CRI: Planning - A Testbed for Compiler-supported Scalable Error Monitoring and Diagnosis for Reliable and Secure Sensor Networks

这是一项计划拨款，重点是传感器网络的嵌入式中间件开发工具，该工具基于普渡大学该团队最近开发的研究原型。对于这项规划拨款，项目团队建议通过购买更大的传感器网络试验台来验证软件开发工具的运行时错误监控和诊断，以扩展他们现有的传感器能力。此外，该项目还将实现室内循环系统二氧化碳监测的应用案例研究。传感器节点在恶劣条件下部署时，通常极易受到硬件故障的影响。由于网络嵌入式系统的自组织和动态特性，其通信协议往往是复杂的，并且经常容易出错。此外，这些网络还面临以下问题：组件和通信链路暴露给潜在的对手，因此受到节点捕获、拒绝服务和恶意代码注入等安全威胁；受限的资源，如存储、带宽、计算能力和能量；以及，即使它们可能被正确设计，网络协议也可能由于编程错误而不正确地实现。该项目的目标是允许网络嵌入式系统(NES)的广泛研究社区使用该项目提出的健壮编程工具来进行运行时错误监控和诊断。该工具将针对这样的问题，即传感器网络的许多组件中的任何一个都可能发生错误，并且需要快速有效地检测这些错误。

This is a planning grant that focuses on an embedded middleware development tool for sensor networks that is based on a research prototype recently developed by this team at Purdue University.For this planning grant the project team proposes to expand their existing sensor capability by purchasing a larger sensor network test-bed to validate software development tools for run-time error monitoring and diagnosis. In addition, the project will enable an application case study for carbon dioxide monitoring for indoor circulation systems.Sensor nodes typically are highly vulnerable to hardware breakdowns when deployed in harsh conditions. Because of their ad hoc and dynamic nature, the communication protocols of networked embedded systems tend to be complex and frequently error-prone. In addition, these networks experience: components and communication links that are exposed to potential adversaries and hence are under security threats such as node capture, denial of service, and malicious code injection; constrained resources such as storage, bandwidth, computing power and energy; and, even though they may be correctly designed, network protocols may be implemented incorrectly due to programming errors.The goal of this project is to permit the broad research community of Networked Embedded Systems (NES) to use the robust programming tool proposed with this project for run-time error monitoring and diagnosis. The tool will target the problem that errors can occur in any of the many components of a sensor network and those errors need to be detected quickly and effectively.