PaperlayerView exact revision on arXiv

Source-linked AI summary

A survey on road safety and traffic efficiency vehicular applications based on C-V2X technologies

Ignacio Soto, Maria Calderon, Oscar Amador, Manuel Urueña

arXiv:2403.17851v1cs.NI

TL;DR

Connected-vehicle applications seek road-safety and traffic-efficiency benefits, while research on C-V2X technologies is still developing. This survey describes the technologies, classifies communication types, and organizes application research; it finds varied implementation approaches and identifies areas needing further work.

  • Problem

    The survey addresses how cellular technologies can support communication-based road-safety and traffic-efficiency applications for connected vehicles.

    Source: §1 · p. 2, §6 · p. 53

  • Method

    The authors describe C-V2X technologies, classify communication types, and use that classification to organize research on connected-vehicle applications.

    Source: §6 · pp. 53–54

  • Results

    The reviewed work shows different ways C-V2X technologies can implement communication-based vehicle applications, while the survey identifies fields needing more research.

    Source: §6 · p. 54

  • Takeaways & Limitations

    The survey concludes that further work is needed to explore evolving C-V2X technologies, including their newer versions.

    Source: §6 · p. 54

    The communication-type classification omits some possibilities because the authors found no literature examples, or because those types may not apply.

    Source: §3 · p. 19

Abstract

from arXiv · show

In recent years, the use of cellular network technologies to provide communication-based applications to vehicles has received considerable attention. 3GPP, the standardization body responsible for cellular networks specifications, is developing technologies to meet the requirements of vehicular communication applications, and the research community is testing and validating the ability of those technologies to implement different applications. This survey presents the body of work dealing with the use of cellular technologies to implement communication-based applications for the connected vehicle. We focus on basic and advanced road safety and traffic efficiency applications, which are critically important for the future of vehicular networks. We start by describing the different cellular-related technologies that have a role to play in providing services to the connected vehicle, propose a classification of types of communication used in vehicular applications, and then apply this classification to organize and present recent research work on the topic. Finally, we identify the main challenges in the use of cellular technologies to develop applications for the connected vehicle.

1. Introduction

Communication-based vehicle applications are motivated by safety and traffic-efficiency goals, while the expected commercial deployment of IEEE 802.11p-based systems has not materialized. This survey reviews C-V2X research in that context and distinguishes its focus from earlier surveys covering broader technologies, networking topics, or applications.

1. Introduction

  • Communication-based applications aim to protect vehicle occupants and other road users while improving fuel and road-infrastructure efficiency.The introduction groups connected-vehicle applications into road safety, traffic efficiency, and other applications.Source: §1 · pp. 2–3
  • Despite more than a decade of IEEE 802.11p research, open-road pilots, and spectrum allocation, its commercial deployment has not progressed as expected.Source: §1 · p. 2
  • 3GPP developed C-V2X, including direct vehicle communications from Release 14, as cellular technologies gained focus for connected-vehicle communications.The passage notes the FCC's decision to change the reserved 5.9 GHz spectrum use from DSRC to C-V2X.Source: §1 · p. 3
  • The survey reviews C-V2X communications for road-safety and traffic-efficiency applications, including other applications when studied alongside them.Source: §1 · pp. 3–4

1.1. Related work

  • Earlier surveys addressed topics ranging from LTE usability and access technologies to autonomous driving, communication standards, IP networking, and connected-vehicle applications.Several related surveys instead focused on 802.11p or specific C-V2X topics such as sidelink, resource allocation, security, or machine learning.Source: §1.1 · pp. 4–5
  • The survey comparison considers technology coverage, tutorial content, communication endpoints and messages, experimental methods, and communication services.Source: §1.1 · p. 5

1.2. Current Survey

The survey combines a C-V2X technology tutorial with a communication-type classification to organize application research and identify research and deployment challenges.

1.2. Current Survey

  • The survey examines how C-V2X technologies suit connected-vehicle applications, including their requirements, limitations, and configuration.Source: §1.2 · p. 7
  • Its tutorial covers C-V2X technologies, including 5G NR cellular and NR sidelink.Source: §1.2 · p. 7
  • The classification extends endpoint categories with message characteristics such as periodicity, event-driven exchange, RSU relaying, and multi-hop communication.It is applied to 28 specific applications, including cooperative collision warning, see-through, and collective perception.Source: §1.2 · p. 7
  • The survey reviews proposals for C-V2X communication services and associated messages required by connected-vehicle applications.Source: §1.2 · p. 7
  • Surveyed papers are categorized by evaluation approach, including focused and system-level simulations, real-equipment tests, field tests, and demonstrations.Source: §1.2 · p. 7
  • The survey identifies challenges for C-V2X application research and deployment.Source: §1.2 · p. 7

2. Background: C-V2X communication technologies

C-V2X technologies span cellular-network access and direct sidelink communication, with newer releases adding capabilities for vehicular use. The section also describes supporting network technologies for meeting application requirements.

2. Background: C-V2X communication technologies

  • Since Release 14, 3GPP has developed C-V2X technologies tailored to connected vehicles, building on earlier research into cellular alternatives to IEEE 802.11p.Source: §2 · pp. 8–9

2.1. LTE Uu and 5G NR Uu

  • 5G NR Uu adds frequency-band options and can provide higher throughput, lower latency, higher reliability, and improved spectrum efficiency; SC-PTM offers a more flexible, lower-latency broadcast allocation method used for V2X.Source: §2.1 · p. 9, §2.2 · p. 10

2.3. LTE PC5 and 5G NR PC5

  • Release 14 LTE PC5 mode 4 supports direct vehicle broadcast without base-station coverage because each vehicle autonomously allocates resources using S-SPS; semi-persistent scheduling can worsen collision bursts when vehicles repeatedly choose identical resources and periods.Source: §2.3 · pp. 11–12
  • NR PC5 mode 2 adds per-message resource allocation suited to event-driven traffic and improves resource selection through HARQ, re-evaluation, and preemption mechanisms.The described changes also provide more reservation-interval options and a 1 ms minimum latency.Source: §2.3 · pp. 12–13

2.4. Additional technologies

  • SDN, NFV, network slicing, and MEC provide network programmability, flexible function deployment, customized services, and edge-hosted applications for low-latency service.The section presents these technologies as requiring integration to meet vehicular application requirements.Source: §2.4 · pp. 13–14

3. Types of communication used in applications for the connected vehicle

The paper extends V2X communication categories beyond endpoint types to include message characteristics, then uses the resulting classification to organize applications and reviewed research. The surveyed literature is concentrated in V2V periodic messaging and V2I, with relatively little work on multi-hop or RSU-relayed communication.

Communication classification

  • The classification extends endpoint-based V2X categories to include message characteristics, such as periodic or event-driven messages, relaying, and multi-hop communication.It focuses on information transfer, one component of a connected-vehicle application.Source: §3 · pp. 15–16

V2V communication

  • V2V periodic messages provide recurring vehicle information, such as location and kinematics, that can support collision warnings and cooperative adaptive cruise control.Examples include ETSI CAMs and SAE BSMs; their transmission period can adjust to traffic or network conditions.Source: §3 · pp. 16–17

Communication types

  • The classification also covers event-driven and bidirectional V2V messages, multi-hop forwarding, V2P and V2I messaging, RSU relaying, and bidirectional V2N communication.Examples include hazard alerts, see-through video, pedestrian information, RSU warnings, and vehicle-to-server traffic updates.Source: §3 · pp. 17–18

Scope and literature coverage

  • The classification is not exhaustive: some endpoint and message-type combinations lack literature examples, either because they remain unexplored or may not apply.The survey catalogs application examples and reviewed papers by communication type.Source: §3 · p. 19

Research coverage

  • Most reviewed research addresses V2V periodic messaging, followed by V2I, while few papers examine multi-hop communication or RSU-relayed messages.Most of these works consider single-hop communication.Source: §3 · p. 19

4. An exploration of the use of C-V2X communications to support connected vehicle applications

The survey organizes reviewed C-V2X application research by communication type, technology, and validation approach. Its mapping shows broad interest in V2V periodic messages and V2I, but much of the evidence remains simulation-based.

Research patterns

  • Research interest in communication types is technology-agnostic: most works study V2V periodic messages and V2I event-driven messages.LTE Uu and sidelink modes are the most popular technologies in the survey mapping.Source: §4 · pp. 21–22

Survey organization

  • Table 3 maps reviewed papers across communication types, C-V2X technologies, and validation approaches, with simulation and field-test categories listed alongside technologies.Its rows represent communication types, while its columns include technologies and evaluation alternatives.Source: §4 · pp. 20–21, Table 3 · p. 22

Validation approaches

  • Most reviewed studies use simulations; relatively few use real-world equipment, although Release 14 LTE equipment has enabled several real-world evaluations.Many real-equipment evaluations use LTE Uu, particularly for V2I applications.Source: §4 · p. 23

4.1. V2V periodic messages

Research on V2V periodic messaging spans early cellular D2D approaches, comparisons with IEEE 802.11p, and Release 14 LTE PC5 mode 4 studies. Findings vary by traffic, distance, configuration, and evaluation setting, while real-equipment tests also demonstrate applications.

Earlier cellular approaches

  • Early studies of Release 12 LTE PC5 modes 1 and 2 examined resource-allocation strategies, while later comparisons found lower latency for D2D than LTE Uu infrastructure forwarding.Frequency-reuse strategies also showed potential benefits for LTE PC5 compared with IEEE 802.11p.Source: §4.1 · pp. 23–24

Release 14 LTE PC5 mode 4

  • Release 14 LTE PC5 mode 4 studies identify S-SPS weaknesses under high traffic, including persistent collisions, half-duplex errors, and hidden-terminal errors; proposed an alternative allocation algorithm.Other work reported poor performance in high-load scenarios and proposed feedback to address repeated resource-selection collisions.Source: §4.1 · p. 25

Comparison with IEEE 802.11p

  • In medium and high traffic, IEEE 802.11p outperformed Release 14 LTE PC5 mode 4 for packet reception ratio and inter-packet gap; it also had lower latency.The cited studies associate mode 4 limitations with shifting message periods, variable message sizes, and waiting for persistent scheduled resources.Source: §4.1 · p. 26
  • At distances up to 300m, Release 14 LTE PC5 mode 4 achieved performance similar to IEEE 802.11p when HARQ was used; at longer distances, mode 4 performed better.HARQ required double bandwidth in that comparison.Source: §4.1 · p. 26

Real-equipment evaluations

  • Real-equipment tests found better radio coverage and obstacle robustness for Release 14 LTE PC5 mode 4 than IEEE 802.11p.The SAE J2945/1 congestion-control algorithm also avoided large performance degradation as network load increased.Source: §4.1 · pp. 26–27
  • Test-track and open-road studies varied antenna placement and other configurations; realistic antenna placement yielded packet error rates as high as 16.65% below 100m.Another study reported minimum losses at ranges up to 1.2 km.Source: §4.1 · p. 27

4.2. V2V event-driven messages

Studies of V2V event-driven messaging examine cellular resource allocation, infrastructure forwarding, and application behavior. Findings include a benefit from short-term sensing in simulations, load-sensitive infrastructure performance, and demonstrations using real equipment.

Infrastructure forwarding

  • Base-station servers provided latency considered adequate for safety services, but performance degraded with network load; direct Release 14 LTE PC5 mode 3 performed better in a single-cell comparison [31].The work also examined event-driven alerts sent through infrastructure, including collision-avoidance warnings.Source: §4.2 · p. 29

Application studies

  • A cooperative lane-merging application achieved a success ratio above 90% when operating delay was below 5ms, with success degrading at larger delays.Delay could arise in communications or application information processing.Source: §4.2 · p. 29

Resource allocation

  • Release 14 LTE PC5 mode 4’s 1s sensing window was considered too long for asynchronous event-driven traffic, motivating investigation of alternative resource-allocation strategies.The standard’s original focus was V2V periodic messaging.Source: §4.2 · p. 29
  • Short-term sensing with a 100ms window performed better for event-driven messages than a 1s sensing window in system-level simulations.The studies found sensing important when event-driven traffic coexisted with periodic messages.Source: §4.2 · pp. 29–30

Real-equipment demonstrations

  • Real-equipment demonstrations of event-driven applications, including emergency electronic brake light and slow stationary vehicle warnings, reported no operational issues.Demonstrations involved two vehicles and a motorcycle under different conditions and speeds.Source: §4.2 · p. 30

4.3. V2V periodic messages and V2V event-driven messages: Platooning

Platooning research spans LTE and NR resource-allocation studies, equipment tests, and a 5G NR Uu framework that met the stated latency requirement. Results also expose constraints from congestion, network load, and application delay.

4.3. V2V periodic messages and V2V event-driven messages: Platooning

  • Initial Rel. 12 LTE PC5 mode 1 studies examined resource allocation for platoon communications, formation, and braking, sometimes using eMBMS for leader-to-member messages.Their simulations modeled only parts of the system or significantly simplified it.Source: §4.3 · pp. 31–32
  • Group scheduling assigns resources to a platoon through its leader, improving handover performance but requiring base-station coordination; high-load scenarios still need improvement.These works considered Rel. 14 LTE PC5 mode 3 and Rel. 16 5G NR mode 1.Source: §4.3 · p. 32
  • Rel. 14 LTE PC5 mode 3 outperformed mode 4 and enabled reduced inter-truck spacing in the studied comparison, which considered only a single cell.For mode 4, concentrated packet losses from half-duplex effects and collision bursts affected platooning despite adequate average packet delivery ratio.Source: §4.3 · p. 32
  • In a test-track trial with Rel. 14 LTE PC5 mode 4 equipment, packet error rate or information age degraded sharply without congestion control.The tested scenarios did not report specific supported inter-vehicle distances.Source: §4.3 · pp. 32–33
  • Below 10ms latency for periodic CAMs and event-driven messages met the 3GPP platooning requirement in a test-track framework using a low-latency slice and a virtualized 5G core at the edge.The framework combined NFV, SDN, MEC, and network slicing.Source: §4.3 · p. 33

4.4. V2V bi-directional communication

V2V bi-directional work covers see-through data sharing, while multi-hop C-V2X research remains limited. The surveyed studies include both practical performance constraints and proposed communication mechanisms.

4.4. V2V bi-directional communication

  • Few works explored bi-directional V2V because pre-Rel. 16 NR PC5 D2D technologies were not designed for vehicular point-to-point communication; see-through use cases need LIDAR or video from another vehicle.The early mmWave study proposed link allocation that considered both link properties and information value, but its simulations did not accurately model communications.Source: §4.4 · pp. 33–34
  • In a test-track cooperative-sensing trial, packet loss and computation delays made safe operation possible only at very low speeds; antenna alignment and obstacles strongly affected performance.The trial sent LIDAR data from one vehicle through a traffic-management server to another vehicle.Source: §4.4 · p. 34
  • Rel. 14 LTE PC5 mode 4 equipment supported only low-quality video in a see-through test, and that performance did not compete with other traffic.The test streamed video between vehicles.Source: §4.4 · pp. 34–35
  • Rel. 14 LTE PC5 mode 3 or Rel. 16 PC5 mode 1 could let direct vehicle links share spectrum with cellular links through network resource allocation, but the surveyed papers did not analyze this configuration at the application level.The cited studies proposed mode 3 resource-allocation algorithms.Source: §4.4 · p. 35

4.5. V2V multi-hop messages

  • Among the surveyed multi-hop proposals, used Rel. 14 LTE PC5 mode 3 and proposed cross-cluster forwarding by boundary vehicles or relaying through infrastructure for event-driven safety messages.Other cited work used IEEE 802.11p or generic D2D, or assumed perfect communications.Source: §4.5 · p. 36

4.6. V2P periodic messages

The surveyed V2P periodic-message studies use infrastructure and MEC for pedestrian protection, while V2I periodic studies include both Uu and PC5 implementations. Results range from Uu latency too large for safety applications to field-tested PC5 coverage and applications.

4.6. V2P periodic messages

  • V2P periodic-message studies used LTE Uu and MEC for pedestrian protection, including infrastructure-based warnings and offloading some collision processing to reduce device energy consumption and timing.The studies did not discuss scalability when many pedestrians and vehicles send messages to the infrastructure for redistribution.Source: §4.6 · p. 37

4.7. V2I periodic messages

  • In a real-equipment LTE Uu test, periodic V2I traffic-sign messages had latency too large for safety applications.The V2N design sent IP-based messages from a network server to particular vehicles, with addressing information obtained from vehicle messages.Source: §4.7 · pp. 37–38
  • With HARQ on, 2RX antennas, 20 dBm power, and 20 MHz bandwidth, Rel. 14 LTE PC5 mode 4 tests achieved 1400m line-of-sight coverage and at least 400m non-line-of-sight coverage.A bridge-mounted RSU test reported at least 550m coverage with packet reception ratios over 90%.Source: §4.7 · p. 38
  • A real-equipment V2I information application using Rel. 14 LTE PC5 mode 4 showed no issues in urban and motorway conditions.In tests with two vehicles and one RSU, PC5-based traffic-light and automated-driving integration had 30ms RSU-to-vehicle latency.Source: §4.7 · pp. 38–39

4.8. V2I event-driven messages

V2I event-driven applications use RSUs or network infrastructure to process local warnings and send them to vehicles. The surveyed work emphasizes edge processing for latency, while application tests and simulations report distinct performance conditions and approaches.

4.8. V2I event-driven messages

  • Vulnerable-road-user warning studies process vehicle and pedestrian CAMs or BSMs near the network edge to detect collision risks, using MEC at an RSU or base station.Evaluations included real equipment in a lab and system-level simulations.Source: §4.8 · p. 40
  • A German motorway pilot achieved latency lower than 20 ms for infrastructure-based collision warning or see-through using MEC.The approach uses MEC for low-latency processing and heavy application processing.Source: §4.8 · p. 41
  • A PC5 mode 4 vulnerable-road-user warning application used cameras and local vehicle collision detection, while a separate camera-based system varied transmission power and warning repetitions without studying complete-system latency.The real-equipment application reported no issues.Source: §4.8 · p. 41
  • A cooperative lane-merging simulation achieved success above 90% when operating delay was below 5ms, with reliability improved by more RSUs but delay little affected once there was at least one RSU per 2 km [107].The V2V approach performed better when vehicles had enough processing capacity.Source: §4.8 · p. 42
  • Reports about risky vehicles can trigger infrastructure actions such as traffic-light changes or warnings, using D2D and cellular Uu exchanges.Source: §4.8 · p. 42
  • A Rel. 14 LTE PC5 mode 4 roadworks-warning demonstration reported no application issues, while infrastructure-to-vehicle road-condition warnings were also tested over pre-5G NR Uu [103].Source: §4.8 · p. 43

4.9. RSU-relayed messages

The survey describes RSU-relayed and bidirectional V2N communications, including applications that send vehicle data to network servers and retrieve traffic information. Evaluations include LTE throughput and network slicing, while other work explores 5G and mmWave applications.

4.9. RSU-relayed messages

  • Cluster-based relaying sends periodic and event-driven V2V messages to infrastructure over Uu, making it V2N and less efficient than RSUs receiving messages directly over D2D.The direct-RSU approach can relay periodic vehicle information to servers for information gathering or collision-risk detection.Source: §4.9 · pp. 43–44
  • RSUs also relay traffic-violation reports and vehicle movement data for traffic-control broadcasts and shockwave-detection processing.Processing can use MEC near the network edge for shorter latency, with cloud computation for operations needing a network-wide view.Source: §4.9 · pp. 44–45

4.10. V2N bi-directional communications

  • LTE Uu media downloading achieved clearly better throughput than IEEE 802.11p, but performance degraded significantly above 90 km/h.The comparison used real equipment on a test track.Source: §4.10 · p. 45
  • Network slicing in a motorway field test offered different bandwidth and latency to users, and unused assigned bandwidth could be used by other slices.The results indicate that slicing can guarantee resources for one application without hindering efficient resource use.Source: §4.10 · pp. 45–46
  • Other V2N studies send information to traffic servers and return incident notifications or path-planning data.One simulated 5G NR Uu system used MEC to process cooperative-perception messages and support downloading local dynamic maps; another tested mmWave and SDN with one vehicle and two RSUs.Source: §4.10 · p. 46

5. Challenges for C-V2X applications research and deployment

The survey finds that C-V2X applications have been explored with simulations and real equipment, but identifies gaps in deployment, evaluation, and coverage of communication types and newer technologies. It highlights challenges spanning resource management, advanced applications, interoperability, experimentation, and deployment models.

Survey findings

  • C-V2X feasibility has been demonstrated across application studies, but periodic V2V is the most studied communication type, while periodic V2P and V2V multi-hop receive little attention.The survey also reports PC5 performance degradation as network load increases, identified S-SPS weaknesses, and few evaluations of NR technologies; MEC latency reduction in V2I and V2N services has been confirmed in different works.Source: §5 · pp. 46–48

Experimentation

  • C-V2X research still has relatively few reported pilot deployments, and proprietary or custom simulation tools make results harder to compare and replicate.Commercial LTE PC5 mode 4 equipment was becoming available, but could cost an order of magnitude more than IEEE 802.11p equipment.Source: §5 · pp. 48, 52

Congestion control and distributed resource allocation

  • Congestion control must handle unpredictable network loads, and its interaction with semi-persistent resource scheduling and repeated HARQ transmissions requires further investigation.The survey also calls for resource-allocation algorithms optimized for advanced applications’ requirements, including links between vehicles and the network.Source: §5 · pp. 49–50

Geobroadcast

  • Geobroadcast forwarding over D2D PC5 needs careful analysis because S-SPS timing granularity and delay may affect CBF, while infrastructure relaying offers another coverage path [141].Message relaying through infrastructure is under study for 3GPP Release 17.Source: §5 · pp. 50–51

Advanced applications

  • More work is needed to explore recent C-V2X solutions and related technologies for advanced applications such as see-through, remote driving, and massive sensor sharing.The survey names 5G NR PC5 modes, NR Uu, MEC, SDN, network virtualization, and network slicing as technologies to explore.Source: §5 · p. 51

Multi-operator scenarios and hybrid solutions

  • Multi-operator safety services require interoperability and coordination across operators and cells, while combining IEEE 802.11p with C-V2X interfaces remains an open issue.Spectrum management is particularly challenging in multi-operator settings; the roles of LTE PC5, NR PC5, and NR Uu in service provision also remain open.Source: §5 · pp. 51, 53

Business models

  • Business models remain a deployment challenge: B2C faces business and consumer concerns, while B2B may offer opportunities that require further investigation, particularly for D2D.Examples include cloud-resource use and multi-operator scenarios for B2C, and an “as a Service” model for B2B.Source: §5 · p. 53

6. Conclusion

The survey reviews C-V2X technologies and classifies communication types used for connected-vehicle applications. It organizes research on their suitability, limitations, and configuration, and identifies areas needing further work for real-world adoption.

6. Conclusion

  • The article reviews literature on C-V2X technologies for road-safety and traffic-efficiency applications in connected vehicles.It considers the technologies’ appropriateness for application requirements, their limitations, and how they can be used and configured.Source: §6 · pp. 53–54
  • The survey proposes a classification of communication types needed to support connected-vehicle applications and uses it to organize research on C-V2X technologies.The reviewed work analyzes and validates technologies specified by 3GPP and explores limitations and improvements that have contributed to specification development.Source: §6 · p. 54
  • Further work is needed on recent C-V2X versions, advanced services, combining technologies, multi-operator interoperability, and business and deployment models.The conclusion specifically notes 5G NR PC5 or NR Uu combined with technologies such as MEC.Source: §6 · p. 54
Loading 2403.17851v1…