TL;DR
This paper introduces a comprehensive methodology for designing automotive Time-Sensitive Network (TSN) systems, focusing on meeting strict real-time communication requirements. They developed a systematic simulation system using OMNeT++ to evaluate various TSN mechanisms and applied it to a realistic automotive scenario, demonstrating TSN's ability to guarantee performance.
What Problem Does This Paper Address?
Modern vehicles are becoming increasingly complex, with a growing demand for reliable, low-latency, and deterministic communication networks. Traditional automotive networks struggle to meet these demands, especially with the integration of advanced features like autonomous driving and sophisticated infotainment systems. Time-Sensitive Networking (TSN), an extension of Ethernet, offers a promising solution for guaranteeing real-time performance. However, existing research often focuses on individual TSN mechanisms in isolation, lacking a holistic design approach and a unified simulation tool to evaluate complex, mixed-mechanism scenarios relevant to the automotive industry. This paper aims to bridge this gap by providing a structured methodology and a robust simulation framework for designing and verifying automotive TSN systems.
Methods
The authors propose a design methodology that prioritizes meeting traffic delay constraints as the primary goal. This involves:
- Traffic Characterization and Mapping: Analyzing different automotive traffic types (e.g., safety-critical signals, sensor data, audio/video) and mapping them to specific priorities and traffic classes based on their criticality and timing requirements. This is visualized in Table 4, which links traffic classes to automotive scenarios and priorities.
- Holistic Design Approach: Considering the entire design lifecycle, from requirements analysis and E/E architecture design to detailed programming, simulation, and testing.
- Systematic Simulation System: Developing a comprehensive TSN simulation system built upon OMNeT++ and the INET framework. This system integrates various TSN standards and mechanisms, including Time-Aware Shaper (TAS), Frame Preemption, and redundancy mechanisms. The framework is designed to be extensible and capable of simulating mixed-mechanism scenarios.
- Scenario-Based Verification: Applying the developed methodology and simulation system to a complex, real-world automotive scenario provided by a major motor company. This scenario involves a zonal E/E architecture and diverse traffic types.
Datasets
The paper utilizes a realistic automotive scenario provided by a major motor company in Shanghai. This scenario includes:
- Zonal E/E Architecture: A hybrid topology with end systems, zonal controllers, domain controllers, backbone networks, and remote controllers.
- Traffic Information: Detailed information on various traffic types, their transmission paths, and characteristics, as presented in Table 5 and Table 6. This serves as the benchmark for evaluating the TSN system's performance.
Key Insights and Ideas
- Prioritizing Delay Constraints: The core design philosophy is to treat meeting the strict delay requirements of different traffic types as the paramount goal for automotive TSN systems.
- Holistic Design is Crucial: Effective automotive TSN design requires a comprehensive approach that considers E/E architecture, software, and traffic characteristics, rather than focusing on isolated mechanisms.
- Unified Simulation is Essential: A systematic simulation platform is vital for evaluating the performance of mixed TSN mechanisms and complex automotive scenarios, which is often lacking in current research.
- Trade-offs Exist: Implementing mechanisms like Time-Aware Shaper (TAS) significantly improves the real-time performance of high-priority traffic but can slightly degrade the performance of lower-priority traffic. This highlights the need for careful balancing.
- Redundancy for Reliability: While redundancy mechanisms have a minimal impact on end-to-end delay under bandwidth limitations, they are crucial for enhancing reliability by ensuring data delivery even in the event of failures.
Key Results
The study successfully demonstrated that their proposed methodology and simulation system can effectively design and verify automotive TSN networks. Specifically:
- Meeting Timing Constraints: In the analyzed complex automotive scenario, applying the Time-Aware Shaper (TAS) and redundancy mechanisms allowed all traffic types to meet their defined end-to-end delay requirements.
- Jitter Reduction: The jitter for critical control traffic and ADAS traffic was reduced to 0 µs after applying the scheduling and redundancy mechanisms.
- Performance Improvement for High-Priority Traffic: The maximum end-to-end delay and jitter for higher-priority traffic were significantly reduced, confirming the effectiveness of TAS in enhancing time certainty.
- Acceptable Impact on Low-Priority Traffic: While lower-priority traffic experienced a slight increase in average delay and jitter, this increase was deemed acceptable.
Limitations and Future Work
- Limitations:
- The paper acknowledges that the complexity of the analyzed scenario is not the absolute highest, as it uses fewer than eight traffic classes.
- While the simulation system is comprehensive, the paper notes that some TSN standards might not be fully implemented or might have minor inconsistencies with the protocols (e.g., PSFP in CoRE4INET).
- Future Work:
- The authors plan to analyze the performance of the Asynchronous Traffic Shaper (ATS) defined in IEEE 802.1Qcr and investigate its suitability for different in-vehicle scenarios.
- They also aim to optimize TSN scheduling algorithms using intelligent algorithms like machine learning.
- Further exploration of how different TSN scheduling mechanisms can work together is also a future direction.
Summary updated