CNS Core: Medium: Detroit – A New End-to-end System for Practical and Accessible IoT

University Of California-Santa Barbara, Santa Barbara CA

Investigators

Abstract

The Internet of Things (IoT) is a rapidly emerging set of technologies in which ordinary objects are equipped with digital intelligence -- the ability to sense, analyze, actuate, and control their environment automatically. By linking our physical and digital worlds, IoT devices make it possible to perform advanced data analysis in place -- to enhance decision-making by humans, to detect, diagnose, and remediate problems without human intervention, and to automate operations throughout our economy (e.g., in health care, military, agriculture, technology, financial, transportation, energy sectors, among others). This project focuses on how to securely program, deploy, and manage IoT systems and applications. Specifically, the project will develop a portable, multi-tier (sensors, edge, cloud) platform, called Detroit, that supports "write-once-run-anywhere" programming for IoT devices. The platform will enable low-latency data analytics and response in or near the sensing tier, while leveraging emerging public cloud computing technologies at the edge and cloud tiers. High-level language support and platform services (for monitoring, security policy enforcement, and deployment management) will be used to hide the complexities of IoT, give developers tools with which they can specify and automatically check security and deployment properties, and automate placement, end-to-end authentication, and repair/update for IoT applications. The project will enable secure IoT innovation to become broadly practiced rather than solely the domain of distributed and embedded systems experts. Such democratization is key to realizing the positive societal transformation and shared economic impact that the IoT promises is possible. In addition, the project will use real-world IoT testbeds and applications, from several other scientific domains (e.g., smart homes, digital agriculture, and ecology), to evaluate and demonstrate the advances and to ensure that they are useful to developers and researchers across disciplines. Finally, the project will develop, collect, and maintain source code, datasets, and results related to this effort that will be shared publicly via the GitHub repository https://github.com/MAYHEM-Lab/Detroit. These artifacts will include the Detroit platform, benchmarks, IoT applications, and sensor data that can be used to drive, compare, and evaluate IoT systems research advances. The project will maintain these artifacts for the duration of the project and two years following. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.