Systematic review maps how 5G and 6G sensing could ease urban traffic congestion

Urban traffic congestion is undermining sustainability, energy efficiency and mobility in smart cities, and a new systematic review argues that the missing ingredient is integration. Researchers Benjamin Simwinga, Aaron Zimba and Mayumbo Nyirenda conducted a PRISMA-guided systematic literature review, synthesizing 49 peer-reviewed studies published between 2019 and 2026, to identify key trends, methodologies and applications that leverage 5G/6G-enabled sensing and communication for vehicular congestion reduction. Their conclusion is blunt: existing 5G/6G-enabled intelligent transportation systems research remains fragmented, with limited integration between wireless communication, sensing and vehicular traffic modeling, which reduces its effectiveness against real-world urban congestion.

The review, published in Discover Networks, set out to fill a gap the authors say previous surveys have left open. Current review studies, they note, primarily focus on isolated aspects such as communication architectures, AI techniques or vehicular networking frameworks, and lack a comprehensive and systematic synthesis that examines intelligent wireless communication and sensing for sustainable traffic modeling. By contrast, this study claims to be the first PRISMA-guided systematic review providing a unified cross-layer synthesis of intelligent wireless communication, sensing technologies and AI-driven traffic modeling specifically for sustainable vehicular congestion reduction in 5G/6G-enabled intelligent transportation systems.

The methodological scaffolding is unusually explicit for this field. The authors searched IEEE Xplore, ScienceDirect, SpringerLink, MDPI, arXiv, Zenodo, BMJ and Google Scholar for publications from 2019 to 2026, using keyword combinations pairing 5G, 6G and beyond-5G terms with vehicular communication, sensing, traffic modeling and routing optimization vocabulary. A total of 1,247 records were initially identified. After removal of 148 duplicate records using DOI, title and metadata matching, 1,099 studies remained for title and abstract screening. Screening excluded 950 studies for lack of relevance to intelligent transportation systems, 5G/6G communication or AI-assisted traffic management, and the remaining 149 articles underwent full-text eligibility assessment, where 100 were excluded for limited methodological detail or insufficient alignment with the study objectives. Ultimately, 49 studies satisfied all inclusion criteria and entered the final qualitative synthesis. No formal meta-analysis was conducted, the authors explain, due to heterogeneity in study designs, methodologies and evaluation contexts.

What did the 49 studies show? Findings indicate that existing research widely explores the integration of 5G/6G-enabled vehicular communication, IoT sensing, AI-driven traffic prediction and edge intelligence for real-time congestion management. Nevertheless, full integration across communication, sensing and traffic modeling remains limited, with privacy, scalability, cost and interoperability continuing to pose major challenges. The authors argue that integrating 5G/6G, IoT sensing and AI traffic modeling enables real-time congestion management and efficient routing, directly supporting sustainability goals such as reduced energy use, lower emissions and improved urban mobility, though they caution that only a limited number of reviewed studies explicitly evaluated sustainability using direct metrics such as energy consumption or carbon reduction.

That sustainability caveat is one of the review’s sharpest findings. In their classification of the 49 studies, the authors report that the majority rely on indirect proxies such as latency, scalability or resource efficiency, while a significant portion of communication-focused works do not consider sustainability at all. Furthermore, none of the studies incorporate life-cycle assessment or deployment cost analysis, confirming, in the authors’ words, that sustainability is treated as an inferred or conceptual attribute rather than a rigorously measured outcome. This matters because the field’s promise rests partly on environmental claims: the review notes that intelligent routing approaches have already contributed to improved energy efficiency in 5G environments, and that integrating 6G technologies with edge computing is anticipated to support even more energy-aware and scalable network operations.

The publication trend analysis reveals a field that boomed and then cooled. Publication activity was low in the early period, with only 3 articles in 2019 and 4 in 2020, before a rapid growth phase lifted output to 8 articles in 2021, 10 in 2022 and 10 in 2023, the peak of research intensity in the dataset. A gradual decline followed, with 8 articles in 2024, 5 in 2025 and 1 in 2026, which the authors attribute partly to a transition toward newer research directions and early-stage publication coverage for the most recent year. Thematically, the largest proportion of studies, 34.7 percent, focus on 6G-V2X and next-generation vehicular networking technologies, followed by federated learning and knowledge-sharing approaches at 16.3 percent, security and blockchain-enabled transportation systems at 14.3 percent, AI/ML-enabled and edge intelligence and 5G-V2X communications at 12.2 percent each, and UAV-assisted communications at 10.2 percent. IEEE Xplore contributed the largest share of publications, 28 of the 49 papers, with MDPI accounting for 10, Elsevier 9 and other sources 2.

Beneath the thematic spread, the review identifies recurring performance trade-offs that it says the literature has persistently failed to resolve. A major conflict exists between security and latency in vehicular communication systems: blockchain-enabled mechanisms improve trust, transparency and data integrity in V2X environments but introduce additional computational overhead that conflicts with ultra-low-latency requirements, while lightweight security schemes reduce communication delay but weaken resilience against malicious attacks. A similar trade-off emerges in distributed intelligence, where edge computing architectures reduce end-to-end latency through localized processing yet lack the global contextual awareness of centralized cloud approaches, and federated learning enhances privacy preservation and scalability through decentralized model training but introduces communication overhead and convergence instability under high mobility conditions. At the infrastructure level, the authors observe that the transition from 5G to 6G shifts limitations from performance bottlenecks toward orchestration complexity, a shift in system design complexity rather than a complete resolution of existing constraints.

Non-terrestrial networks add another layer of tension. UAV-based communication systems provide flexible and low-latency connectivity in dynamic urban environments but are constrained by limited endurance and energy capacity, while satellite-based systems offer wide-area coverage and resilience yet suffer from higher propagation delays. Hybrid terrestrial–non-terrestrial architectures improve overall system robustness but significantly increase coordination and interoperability complexity. The review also notes that while advancements in low Earth orbit satellite constellations have improved latency and link performance, these gains remain context-dependent and have not yet been consistently validated in large-scale real-world deployments. UAV-assisted studies demonstrate enhanced aerial–edge connectivity and flexible communication, the authors write, but remain limited by energy constraints, regulatory restrictions and deployment complexity.

On the maturity question, the review draws a clear line between what is deployable and what is still aspirational. The evidence reviewed suggests that 5G-enabled ITS technologies are approaching deployment maturity, whereas many proposed 6G vehicular networking capabilities should be interpreted as promising future research directions that require substantial experimental validation before widespread implementation. Current 5G-V2X and edge intelligence solutions have demonstrated tangible improvements in latency reduction, connectivity and distributed learning efficiency, particularly when integrated with AI and federated learning frameworks. In contrast, 6G-enabled ITS architecture remains predominantly conceptual and vision-based, focusing on ultra-low latency communication, pervasive intelligence and large-scale autonomous coordination, but these frameworks have not yet reached practical implementation and are mainly supported by simulation or theoretical studies. Federated learning itself, the authors add, remains constrained in vehicular environments by mobility, intermittent connectivity, data heterogeneity and model convergence instability, placing FL-based solutions at an early to moderate maturity stage rather than fully deployable large-scale frameworks.

The practical implications the authors draw are correspondingly grounded. They recommend prioritizing hybrid edge–cloud architectures with embedded AI inference at roadside units and onboard vehicles to enable real-time congestion prediction and adaptive traffic control, and they suggest UAV-assisted sensing and NTN-based connectivity can enhance monitoring coverage in dense urban areas, accident-prone zones and infrastructure-limited regions. At the policy level, they call for spectrum harmonization between terrestrial and non-terrestrial networks to enable seamless handovers and consistent quality of service, standardization of federated learning evaluation metrics covering privacy, convergence and communication efficiency within ongoing 3GPP Release 18 and beyond, and strengthened regulatory frameworks for cross-domain data sharing. The thematic synthesis indicates, they conclude, that the most practical pathway toward scalable and real-time intelligent transportation systems lies in the deployment of hybrid edge–cloud architectures supported by 5G/6G connectivity and privacy-preserving federated learning, combining edge-based real-time inference with cloud-level global optimization. Whether that pathway can deliver the reduced emissions and energy savings the field promises will depend, on this evidence, on researchers finally measuring sustainability directly rather than inferring it from faster networks.

Subject of Research: A systematic review of 5G/6G-enabled wireless communication, sensing and AI-driven traffic modeling for sustainable urban congestion reduction

Article Title: A systematic review of intelligent wireless communication and sensing for sustainable 5G and 6G enabled intelligent transportation systems

Article References: Simwinga, B., Zimba, A., & Nyirenda, M. (2026). A systematic review of intelligent wireless communication and sensing for sustainable 5G and 6G enabled intelligent transportation systems. Discover Networks, 2(1), Article 26. https://doi.org/10.1007/s44354-026-00043-0

Image Credits: AI Generated

DOI: 10.1007/s44354-026-00043-0

Keywords: 5G, 6G, intelligent transportation systems, V2X, traffic congestion, federated learning, edge computing, UAV sensing, IoT, sustainability, systematic review, PRISMA