Doing More with Less Wiring: Mission-Critical and Intelligent Communication Protocols for Future Vehicles Using Power Lines

随着汽车领域自动化任务的兴起，车载通信的发展越来越重要，并受到新的应用。尽管有线和无线通信已经在很大程度上用于支持各种应用，但大多数具有关键任务性质的车载应用，如刹车和发动机控制，仍然更喜欢专用的有线网络，以实现可靠和安全的传输。数据布线的关键挑战之一是促进越来越多的设备（如传感器和电子控制单元（ECU））的互连，有效地创建一个响应延迟低、可靠性高、复杂性低的车载网络。这些电线的空间要求、重量和安装成本可能会变得非常大，尤其是在未来的汽车中，这些汽车都是高度复杂的电子系统。鉴于车辆部件、传感器和ecu已经连接到电线上，我们在本提案中将最近用于车载物理层通信的车载电力线应用于车载网络。以质量空气流量传感器为例，它有一根电源线和两根信号线，用电力线通信代替目前的信号线将会高效，因此可以减少66%的布线。车载电力线通信（VPLC）的进步可以为车载网络提供一个非常低复杂性和自由的平台，这是满足日益增长的应用需求的理想选择，特别是未来的车辆。然而，新兴的VPLC受到缺乏协议支持的限制，这对在实践中部署它并确保关键任务通信构成了重大挑战。下面的例子说明了这个提议的动机。举个例子来说明动机：未来的车辆配备了先进的驾驶员辅助系统（ADAS），可以连接多个传感器和ecu，以提供安全监测和控制。这个场景的一个重要要求是，作为源的系统应该与所有ecu或网络目的地具有稳定的连接。同样重要的是，此类车载网络必须保证新兴控制服务的超低延迟，因为任何一秒钟的延迟都可能导致致命事故。因此，有效的协议设计对于VPLC支持具有关键任务和高带宽需求的未来应用至关重要。该项目的目的是提高网络的可靠性，并保证车载电力线通信中车载应用的严格关键任务要求。我们将与汽车专家合作，共同构建该项目，以开发创新的智能车载通信协议。该提案寻求的解决方案是双重的。一是通过充分探索实践和理论分析、通道/流量模式的动态性和自学习技术，追求智能接入和拥塞控制解决方案的新设计，这为提案提供了理论方面的支持。第二步是实践方面，提出的电力线方法需要能够与现有的最先进的解决方案共存和协作，并通过实际的车载交通数据来验证其性能。显然，这两者是不可分割的，不仅因为车载网络可靠通信的最终目标只有在这两部分都完成的情况下才有可能实现，而且因为这两部分之间的交互是有效系统设计的关键。

With the emerging automated tasks in vehicle domain, the development of in-vehicle communications is increasingly important and subjected to new applications. Although both wired and wireless communications have been largely used for supporting diverse applications, most of in-vehicle applications with mission-critical nature, such as brake and engine controls, still prefer dedicated wired networks for reliable and secure transmission. One of the key challenges for data wiring is to facilitate the interconnectivity of increasing devices, e.g., sensors and electronic control units (ECU), effectively creating an in-vehicle network with low response latency, improved reliability and less complexity. The space requirement, weight, and installation costs for these wires can become significant, especially in future vehicles, which are highly sophisticated electronic systems. Given that vehicle components, sensors and ECUs are already connected to power wires, we apply vehicle power lines, which have recently been utilized for in-vehicle communications at the physical layer, to in-vehicle networks in this proposal. Taking mass air flow sensor as an example, it has one power wire and two signal wires, it will be efficient to use power line communications to replace the current signal wires, so 66% of wiring can be reduced. The advancement of vehicular power line communications (VPLC) can provide a very low complexity and free platform for in-vehicle networks, which is ideal for the increasing demand of applications in particular with future vehicles. However, the emerging VPLC is constrained by lack of protocol support, which pose significant challenges to deploy it in practise and ensure mission-critical communications. The following example illustrates the motivation of this proposal.An example for the motivation: A future vehicle is equipped with advanced driver assistance systems (ADAS) which can be connected with multiple sensors and ECUs to provide safety monitoring and control. An important demand of this scenario is that the systems, viewed as sources, should have stable connections with all ECUs, or network destinations. And it is also important that such in-vehicle networks must guarantee ultra-low latency for emerging control services since any seconds of delay may cause fatal accident. Therefore, an effective protocol design is crucial for VPLC to support future applications with mission-critical and high-bandwidth demands. The aim of the project is to improve the reliability of the network and guarantee stringent mission-critical requirements of in-vehicle applications in vehicular power line communications. We will partner with automotive specialists and construct the project to develop innovative and intelligent in-vehicle communication protocols. The solution this proposal is seeking is two fold. One is to pursue new design of intelligent access and congestion control solutions by fully exploring the practical and theoretical analysis, dynamic nature of channels/traffic patterns and self-learning techniques, which provides the theoretic aspect of the proposal. Then, the second step is from the practical aspect, where the proposed power line method shall be able to coexist and cooperate with existing state-of-the-art solutions, and its performance will be validated by practical in-vehicle traffic data. Obviously the two are inseparable not just because the ultimate goal of reliable communication for in-vehicle networks is only possible with the accomplishment of the both two parts, but also because the interaction between the two parts is the key for effective system design.