The next generation of automobiles integrate a multitude of electronic control units (ECUs) to implement various automotive control and infotainment applications. However, recent works have demonstrated that these pervasively computerized modern automobiles are susceptible to security attacks that could compromise the physical safety of the driver and/or passengers. In this paper, we propose a novel ECU architecture for automotive cyber-physical systems (CPS) that simultaneously integrates both security and dependability primitives in the design with negligible performance, energy, and resources overhead. We implement our proposed ECU architecture on Xilinx Automotive (XA) Spartan-6 FPGA. We demonstrate the effectiveness of our proposed architecture using a steer-by-wire (SBW) application over controller area network (CAN) with flexible data rate (CAN FD) as a case study. We also optimize and implement a prior secure and dependable automotive work on NXP quad-core iMX6Q SABRE automotive board. We quantify the performance, energy, and error resilience of our proposed architecture for the SBW case study. Results reveal that our proposed architecture can attain a speedup of 47.9× while consuming 2.4× lesser energy than the optimized SABRE board implementation of security and dependability primitives. We further perform a comparative analysis of prior designs and the proposed ECU architecture for different in-vehicle networks, viz., CAN, CAN FD, and FlexRay. Results verify the feasibility as well as the superiority of the proposed ECU over other prior designs in terms of response time, energy efficiency, and error resilience.

中文翻译：

---

用于安全可靠的汽车 CPS 的可重构 ECU 架构的设计和评估

下一代汽车集成了大量电子控制单元 (ECU)，以实现各种汽车控制和信息娱乐应用。然而，最近的工作表明，这些普遍计算机化的现代汽车容易受到可能危及驾驶员和/或乘客人身安全的安全攻击。在本文中，我们为汽车网络物理系统 (CPS) 提出了一种新颖的 ECU 架构，该架构在设计中同时集成了安全性和可靠性原语，而性能、能源和资源开销可忽略不计。我们在 Xilinx 汽车 (XA) Spartan-6 FPGA 上实现了我们提出的 ECU 架构。我们使用具有灵活数据速率 (CAN FD) 的控制器局域网 (CAN) 上的线控转向 (SBW) 应用程序作为案例研究来证明我们提出的架构的有效性。我们还在 NXP 四核 iMX6Q SABRE 汽车板上优化和实施了先前安全可靠的汽车工作。我们量化了我们为 SBW 案例研究提出的架构的性能、能量和错误恢复能力。结果表明，与安全性和可靠性原语的优化 SABRE 板实现相比，我们提出的架构可以实现 47.9 倍的加速，同时消耗的能量少 2.4 倍。我们进一步对不同车载网络（即 CAN、CAN FD 和 FlexRay）的先前设计和提议的 ECU 架构进行比较分析。