SANE: an Integrated Systematic Framework for Automotive In-Vehicle Network Emulation

Recently, in-vehicle networks have increasingly become one of the most important components in vehicles, es-pecially electric automobiles. Considering the high complexity and stringent reliability requirements of automotive software and hardware, it is highly desirable to test new applications in an emulated environment before deploying them in real vehicles. This approach is advantageous for Original Equipment Manufacturers (OEMs), which can help them identify hidden problems and reduce costs. However, due to the heterogeneity of in-vehicle electronic control units (ECUs) and the diversity of protocols, the complexity of in-vehicle networks has increased rapidly. Consequently, the emulated environment for any single ECU can hardly meet all requirements. In this paper, we propose a vehicle-level emulation framework for in-vehicle networks, Systematic Automotive Network Emulation (SANE), which integrates functionalities from the operating system layer to the network application layers, and is specifically designed for emulating and evaluating in-vehicle networked soft-ware services. Recognizing the complexity of in-vehicle network environments, we emphasize the scalability of the framework, ensuring it is easy to deploy and modify on a large scale. We present and analyze a series of experimental results with different configurations to identify the impacts of various factors, such as in-vehicle Electrical/Electronic (E/E) architecture, on network performance and the trade-offs between different Quality of Service (QoS) strategies, while also confirming the framework's feasibility. To the best of our knowledge, our work is a pioneering effort to comprehensively consider the entirety of vehicle systems within an emulated network framework. SANE is the first of its kind to offer a detailed and adaptable emulation environment, specifically designed for in-vehicle networks. Through evaluations across various configurations, we also systematically quantify the influence of different in-vehicle Electrical/Electronic (E/E) architectures on network and service performance.