CAN-XL (Controller Area Network eXtra Long) and Automotive Ethernet represent two different approaches to in-vehicle networking. CAN-XL is the upcoming third-generation CAN bus protocol, evolving the classical CAN/CAN-FD architecture to support higher data rates (on the order of tens of Mbit/s) and larger message payloads while retaining CAN’s low-cost, real-time control advantages. Automotive Ethernet, on the other hand, refers to Ethernet-based networks adapted for cars, with data rates from 100 Mbit/s up to multi-gigabits and a scalable switched network topology. Ethernet in cars enables the high bandwidth and IP-based communication needed for modern features like high-resolution cameras, advanced driver assistance systems (ADAS), and connected infotainment. In essence, CAN-XL extends the CAN bus concept into moderately higher speeds, whereas Automotive Ethernet brings IT network speeds and protocols into the vehicle. Both will likely coexist in future vehicles, each serving areas best suited to its strengths. Below, we compare their technical differences, implementation aspects, and regional adoption trends in the automotive industry.
Troubleshooting Automotive Computer Systems: Automotive Computers, Sensors & Network
This book provides a thorough and practical guide to diagnosing issues related to no-network conditions, no-start conditions, and loss of communication between sensors, control modules, and the PCM. It explains diagnostic procedures for automotive computer systems in a clear and structured manner.
The content is presented in an accessible way, making it understandable not only for professional mechanics but also for technically inclined readers without formal training. Common diagnostic trouble codes such as P0300, P0171, P0172, and P0174 are addressed in a straightforward and easy-to-follow format.
Drawing on extensive experience in engine performance, electrical systems, and automotive electronics, the author offers practical insight that can help readers build confidence in diagnosing and resolving vehicle computer and engine performance issues. More information…
Technical Differences and Implementation
Key technical and implementation differences between CAN-XL and Automotive Ethernet include:
-
Data Rate & Bandwidth: CAN-XL operates up to about 10 Mbit/s (with CAN SIC XL transceivers supporting up to ~10 Mbit/s on typical cabling, and up to 20 Mbit/s in ideal conditions). This is a major jump from classical CAN’s 1 Mbit/s, but it remains an order of magnitude lower than Ethernet speeds. Automotive Ethernet links typically run at 100 Mbit/s (100BASE-T1), 1 Gbit/s, or higher, with new standards reaching 2.5/5/10 Gbit/s and even 10 Gbps on the horizon. In raw bandwidth, CAN-XL’s maximum (~10 Mbps) is far below the multi-megabit or gigabit throughput of Ethernet. This means Ethernet can carry high-resolution sensor data, video streams, and large data transfers that CAN-XL cannot. However, CAN-XL’s boost into the low tens of Mbps aims to cover many mid-speed needs (e.g. large sensor readings, faster ECU reprogramming) that were previously out of reach for CAN.
-
Payload Capacity: CAN-XL dramatically increases frame payload size. It supports up to 2048 bytes of data in a single frame, compared to 8 bytes in classical CAN and 64 bytes in CAN-FD. This extra-long payload (hence “XL”) allows CAN-XL messages to carry large chunks of data that would require many frames in CAN/CAN-FD. For example, a single CAN-XL frame can encapsulate what used to be 32 CAN-FD frames, or even an entire standard Ethernet packet (~1500 bytes) in one go. By contrast, standard Ethernet (with a typical MTU) handles roughly 1500 bytes per frame, so in terms of per-frame payload the two are of the same order (Ethernet can be extended with jumbo frames if needed). The benefit is that CAN-XL can transmit large data (like a diagnostic message or sensor snapshot) more efficiently than CAN-FD, with less protocol overhead per byte. However, sending a full 2 KB CAN-XL frame at 10 Mbps takes ~2 ms on the bus. This is still relatively fast, but longer than typical short CAN frames, so CAN-XL includes a frame preemption mechanism – a high-priority message can interrupt a large low-priority frame to keep critical control latency low. Ethernet frames, being shorter in duration at high speed, can deliver large data quickly, though Ethernet’s efficiency shines when continuously streaming large volumes (e.g. video) that would overwhelm CAN-XL’s lower bandwidth.
-
Latency & Determinism: CAN-XL retains CAN’s deterministic arbitration mechanism. It uses a priority-based Carrier-Sense Multiple Access with Collision Resolution (CSMA/CR) on a shared bus. Every frame carries an 11-bit priority field (separate from a larger ID field in CAN-XL), and if two nodes transmit at once, the node with higher priority (lower ID) wins arbitration without destroying the message. This means high-priority CAN-XL messages get through with microsecond-level latency once the bus is free, guaranteeing a bounded worst-case latency for critical signals. Ethernet, by default, is not deterministic – it uses asynchronous packet switching, and in standard Ethernet, collisions or queueing can introduce variable delay. However, modern Automotive Ethernet can be made deterministic via Time-Sensitive Networking (TSN) extensions. Standards like IEEE 802.1Qav/Qbv (traffic shaping and scheduled traffic) and 802.3br (frame preemption) allow Ethernet networks to ensure high-priority, time-critical traffic is delivered with bounded latency, similar in spirit to CAN’s behavior. For instance, Ethernet’s frame preemption lets a switch interrupt a large low-priority Ethernet frame to send a critical frame immediately, mimicking CAN’s ability to prioritize urgent data. In practice, achieving strict real-time performance on Ethernet requires more complex configuration (time schedules, sync protocols like gPTP, etc.), whereas CAN-XL’s prioritization is simple and inherent. In summary, CAN-XL has built-in low latency for brief control messages, while Automotive Ethernet can be engineered for real-time but with added network design complexity. For most control applications with short messages, CAN-XL offers very predictable timing. Ethernet shines when high-bandwidth data must be moved with reasonable (and configurable) latency across the car, especially when using TSN for tasks like sensor fusion where synchronization is critical.
-
Network Topology: CAN-XL continues CAN’s use of a multi-drop bus topology. All nodes connect to a shared two-wire twisted pair bus (with terminators at the ends), and any node can transmit to all others. This broadcast nature means any CAN-XL message is seen by all nodes (useful for one-to-many communications like a sensor value broadcast). The wiring is simple and typically runs in a daisy-chain or linear bus fashion. In contrast, Automotive Ethernet employs a point-to-point link topology. Each Ethernet connection is a dedicated link (often a single twisted pair for 100BASE-T1 and above) between an ECU and an Ethernet switch (or between switches). Multiple Ethernet ECUs form a network via switches, arranged in star, tree, or ring topologies depending on the architecture. There is also a special 10BASE-T1S Ethernet variant that supports multi-drop (bus topology) at 10 Mbps, functioning somewhat like a competitor to CAN-XL’s bus while still using Ethernet protocols. Generally, however, most automotive Ethernet links are switched point-to-point. The topology difference means that CAN-XL is inherently a one-bus-for-many nodes solution, easy to wire for local domains, whereas Ethernet requires switches or hubs to connect multiple nodes, introducing additional hardware. Ethernet’s switched network can isolate traffic (reducing busload on individual links) and scale to many nodes with high aggregate bandwidth, which is ideal for a backbone. CAN-XL’s bus has simpler connectivity but all nodes share the bus bandwidth. This typically limits CAN(-XL) to on the order of 30–50 nodes per bus for performance reasons, whereas Ethernet networks can accommodate dozens of nodes per switch and multiple switches interconnected for larger systems.
-
Implementation Complexity & Cost: A major reason CAN (and by extension CAN-XL) remains attractive is its low cost and simplicity. A CAN or CAN-XL node only requires a microcontroller with an integrated CAN controller and a CAN transceiver chip – components that are inexpensive and readily available. The wiring is a single unshielded twisted pair, and no active network infrastructure (like a switch) is needed for a CAN bus. This simplicity scales well for adding devices: adding another CAN-XL ECU is relatively cheap and straightforward. Automotive Ethernet, although much cheaper now than in the past, still involves higher complexity and cost per node. Each Ethernet-enabled ECU needs a PHY transceiver (and often magnetics or coupling circuits), and connecting multiple ECUs demands an Ethernet switch or multiple switch ports in a central gateway. The processors in Ethernet ECUs also tend to be more powerful (and costly) because they often run a full TCP/IP stack or at least some Ethernet frame handling and protocol software (e.g. SOME/IP for service discovery). All this means Ethernet hardware and development can be more expensive, and testing is more involved (Ethernet PHY compliance tests, etc., unlike the relatively plug-and-play nature of CAN). Automotive Ethernet cabling has been optimized (using single-pair unshielded cables to save weight and cost), but it is still point-to-point – so a complex system might need more total cable length and connectors than an equivalent CAN bus network. CAN-XL inherits CAN’s cost advantages: it runs on the same two-wire bus (with slightly improved transceivers for higher speed) and doesn’t necessarily require any new infrastructure beyond what CAN buses use. For a subsystem that only needs, say, ~10 Mbps aggregate throughput, using CAN-XL can be far more cost-efficient than equipping every device in that domain with Ethernet capability and a switch port. In summary, Ethernet is justified for data-heavy domains despite higher cost, whereas CAN-XL targets a “sweet spot” where moderate bandwidth is needed without the overhead of a full Ethernet network.
-
Protocol Overhead and Integration: CAN-XL and Ethernet also differ in how they integrate with higher-level protocols. CAN-XL remains a message-based protocol with minimal overhead – messages are identified by IDs (now 32-bit IDs in CAN-XL for content, plus an 11-bit priority) and broadcast to all, which is excellent for low-level control and diagnostics where any node can easily listen to any message. Ethernet is typically used with the TCP/IP stack or service-oriented middleware, meaning a lot more overhead (MAC addresses, IP headers, etc.) but it seamlessly supports complex communications: IP networking, streaming, and internet connectivity. In practice, automotive software frameworks (like AUTOSAR) are evolving to use service-oriented communication over Ethernet (e.g. using SOME/IP and RESTful APIs) for the flexibility of a “software-defined vehicle”. CAN-XL is adapting by introducing a protocol type field (similar to an EtherType) in its frame, so it can carry and distinguish higher-layer protocols or even encapsulate Ethernet frames. This makes CAN-XL more interoperable with Ethernet/IP-based data – for instance, a gateway can transfer an Ethernet payload via CAN-XL if needed. Security is another consideration: Ethernet benefits from mature IT security protocols (TLS, IPsec, etc.), whereas CAN-XL is just beginning to incorporate security at the data link layer (the emerging CANsec standard for frame encryption/authentication). Overall, Ethernet aligns with the broader IT ecosystem, enabling easier integration with cloud services, OTA updates, and V2X communication, while CAN-XL is an evolution in the traditional automotive ecosystem, aiming to boost capacity without abandoning CAN’s simplicity and reliability.
Use Case Orientation: In practice, CAN-XL and Automotive Ethernet serve complementary roles in a vehicle. CAN-XL is well-suited for real-time control networks that need up to tens of Mbps – for example, powertrain, chassis, body control, or sensor/actuator networks that are currently on CAN/CAN-FD can eventually be upgraded to CAN-XL to carry more data (such as high-rate sensor feedback, larger diagnostics, or firmware updates) without losing the simple bus architecture. Its large frames and moderate speed can also be useful for over-the-air (OTA) updates on smaller ECUs or as a backup channel – e.g. downloading firmware over CAN-XL is much faster than over classic CAN, though still slower than Ethernet. Automotive Ethernet, by contrast, is chosen as the backbone network for data-intensive domains: high-resolution camera and radar feeds, centralized sensor fusion computers, high-bandwidth infotainment systems, and fast telemetry/telematics links all lean on Ethernet’s high throughput. It’s also key for the emerging zonal architectures and “software-defined vehicle” concept, where a lot of data is routed across the car and new software features can be deployed requiring an IP/Ethernet environment. In short, Ethernet enables what CAN (even CAN-XL) cannot, and CAN-XL aims to cover the gap between CAN-FD and Ethernet. We can expect most cars to use a mix: an Ethernet network (or several) for heavy data and central gateway connectivity, and CAN-XL (along with CAN-FD/LIN in smaller roles) for distributed control where its efficiency and real-time behavior are advantageous. Industry consensus is that Ethernet will take over as the vehicle backbone, but CAN/CAN-XL will remain for “peripheral” networks where it is more cost-effective.
Automotive Ethernet
Explore the latest developments in Automotive Ethernet technology and implementation with this fully revised third edition. Featuring approximately 20% new material and expanded technical depth, the book now includes detailed coverage of emerging PHY technologies such as 10BASE-T1S (including multidrop operation) and 2.5, 5, and 10GBASE-T1. It also provides in-depth discussion of EMC interference models and the latest Time-Sensitive Networking (TSN) standards tailored for automotive applications.
The edition further addresses security concepts, power-saving strategies within Automotive Ethernet systems, and functional safety considerations in automotive environments. Readers will also find an overview of test methodologies and key lessons learned from real-world deployments.
Industry pioneers share both technical and strategic insights that have shaped the success of Automotive Ethernet. Topics range from electromagnetic requirements and physical layer technologies to Quality of Service (QoS), VLANs, IP-based communication, service discovery, network architecture, and system validation.
This book serves as a comprehensive guide for engineers, technical managers, and researchers involved in the design of in-vehicle electronic systems, as well as for decision-makers evaluating the strategic introduction of new networking technologies. More information…
Regional Adoption Trends
The adoption of CAN-XL versus Automotive Ethernet is evolving differently across regions. Broadly, Automotive Ethernet has global momentum – virtually all major OEMs worldwide joined the OPEN Alliance to standardize one-pair Ethernet for vehicles in the 2010s – whereas CAN-XL’s uptake is just beginning and is currently driven mostly by European (especially German) manufacturers. Below is a regional overview of how each technology is being favored:
Europe (EU)
European automakers, particularly in Germany, are leading the push for CAN-XL. Germany is the birthplace of CAN, and German companies (OEMs and suppliers) have been heavily involved in CAN-XL development from the start. For instance, Volkswagen (VW) initiated the CAN-XL project through the CAN in Automation consortium, seeking a CAN successor that would meet higher bandwidth needs while preserving CAN’s benefits. Volkswagen has explicitly included CAN-XL in its next-generation E/E architecture plans: in a 2021 conference, VW outlined a zonal architecture using CAN-FD and CAN-XL to replace legacy CAN buses, with Ethernet as the backbone for ultra-high data devices. In this model, high-speed domains like infotainment and ADAS sensors feed into an Ethernet network, but zonal controllers and traditional domains (powertrain, body, etc.) would communicate via CAN-FD or CAN-XL, depending on bandwidth needs. This strategy shortens and localizes CAN segments (reducing cable weight) and uses CAN-XL where payloads or data rate demand it, while still relying on Ethernet to carry heavy data centrally. VW’s backing is significant, suggesting its mid-to-late 2020s models (especially higher-end or electric platforms) could be among the first to feature CAN-XL in production. Other German automakers are also interested: Audi/Porsche (part of VW group) will likely follow VW’s lead, and companies like BMW and Mercedes-Benz(Daimler) – which had pioneered prior bus technologies like FlexRay – are now focusing on Ethernet for their high-speed needs but are evaluating CAN-XL as a supplement for mid-speed networks currently on CAN-FD. In Germany’s supplier industry, Bosch (which co-developed the CAN-XL spec and first silicon) and toolmakers like Vector have prepared a full CAN-XL ecosystem (IP cores, transceivers, software stacks, test tools) to support OEM adoption. This strong ecosystem and the AUTOSAR software support for CAN-XL (influenced by German OEMs) mean European manufacturers have the pieces in place to roll out CAN-XL in the near future. At the same time, Automotive Ethernet is already well-established in Europe. Many European cars (especially luxury brands) have used Ethernet for certain systems for years (e.g. BMW introduced 100 Mbps Ethernet in a 2013 model for camera data). Europe as a region has been at the forefront of automotive Ethernet adoption, in part due to these early deployments and initiatives like the OPEN Alliance (whose founding members included BMW, VW, and Volvo among others). In summary, European OEMs are embracing a hybrid approach: Ethernet will be the backbone for data-intensive functions, while CAN-XL is poised to be adopted for the next generation of zonal and control networks as an evolutionary step from CAN-FD. German automakers are expected to launch the first CAN-XL implementations (likely in high-end vehicles by the late 2020s), setting a precedent that others in Europe (and globally) will watch closely.
United States (USA)
In the U.S., Automotive Ethernet is gaining broad acceptance, whereas CAN-XL is being approached cautiously. The traditional Big Three American OEMs (GM, Ford, Stellantis) have not announced any CAN-XL deployments as of 2025. Instead, their focus has been on migrating from classical CAN to CAN-FD and on integrating Ethernet for the newest features. For example, General Motors (GM) started using CAN-FD around 2019 in some platforms (to speed up ECU reflashing and to support ADAS sensors), and it has invested heavily in Automotive Ethernet for backbone networks, especially to support its Ultra Cruise and autonomous driving developments. Likewise, Ford uses CAN-FD for many modules and Ethernet for ADAS and infotainment systems. These OEMs appear to view CAN-XL as an “optional” upgrade that they might enable later if needed – since upcoming microcontrollers from their suppliers will likely include CAN-XL capability by default, they can adopt it down the line via software if a clear use case emerges. For now, U.S. automakers seem more intent on simplifying architectures and leveraging proven technologies (consolidating networks with CAN-FD and Ethernet) rather than introducing a brand-new CAN variant immediately.
The standout case in the U.S. is Tesla, which is taking a markedly different path. Tesla initially used CAN buses like any other manufacturer, but found even CAN-FD (up to 8 Mbit/s) inadequate for the data-heavy, high-speed needs of their EV and Autopilot systems. In recent years Tesla has been developing a next-gen in-vehicle network that essentially bypasses CAN altogether. According to their public patents and reports, Tesla is implementing a time-division multiplexed network (TDMA-based) over a 48V power architecture, likely leveraging Automotive Ethernet physical layers or even power-line communication on the same wires. This centralized, scheduled network will allocate time slots to each node for communication (akin to a deterministic Ethernet/TSN system), eliminating collisions and guaranteeing latency for critical data. The upcoming Tesla Cybertruck and future models are expected to use this new architecture, effectively treating Ethernet (or a similar high-speed medium) as the single network for both control and high-data traffic. Tesla has thus indicated it will not adopt CAN-XL – it is “leapfrogging” directly to an Ethernet-based approach with >100 Mbit/s capacity and microsecond-scale timing precision, considering any intermediate step (like CAN-XL’s 10 Mbps) too slow. This approach is influencing others; some tech-focused EV startups and projects in the US may also opt to skip CAN-XL and go straight to robust Ethernet networks, using CAN/CAN-FD only in minimal roles for legacy compatibility. In summary, no mainstream US-built vehicles use CAN-XL yet (as of 2025). American OEMs continue to rely on classical CAN/CAN-FD widely (due to cost-effectiveness and familiarity) and deploy Ethernet for new high-bandwidth needs. The U.S. industry’s stance is somewhat skeptical on CAN-XL’s necessity: if a major network overhaul is needed for future vehicles, many would rather invest in Ethernet or other advanced network concepts directly, instead of adding a new CAN flavor that still falls short of Ethernet’s capabilities.
Japan & South Korea
In Japan and South Korea, the preference leans toward established solutions (CAN-FD and Ethernet), with a conservative approach to adopting CAN-XL so far. Japanese automakers (like Toyota, Honda, Nissan) and Korean automakers (Hyundai/Kia) tend to introduce new network technologies cautiously, often after they are proven elsewhere, due to the priority on reliability and the cost of re-tooling. As of 2025, there have been no public announcements from major Japanese or Korean OEMs about CAN-XL adoption, and it appears CAN-XL is not yet a focal point in these regions’ production vehicles. Instead, manufacturers have been incrementally upgrading from classical CAN to CAN-FD for higher data rates in certain domains, and adopting Automotive Ethernet for the cutting-edge infotainment and ADAS systems in premium models. For instance, Japanese OEMs are members of the OPEN Alliance and have recognized the need for Ethernet – Toyota joined the Alliance early on, and Hyundai Motor was even a founding member, reflecting their commitment to Ethernet as the future backbone. In practice, luxury models or advanced EVs from Japan/Korea have begun using Ethernet (for example, some Nissan and Toyota models use Ethernet-based camera systems, and Hyundai’s latest EV platform includes Ethernet for its driver-assist sensors, according to industry teardowns). Meanwhile, CAN-FD is used for powertrain and other control networks where more bandwidth than classical CAN is needed. CAN-XL’s appeal in these regions will likely depend on global validation. Japanese and Korean automakers often wait to see a technology mature and gain ecosystem support before mass adoption. Now that CAN-XL is standardized (ISO 11898-1/-2 updated in 2024) and supported by major silicon vendors (including Japan’s own Renesas, which has CAN-XL capable microcontrollers), the foundation is there. We can expect Japan and Korea to evaluate CAN-XL through joint projects or Tier-1 suppliers, especially if their European or American partners start using it. However, given that even CAN-FD rollout took time (most Japanese/Korean cars today still heavily use classical CAN), CAN-XL may take several years to appear. When it does, it will likely be in specific domains where it adds value without needing the full jump to Ethernet – for example, body electronics or zone controllers that need ~10 Mbps could use CAN-XL instead of adding an Ethernet switch port. In summary, Japanese and Korean automakers currently favor a combination of CAN-FD and Automotive Ethernet, and while they are not leading the CAN-XL charge, they are positioned to adopt it later on if it proves beneficial and cost-effective. Regional industry inertia plays a role: engineers in these countries have decades of experience with CAN and LIN, and any new protocol undergoes rigorous reliability vetting. As one analysis noted, across the global industry the transition from classical CAN to CAN-FD (and eventually CAN-XL) is slow – as of 2024 the vast majority of vehicles still use Classical CAN, with CAN-FD in a minority and CAN-XL in virtually 0% (pre-production only). Japan and Korea exemplify this cautious transition, ensuring that new network tech like CAN-XL will be adopted only after it’s clearly proven in real-world use by early adopters.
China
China’s automotive sector is extremely dynamic and is watching both technologies closely. Chinese automakers (including EV giants like BYD, NIO, Xpeng, Geely, etc.) historically built their vehicles with multiple CAN buses similar to Western designs. In recent years, as these companies push into advanced EVs and connectivity, they have started integrating CAN-FD and Ethernet as well – for instance, CAN-FD is often used for battery management systems and critical control in Chinese EVs, while Automotive Ethernet is introduced for high-bandwidth components like infotainment units or ADAS domain controllers. CAN-XL is not yet deployed in Chinese production vehicles (understandable, since it’s newly standardized), but there is growing interest. Chinese tech forums and local suppliers have begun discussing CAN-XL as part of next-gen vehicle network solutions, often alongside Automotive Ethernet in the same conversation. Notably, the IEEE 10BASE-T1S (10 Mbps single-pair multi-drop Ethernet) is highlighted in China as a potential alternative to CAN buses for low-cost domains. Some see 10BASE-T1S as “a direct competitor of CAN-XL” because it offers a similar 10 Mbps bus-like network but with native Ethernet compatibility. Chinese automakers are very focused on connectivity and the concept of the software-defined vehicle (which aligns naturally with Ethernet/IP architectures). This could incline them to favor Ethernet-based solutions (including 10BASE-T1S for lower-speed networks) rather than champion a new CAN variant on their own. However, cost and supply chain considerations are critical in China’s high-volume market. If CAN-XL can be delivered cheaply by local silicon suppliers and if it simplifies certain network designs, Chinese OEMs will use it pragmatically. Essentially, China tends to move quickly once a technology is proven elsewhere. If Volkswagen and others roll out CAN-XL successfully in production and demonstrate clear benefits (like reduced weight from fewer cable runs, or cheaper implementation for mid-speed zones), Chinese manufacturers are likely to follow suit in short order. We are already seeing Chinese OEMs redesigning their E/E architectures in line with global trends (for example, adopting zonal architectures similar to those proposed in Europe/U.S.). In those, they will be deciding whether to stick with CAN-FD or move some networks to CAN-XL under the Ethernet backbone. Additionally, China’s commercial vehiclesector (buses, trucks) widely uses CAN (often CAN-based J1939 protocol). As these vehicles become more connected (telematics, driver assist systems for commercial fleets), the higher bandwidth of CAN-XL could be valuable to transmit more data without jumping to Ethernet in cost-sensitive heavy vehicles. In summary, Chinese automakers are taking a pragmatic, wait-and-see approach: they are open to CAN-XL but not committed to it until its advantages are proven, and in the meantime they continue to deploy Automotive Ethernet for the high-data needs of their rapidly evolving EV platforms. Given the fast pace of the Chinese market, if and when CAN-XL gains global traction, China’s adoption could ramp up very quickly, potentially even outpacing other regions in volume once the technology is deemed ready and cost-effective for mass production.
Networking Vehicles to Everything: Evolving Automotive Solutions
Communication between vehicles and infrastructure is poised to transform the way traffic, pedestrians, and the surrounding environment are managed, fundamentally reshaping the future of mobility. Networking Vehicles to Everything provides a comprehensive, state-of-the-art resource for readers seeking a deep understanding of the technologies converging to enable connected and automated transportation.
This in-depth volume examines technical challenges, practical case considerations, standardization activities, product implementation strategies, and emerging trends such as software reconfiguration, millimeter-wave (mmWave) communication, and advanced control theory methods. It presents a balanced view of the evolving ecosystem that underpins vehicle-to-everything (V2X) systems.
Readers will gain insight into the key organizations and regulatory bodies driving development and governance in this field, as well as the ongoing technological competition—most notably between IEEE 802.11p and 3GPP LTE-V2X solutions—within a rapidly expanding, multi-billion-dollar market. The book also explores open technical questions, market dynamics, application domains, and the future possibilities for connected systems across all vehicle types. More information…
Conclusion
CAN-XL and Automotive Ethernet address different needs in automotive networks, and both will play roles in future vehicles. Technically, CAN-XL fills the niche between traditional CAN and Ethernet – it offers an order-of-magnitude more bandwidth than CAN-FD (up to 10–20 Mbps) and much larger frame sizes, all while preserving CAN’s deterministic low-latency bus arbitration and low implementation cost. This makes it ideal for mid-speed, real-time control domains that are reaching the limits of CAN-FD. Automotive Ethernet vastly exceeds CAN-XL in raw speed (scaling to 100 Mbps, 1 Gbps, 10 Gbps and beyond), and with proper configuration it can meet real-time requirements, which is why it is becoming the backbone for high-data and high-performance systems. Implementing CAN-XL is relatively straightforward as an evolution of CAN (it can often use existing wiring with upgraded transceivers, and new controllers support CAN/CAN-FD/CAN-XL triple modes). In contrast, implementing Automotive Ethernet requires a more complex network design (point-to-point links, switches, and a shift to IP-based software), but it unlocks capabilities that are indispensable for modern connected and autonomous vehicles.
In terms of industry adoption, Europe (especially Germany) is championing CAN-XL’s introduction – with VW and others planning to integrate it alongside Ethernet in upcoming models – whereas North American OEMs are largely skipping directly to Ethernet-centric architectures, using CAN-FD for legacy needs and showing hesitance to add another CAN variant. Tesla’s approach in the US exemplifies the Ethernet-first strategy, eliminating traditional CAN where possible in favor of a unified high-speed network. In Japan and South Korea, automakers are gradually adopting Ethernet as needed but remain conservative about CAN-XL until it’s a proven path; they will likely follow the global consensus once it forms, rather than lead it. And in China, the rapidly growing EV industry is watching both technologies: they are quick to implement Ethernet for advanced features and will adopt CAN-XL if it proves to offer clear cost or integration advantages in their pursuit of scalable, software-defined vehicle platforms. Overall, we anticipate a coexistence of CAN-XL and Automotive Ethernet in future vehicles – CAN-XL extending the life and capability of the CAN ecosystem for efficient, real-time control in many subsystems, and Ethernet serving as the high-speed backbone for data-intensive systems and centralized computing. The balance of these technologies in any given region or manufacturer’s designs will depend on their legacy investments, supplier ecosystem, and strategy for meeting the ever-increasing performance demands of tomorrow’s cars.
Sources: The comparisons and trends above are based on recent technical analyses and industry reports, which reflect the current state (as of 2025) of CAN-XL development and the adoption of automotive Ethernet across different automakers and regions. These sources include academic-style reviews, industry conference data, and statements from automakers and consortia, providing a comprehensive view of how CAN-XL and automotive Ethernet stack up and where each is headed in the global automotive landscape.
Comments are closed.