Building a Safe, Secure and Easy to Use Internet-of-Things Infrastructure
Title: Building a Safe, Secure and Easy to Use Internet-of-Things Infrastructure
Abstract: The first wave of computing was large main frames, the second the PC, and the third the smartphone and mobile computing. We are arguably living in the fourth wave of computing led by billions of connected IoT devices, connected to each and every facet of our life, powered by rapid advances in machine learning, and hardware/software systems. However, there are numerous challenges that still hinder the widespread realization of this vision of pervasive IoT – namely security, safety, privacy concerns and those related to scalability and efficacy of ML systems to make sense of the sheer volume and diversity of IoT data.
In this talk, I will present our recent work on building a safe, secure and easy to use IoT infrastructure from the ground up. I will first present GioTTO – an end-to-end IoT stack that we have been building for a few years now, which includes our Mites.io hardware platform coupled with our distributed IoT OS and Machine Learning layers to enable intelligent inferences from IoT sensors without having to write (much) code. I will also describe numerous safety and security challenges common with current IoT ecosystems and our recent approaches towards addressing these issues for existing platforms and for clean slate designs. Finally, I will discuss the numerous privacy issues with these emerging IoT devices and deployments and highlight our multi-faceted approaches to address them by involving various stakeholders — developers, manufacturers and end-users. I will summarize by highlighting challenges and opportunities ahead including briefly mentioning some ongoing work.
Bio: Yuvraj Agarwal is an Associate Professor of Computer Science (and by courtesy in ECE) at Carnegie Mellon, and the director of SynergyLabs @ CMU (www.synergylabs.org). His lab focuses on research at the intersection of hardware (embedded) and software systems, and in the recent past his group has focused on energy efficiency, security & privacy as a central research theme. In additional to publications at top conferences (MobiSys, Ubicomp, SenSys, IPSN, NSDI, USENIX, BuildSys, CHI), SynergyLab specifically strives to have real world impact and benefit society. His group leads research around smart buildings, systems for the Internet-of-Things, privacy and security, and have also had significant impact in mobile computing and energy efficient computing. Many of the systems that Yuvraj and his students have built have been deployed at scale on the UCSD and CMU campuses across several buildings. The systems that he has built for studying privacy on mobile platforms have been used by hundreds of thousands of users worldwide and have had significant industry adoption. He is a member of the IEEE, ACM and USENIX.