide high data rates, while multiple access technologies and resource allocation target low-latency, reliable exchange.
- NOMA supports massive connectivity and faster random access for low-latency tactile V2X, while blockchain is discussed for securing such networking.
- 6G-V2X is expected to improve electric-vehicle driving and battery efficiency through predictable roads, optimized driving modes, and charging guidance.
III. REVOLUTIONARY TECHNOLOGIES FOR 6G-V2X
The paper introduces revolutionary technologies for 6G-V2X, including programmable radio environments, while highlighting high-mobility channel challenges and coverage limitations.
- Vehicular mobility creates doubly selective channels that distort signals and constrain channel capacity while increasing training overhead.
- Large subcarrier spacing and dense pilots improve robustness in high-mobility V2X but reduce spectral efficiency and increase receiver complexity.
- A. Programmable V2X Environment: Intelligent reflecting surfaces are programmable metasurfaces with passive reconfigurable elements that customize radio-wave propagation.
- A. Programmable V2X Environment: IRSs can enhance channel conditions and coverage for mmWave or THz links operating in coverage-limited or non-line-of-sight scenarios.
- A. Programmable V2X Environment: At blocked intersections, IRSs installed on surrounding buildings are proposed to mitigate weak perpendicular-street V2V links.
B. Tactile Communication in V2X
Tactile V2X shifts communication toward real-time haptic and control information, supporting immersive and safety-related vehicular applications. Its stringent rate, latency, reliability, and security requirements remain difficult to satisfy in mobile environments.
- Tactile Communication in V2X: Tactile communication enables real-time transmission of touch, motion, vibration, and surface-texture information for steer/control-oriented applications.The paper connects tactile V2X to remote driving, platooning, driver training, and haptic safety warnings.
- Tactile Communication in V2X: Tactile V2X can support immersive in-vehicle experiences, vehicular control applications, and haptic warnings for improving driving safety.The paper also describes potential assistance for vulnerable road users through haptic signals.
- Tactile Communication in V2X: A demonstrated system enhanced cycling behavior without negatively impacting concentration levels.
- Tactile Communication in V2X: Tactile communication requires extremely high speed, extremely low latency, and reliable exchange of large haptic-information volumes.These requirements are particularly difficult in highly mobile vehicular environments and motivate highly reliable high-rate low-latency communications.
- Tactile Communication in V2X: Brain-controlled vehicles could improve independence for people with disabilities, but scalable operation requires wireless coverage, availability, speed, and low latency.THz communications are identified as a potential high-throughput, low-latency enabler for brain-vehicle interfacing.
- Tactile Communication in V2X: V2X security requirements vary by service, while blockchain adoption remains constrained by high latency, limited throughput, and scalability problems.Mission-critical messages require ultra-resilient security, whereas multimedia services may use lightweight security.
E. Terahertz-assisted V2X Networks
Terahertz communication offers very high throughput for future V2X, while quantum computing is discussed as a possible security and optimization technology. Both areas face substantial deployment and maturity constraints.
- Terahertz-assisted V2X Networks: THz communication operates at 0.1-10 THz and may provide transmission rates from hundreds of Gbps to several Tbps.The paper links this throughput to ultra-fast massive data transfer and other advanced V2X scenarios.
- Terahertz-assisted V2X Networks: THz-assisted V2X still requires advances in transceivers, materials, antennas, propagation measurement, channel modeling, and waveforms.Propagation must be characterized across highway, urban, and in-vehicle scenarios.
- Terahertz-assisted V2X Networks: Practical quantum computing and communication systems remain in their infancy and may not contribute until late 6G or beyond.Current quantum chips operate near zero Kelvin, making them at best usable on vehicular infrastructure without further thermal-stability research.
- Terahertz-assisted V2X Networks: Quantum computing may enhance V2X security because quantum entanglement cannot be cloned or accessed without tampering, according to the paper.The paper presents quantum key distribution as the security framework associated with this potential.
- Terahertz-assisted V2X Networks: Quantum computing is proposed to accelerate complex optimization and potentially achieve optimality with reduced complexity in data-intensive tasks.The example is finding an optimum geographic route with multiple objectives while processing and training advanced machine-learning algorithms.
IV. EVOLUTIONARY TECHNOLOGIES FOR 6G-V2X
The paper surveys evolutionary 6G-V2X technologies that build on established research while addressing demanding future requirements. Hybrid RF-VLC communication and NOMA offer complementary routes toward higher capacity, connectivity, and reliability, but deployment and standardization issues remain.
- IV. EVOLUTIONARY TECHNOLOGIES FOR 6G-V2X: Evolutionary technologies have reached some maturity through prior research, testing, and deployment but require further development and trials for 6G-V2X.
- IV. EVOLUTIONARY TECHNOLOGIES FOR 6G-V2X: Standalone RF-based V2X may struggle with interference, latency, and packet delivery in dense scenarios requiring extremely high rates and low latency.The paper presents RF-VLC integration as one alternative.
- IV. EVOLUTIONARY TECHNOLOGIES FOR 6G-V2X: VLC can provide up to 100 Gbps, low power consumption, enhanced security, anti-electromagnetic interference, and low setup cost using existing vehicle or street lighting.
- IV. EVOLUTIONARY TECHNOLOGIES FOR 6G-V2X: VLC supports V2V links through headlights or backlights, V2X links through traffic lights, and links through street lights, including potential optical backhaul.
- IV. EVOLUTIONARY TECHNOLOGIES FOR 6G-V2X: Hybrid RF-VLC deployment still requires interoperability and solutions for ambient-light interference and mobility-induced received-signal variation.VLC had not been included in the 5G-V2X standard discussed by the paper.
- IV. EVOLUTIONARY TECHNOLOGIES FOR 6G-V2X: NOMA enables multiple users to share time-frequency resources concurrently and can support massive connectivity and distributed V2V scheduling.Open issues include coexistence with OMA, balancing overloading, reliability, and fairness, and designing scalable CAV schemes.
C. Exploration of Multiple Radio Access Technologies
6G-V2X may need coordinated use of mmWave, THz, and other radio-access technologies to meet demanding throughput and latency targets. Directional propagation and multi-radio coordination create unresolved challenges for beam management, access, scheduling, and resource allocation.
- Exploration of Multiple Radio Access Technologies: mmWave and THz bands provide richer frequency resources than sub-6 GHz and are vital for targets such as Tbps rates and sub-millisecond access latency.
- Exploration of Multiple Radio Access Technologies: Propagation loss and blockage at mmWave and THz frequencies necessitate directional beamforming, complicating V2V connectivity at high vehicle speeds.
- Exploration of Multiple Radio Access Technologies: Multi-radio 6G-V2X requires new coordination schemes for fast link configuration, beam management, channel access, autonomous scheduling, congestion control, and interference management.
- Exploration of Multiple Radio Access Technologies: Intelligent reflective surfaces combined with high frequencies may help alleviate some multi-radio challenges.
- Exploration of Multiple Radio Access Technologies: Radio resource management must support QoS across multi-radio technologies and increased algorithmic complexity.The paper identifies context awareness and cross-layer design as support for advanced resource allocation.
- Exploration of Multiple Radio Access Technologies: A hybrid RRM framework can combine dedicated resources that guarantee basic QoS with shared resource pools for connected-vehicle communications.
E. New Multicarrier Scheme
6G-V2X must support reliable, high-rate communication under extreme mobility and diverse network conditions. The section surveys OTFS, aerial platforms, satellite links, and fog computing as enabling approaches.
- Vehicles moving at 1000 km/h or higher can experience reduced channel coherence time and rapidly varying fading.
- OTFS spreads symbols across the time-frequency domain to convert time-varying multipath channels into relatively static delay-Doppler representations.
- UAVs can act as aerial radio access points providing relaying, caching, and computing, particularly in dense vehicular environments.
- Satellite-assisted V2X is a potential communication platform beyond current standards, where satellites are presently used primarily for localization.
- Fog and edge nodes support low-latency vehicular services by processing data near users and offloading complex computations from vehicles.
H. Integrated Sensing, Localization and Communication
Connected vehicles require communication together with high-resolution sensing and accurate localization for situational awareness. The paper presents their integration as a converged 6G capability while identifying radio-resource coordination as a challenge.
- Situational awareness depends on rapid communication, high-resolution sensing, and high-accuracy localization for vehicles, pedestrians, obstacles, and infrastructure.
- Sensing, localization, and communication share radio-wave processing operations and hardware, motivating their integration into one converged RF system.
- ISAC and ILAC are emerging design paradigms for integrating sensing or localization with communication in 6G mobile networks.
- Centimeter-level localization accuracy is expected through ultra massive MIMO, mmWave technologies, and UAV or satellite networks.
- Integrated communication and control can support coordinated vehicle platooning, which is associated with increased road capacity and fuel efficiency.
- ML can process camera, LiDAR, GPS, and sensor streams for data-driven decisions, while this paper focuses on its network perspective.
A. ML for New Physical Layer
The paper examines ML as a tool for adapting physical-layer and resource-management functions to highly mobile, heterogeneous 6G-V2X environments. It also highlights data, convergence, coordination, and model-mismatch challenges.
- Adaptive PHY design is desirable because diverse V2X services require different coding, modulation, waveform, and multiple-access schemes.
- High Doppler spread causes rapid fading and can undermine conventional channel estimation, while offline-trained ML models may mismatch real channels.
- Handcrafted channel codes based on simple models may lose error-correction capability in high-mobility environments with rapidly varying interference.
- Frequent beam switching and beam training remain challenges for vehicular mmWave beamforming and massive MIMO.
- Radio-resource management must address high mobility, heterogeneous networks, and varied QoS requirements without repeatedly rerunning conventional algorithms.
- ML-based prediction can forecast bursty traffic loads and assign channels to links, helping avoid congestion and accelerate channel allocation.
- RL and DRL address resource allocation when labeled datasets are scarce or state-action spaces are large, but convergence and multi-agent coordination remain difficult.
C. ML for Security Management
6G-V2X security is complicated by broadcast communication, heterogeneous access, and exchanged private information. The paper reviews ML-based intrusion detection and control while stressing validation and coordination requirements.
- Broadcast vehicular communication and diverse connectivity expose 6G-V2X to authentication, authorization, and data-forgery attacks.
- Because V2X exchanges identities and trajectories over wireless links, secure user identification and authentication are particularly important.
- Supervised learning can identify abnormal vehicle behavior, but dependence on labeled data may limit detection of novel or unknown attacks.
- Physical-layer security can complement cryptographic methods when secret-key management is difficult in dynamic networks.
- ML can support control and communication mechanisms intended to prevent data-injection attacks in vehicular networks and platoons.
- ML-based security solutions require end-to-end validation and synchronization across network layers to maintain secure communications.
D. Federated Learning for 6G-V2X
Federated learning is presented as a privacy-preserving approach for training machine-learning models in 6G-V2X, while its effective deployment faces substantial vehicular and wireless-network constraints.
- Local model training can reduce the time, cost, and latency of transferring edge-generated samples to remote clouds.Training data are generated at base stations and vehicles, where changing network conditions can make cloud-based training slow to respond.
- Federated learning addresses privacy and communication-overhead concerns associated with jointly training models across base stations and vehicles.Joint training can improve model accuracy and generalization, but participating nodes may not want to share their training samples.
- Effective federated learning for 6G-V2X requires scalable reinforcement-learning frameworks, stable initialization, vehicle connectivity support, and deeper study of wireless-channel effects.Vehicles may have brief connectivity or be out of range, while wireless errors and delays can affect federated-learning performance.
- The paper surveys federated learning among recent machine-learning advances and identifies associated challenges and opportunities for next-generation 6G-V2X.