As automotive networks transition to Optical Ethernet, ensuring link reliability over the lifetime of the vehicle becomes just as important as achieving higher bandwidth.
1. What are the most common failure mechanisms in automotive optical links?
The most common failure mechanisms in automotive optical links are similar to those affecting electrical harnesses: mishandling, crashes, manufacturing defects and aging. However, automotive optical links are inherently robust, and many of these events are less likely to result in communication failures than in conventional electrical wiring.
Automotive optical fibers tolerate smaller bend radii, are highly resistant to bending-induced attenuation, and exhibit excellent resistance to repetitive bending fatigue. They also show high tolerance to cable compression and crushing, making them well suited to the automotive environment.
In extreme cases, damage to the cable harness may lead to optical fiber pinching or localized fiber damage. Excessive bending attenuation effects, mitigated by the bend-insensitive features present in current automotive optical harnesses, may lead to similar effects.
Faulty or contaminated connectors resulting from manufacturing defects or mishandling may increase channel losses and lead to intermittent link unavailability.
Sustained bending stress and aging of both the cable harness and the transceiver may progressively degrade channel performance throughout the vehicle lifetime.
2. How can runtime monitoring detect degradation before functional failure occurs?
Runtime monitoring can detect degradation before functional failure occurs using two complementary mechanisms. Qualitative link-state indicators signal different levels of availability on both optical fibers, while reliable link-margin monitoring based on FEC correction capability allows progressive channel degradation to be detected before communication is affected. This monitoring remains available even in the absence of payload traffic, ensuring a Mean Time To Packet Acceptance (MTTPA ) of 1010 years.
3. What data is most useful for predictive maintenance models?
The combination of Input Average Optical Power and link margin provides a useful indication of channel health and degradation over time. In addition, bidirectional latching-low link-state monitoring flags detect intermittent events such as microcuts that may otherwise go unnoticed during normal operation.
4. How can OEMs balance monitoring granularity with system complexity?
OEMs can balance monitoring granularity and system complexity by relying on monitoring functions integrated in BASE-AU Automotive Optical Ethernet. The PHY can autonomously monitor link status using user-configurable thresholds and interrupt generation, reducing host software complexity while providing timely diagnostics. When additional visibility is required, the Operations, Administration and Maintenance (OAM) channel provides acknowledged communication between link partners. This acknowledgment mechanism confirms both reception by the remote PHY and access by the remote host. Because OAM information is carried within the 802.3cz PHD header, monitoring data can be exchanged without consuming application payload bandwidth.
Don’t miss Luis and Juan’s presentation ‘Improving Predictive Maintenance in ADAS Sensor Networks through Runtime Optical Link Monitoring‘ at AutoSens Europe this year!
Interested in In-Cabin monitoring technology?
With a pass to AutoSens Europe, you’ll also get full access to our co-located sister event, InCabin. Take a look at the full agenda for InCabin here >>
