Vehicle-road-cloud Integration and C-V2X Industry Research Report, 2025

Vehicle-side C-V2X Application Scenarios: Transition from R16 to R17, Providing a Communication Base for High-level Autonomous Driving, with the C-V2X On-board Explosion Period Approaching

In 2024, the C-V2X pre-installations of passenger cars in China was approximately 500,000 units, with an assembly rate of 2.21%. It is expected that by 2028, the installations will exceed 2 million units, and the installation rate will exceed 8%. Once the pre-installed penetration rate of C-V2X exceeds 10%, the industry will enter a mature stage.

The main driving force lies in the significant expansion of application scenarios of R17 protocol. The R17 C-V2X module achieves three breakthroughs of ""global connection - collaborative perception - safe and controllable"", providing a communication base for high-level autonomous driving. With key technologies such as RedCap, NTN satellite networking, and MBS broadcasting entering the mass production stage in 2025, the Internet of Vehicles will enter a new stage of ""global intelligence"".

Key Technologies and application scenarios of R17 protocol in autonomous vehicles:

Vulnerable Road User Protection (V2P): R17 focuses on optimizing the reliability and delay of the direct communication link (PC5 interface), supporting vulnerable road users such as pedestrians and non-motorized vehicles to share location information in real-time through low-power terminals (such as wearable devices). Vehicles can receive collision warnings within 3ms, significantly improving safety in high-risk areas such as urban intersections and school surroundings. It is applicable to scenarios such as early warning for pedestrians suddenly crossing the road and blind spot prompts for non-motorized vehicles.

High-precision Environmental Perception Sharing: MBS adopts the point-to-multipoint (PTM) transmission mode. The Road Side Unit (RSU) distributes the same point cloud/video data to multiple vehicles simultaneously through shared air interface resources, saving 70% of air interface resources compared with the unicast mode and making up for the blind spot of single-vehicle perception. It enhances the vehicle's ability to identify obstacles and construction areas, with end-to-end data transmission < 20ms and warning response time ≤ 50ms.

Group Safety Information Broadcasting: MBS supports broadcast/multicast modes. The Road Side Unit (RSU) can uniformly push emergency accident, bad weather warning, or traffic control information to regional vehicles, saving 70% of air interface resources compared with unicast, alleviating network congestion in dense scenarios such as crossroads and highway ramps, and applicable to scenarios such as accident warning in tunnels and coordinated scheduling of traffic lights.

Emergency Communication in Remote Areas: Integrating Non-Terrestrial Network (NTN) satellite communication technology, vehicles in areas without ground base station coverage (such as forests and deserts) can send distress signals or receive navigation information through low-orbit satellite links, supporting emergency rescue and path planning, and filling the coverage blind area of traditional cellular networks. It is applicable to scenarios such as off-road adventure vehicles getting out of trouble and seeking help, and long-distance freight vehicles monitoring in remote sections.

Navigation in Complex Urban Environments: R17 improves the positioning accuracy to centimeter level (integrating GNSS + RTK). Combined with multipath/NLOS suppression algorithms, it improves the positioning reliability in satellite signal blind areas such as tunnels and under viaducts. Vehicles share real-time positions through V2V to dynamically optimize detour paths in congested sections. It is applicable to complex scenarios such as automatic parking in underground parking lots and precise guidance at urban viaduct ramps.

RedCap Terminal Support: Reducing the number of antennas (1-2) and bandwidth requirements (20MHz Sub-6GHz), the module cost is reduced, and power consumption is reduced by 20%-30%.

C-V2X modules have evolved from the initial R14 LTE-V2X to R16 5G NR C-V2X. R17 modules have been released one after another in 2024 and will be mass-produced and installed in vehicles in 2025.

R16 Protocol:

5G NR C-V2X has achieved large-scale mass production and on-board installation in 2024. For example, ZTE's 5G R16 ZM9300 module is equipped with self-developed S1V 5G chip, supporting 5G and C-V2X dual-mode communication. It is the first domestic chip supporting Cellular Vehicle-to-Everything (C-V2X) and was installed in models of SAIC, GAC, FAW, etc. in 2024.

R17 Protocol:

Favalon 5G Redcap automotive-grade module AN931: Built based on the Qualcomm SA510M platform, it integrates cost, performance, module hardware compatibility, and software platform architecture. It supports the latest 5G 3GPP R17, supports 5G SA (Standalone) mode, and is downward compatible with LTE and NR - FR1 networks.

Quectel's automotive-grade 5G RedCap module AG53xC series: Built based on the Qualcomm SA510M platform, it supports the 3GPP R17 standard and performs excellently in cost-performance balance, hardware compatibility, and software architecture, bringing more efficient and economical solutions to the in-vehicle communication field. At present, this series of modules has entered the mass production stage and is expected to support mass shipments for many automotive customers within the year.

MeiG Smart MA922 series modules:

Support 3GPP Release 17 standard 5G communication, support 5G NR Standalone (SA) and Non-Standalone (NSA) networks, and are downward compatible with 4G/3G/2G networks, and can be compatible with the frequency band requirements of major countries and regions around the world.
Support C-V2X function, use the globally unified ITS 5.9GHz frequency band to deploy V2X applications, and support PC5/Uu communication modes.
Support dual-frequency GNSS + RTK high-precision positioning, which can provide high-precision position information for V2X applications to ensure accurate interaction between vehicles and other traffic participants, thereby improving driving safety and traffic efficiency.
The module has a built-in ECDSA hardware acceleration engine, supporting 6000 signature verifications per second.
It integrates a CPU processor with 20K DMIPS computing power, eliminating the need for a separate external application processor for developing V2X applications, which greatly optimizes and improves the cost.
Based on the 3GPP R17 protocol, it supports 5G NB-NTN (Narrowband Non-Terrestrial Network) satellite communication technology, supports cellular communication through satellites, ensures global communication connections, and provides service continuity and service availability anywhere.

Road-side Infrastructure: Gradually Transitioning to 6G Based on 5G-A, Realizing ""Integration of Communication, Perception, and Computing"", and Gradually Enabling Road-side Data to Be Transmitted to Vehicles

CVIS relies on road-side perception devices (including various traffic sensors such as cameras, lidar, and radar) to collect original information of traffic targets (including 2D video images and 3D point clouds, etc.), which is then sent to road-side edge computing devices for analysis and calculation (including target detection and target classification) as well as perception fusion, generating structured data to represent the attributes of traffic targets (such as vehicle speed and heading, type and influence range of traffic events, etc.).

The structured data on the road side is further processed into V2X messages, specifically I2V messages. These I2V messages are sent by the RSU (Road Side Unit) via the PC5 wireless air interface, or by 5G/4G base stations via Uu wireless air interface to road traffic participants including motor vehicles and pedestrians.

In the field of road-side perception for autonomous driving, commonly used sensors include cameras, lidars, radars, and radar-vision integrated machines.

Baidu Auto Cloud 3.0: vehicle-road-cloud coordination capabilities include functions such as integrating traffic management data and optimizing path planning. Efficient computing power and data management technology provide underlying support for the intelligent upgrading of new energy vehicles. Facing the extremely high computing power requirements of current mainstream end-to-end simulation, Baidu Auto Cloud 3.0's heterogeneous computing power platform has the advantages of large computing power reserve and efficient operation, supporting domestic chips such as Kunlunxin P800.

With the support of large computing power, breaking through barriers of CVIS, Baidu Auto Cloud 3.0 is empowering the development of end-to-end autonomous driving.

ZTE, together with China Mobile and Huawei, released the ""5G Integration of Communication, Perception, and Computing"" Internet of Vehicles architecture: It has the advantages and highlights of connecting to vehicles, providing computing power to the edge, and enabling perception to the network. This architecture has three highlights: unified air interface, integration of communication and perception, and fusion of communication and computing:

For communication connection: Migrate the originally scattered PC5 network to the 5G network, uniformly carry V2X vehicle-road information, realize wide-area network connection at a lower cost, achieve high-reliability connection guarantee based on 5G QoS and slicing, further improve network performance, reduce construction costs, and speed up deployment.

For perception: Replace road-side perception devices such as millimeter-wave radar with communication-perception integrated base stations, which have wireless communication-perception integration capabilities, provide full-process and whole-network wireless perception computing, and realize multiple functions with one network through air interface resource sharing, further improving perception performance.

For computing power: Including two levels of computing power, cloud and wireless edge, to realize V2X cloud-edge collaboration. The first-level computing power realizes wide-area management and control, and the second-level computing power realizes integration of communication and computing with the base station, supporting real-time service sinking and intelligent data offloading, which can realize low-latency edge computing and local precise push.

Please Note: PDF E-mail from Publisher purchase option allows up to 10 users and does not allow printing or editing. This functionality will require a Global Site License. Show more