Computer Science Colloquia
Xiaofan (Fred) Jiang, co-Founder and CTO of Air Scientific Inc.
Friday, March 27, 2015
10:45 AM, Rice Hall, Rm. 242
HOST: Jack Stankovic
Data, Knowledge, and Action - Bringing Together Intenet-of-Things with Physical Analytics
With the rapid advances in mobile and wearable systems, we are increasingly connected to both cyberspace and to our physical surroundings. And with analytics in the cloud, our environment is also becoming smarter and more aware of us. By bringing together Internet-of-Things with physical data analytics, CPS research can have a profound impact on our environment, improving our quality of life, and help creating a smarter and sustainable future.
In this talk, I'll highlight three projects - building energy, precise indoor-location, and air quality monitoring, all of which start with sensing of some physical phenomenon, followed by data analytics in the cloud to derive knowledge, and finally into actions that complete the feedback loop in the physical world.
I'll first present the pervasive air-quality monitoring (PAM) project. PAM employs a network of hybrid sensors with data analytics in the cloud for accurate high-density air-quality monitoring at low-cost. At the front-end, two types of Internet-connected particulate matter (e.g. PM2.5 and PM10) monitors are created with carefully designed mechanical structures for optimal airflow. On the cloud-side, an artificial neural network model and a Gaussian process based prediction module are used to calibrate sensors in real-time, and to analyze and predict PM2.5 concentrations. We also provide an open platform and a set of web services for third-party developers to innovate on top of. In the indoor-location space, I'll present LiveSynergy - a platform that brings together physical sensors, actuators, and virtual services in the cloud to enable humans to easily program and interact with their physical environment. An important component of LiveSynergy is a novel magnetic-based wireless proximity detection platform, with accuracy and consistency better than existing solutions such as BLE beaconing, WiFi fingerprinting, ZigBee, and long-range RFID. Building on top of this platform, LiveSynergy provides cloud-based APIs that enables a range of indoor applications such as location-based targeted advertising, geo-fencing applications, and presence detection. Finally, I'll present the ACme project, which combines data from custom-designed wireless IPv6 plug-load meters with various data feeds to create an energy model of the building. A light-weight data protocol is developed that enables us to uniformly represent different types of monitors and actuators, and provide consistent interfaces to access their data, configuration, historical profiles, and meta-data. Using this sensing and networking fabric, various data analysis can be applied to derive models of building loads at the leaves of the "load-tree", which allows us to answer questions such as what is the breakdown of energy consumption by types of loads, where are energy hotspots in the building, or who is using the most energy.
Bio: Dr. Fred Jiang is the co-Founder and CTO of Air Scientific Inc. Before founding Air Scientific, Dr. Jiang was Senior Staff Researcher and Director of Analytics and IoT Research Laboratory at Intel Labs China. He received his B.S., M.S., and Ph.D. from UC Berkeley, advised by Professor David Culler. His primary research interests include cyber physical systems, mobile and embedded systems, connected health and fitness, and building energy.
Dr. Jiang has published extensively in top international conferences and workshops. His ACme platform has been widely adopted by universities and industries, including Lawrence Berkeley National Laboratory, Stanford University, National Taiwan University, and several startups. His project on wearable and mobile fitness, in collaboration with University of Virginia, was featured on New Scientist and the Economist magazine. His air-quality monitoring project has been featured on China Central Television and People's Daily, and was successfully incubated into a startup. He is actively serving on several technical and organizing committees including ACM SenSys, ACM/IEEE IPSN, and ACM MobiSys. He was also a National Science Foundation (NSF) Graduate Fellow. Before joining Intel, Dr. Jiang has worked in Google, Xilinx, Arch Rock (now part of Cisco), People Power, and Microsoft Research Asia.