Project OPERA

Project OPERA is a multi-institutional EPSRC-funded research project

Project OPERA

OPERA will investigate a new unobtrusive sensing听technology for CONTEXUAL SENSING - defined as听concurrent physical activity recognition and indoor听localisation - to facilitate new applications in indoor听monitoring and Ambient Assisted Living (AAL). The听OPERA platform will be integrated into the "SPHERE听long term behavioural sensing machine" to gather听information alongside various other sensors around听the home听so as to听monitor and track the signature听movements of people.鈥嬏�

Technology and research

The OPERA system will be built around passive sensing听technology: a receiver-only radar network that detects the听reflections of ambient radio-frequency signals from people - in听this case, principally, the WiFi signals in residential听environments. These opportunistic signals are transmitted from听common household WiFi access points, but also other wireless听enabled devices which are becoming part of the Internet of听Things (IoT) home ecosystem.鈥�
鈥�
The project will make use of cutting-edge hardware听synchronisation techniques, and recent advances in direction听finding techniques to enable accurate device-free (non-cooperative) localisation of people. It will also employ the听latest ideas in micro-Doppler radar signal processing, bio-mechanical modelling and machine/deep learning for听automatic recognition of both everyday activities e.g. tidying听and washing-up, to events which require urgent attention such听as falling. OPERA is expected to overcome some of the key听barriers associated with the state-of-the-art contextual sensing听technologies. Most notably non-compliance with wearable听devices, especially amongst the elderly, and the invasion of听privacy brought about by the intrusive nature of听video based听technologies.鈥�

People

Rob Piechocki: Rob is full Professor in the School of Computer Science, Electrical and Electronic Engineering and Engineering Maths. His research interests span the areas of Connected Intelligent Systems, Wireless Networks, Information and Communication Theory, Statistics and AI. His engineering domain expertise is wireless sensing for eHealth and Connected and Automated Vehicles (CAV).听He has published over 200 papers in international journals and conferences and 听holds 13 patents in these areas. He has lead wireless connectivity and sensing research activities for the IRC SPHERE project (winner of 2016 World Technology Award). He is CI on research grants totaling over 拢20M, and PI for several current high-profile projects in networks, connectivity and sensing funded by industry, Innovate UK and EPSRC such as VENTURER, FLOURISH, NG-CDI, OPERA, CAVShield (>拢4M). Rob regularly advices 听industry and the Government on many aspects related to connected intelligent technologies and data sciences.
Kevin Chetty: Kevin is an Associate Professor at 果冻影院. His research expertise lies in the field of radio-frequency (RF) sensing, and radar signal processing using machine learning. Major achievements in RF sensing include demonstrating the first detections of personnel targets using passive WiFi radar, and proving the ability of these systems to perform through-the-wall sensing at standoff distances. In the signal processing domain Kevin鈥檚 work has focused on classifying peoples actions based on their radar micro-Doppler signatures for applications in both security and e-healthcare. His work in this area also covers indoor mapping, target tracking, and high-throughput data processing. Kevin is author to over fifty peer reviewed publications and has been an investigator on grants in passive wireless sensing and FMCW radar for microDoppler target classification funded by EPSRC, EU Joint Research Centre, UK Ministry of Defence/Dstl and the UK Maritime & Coastguard Agency. He also sits on the Technical and Organising Committees for the International Security & Crime Science conference and is a member of the IEEE and IET, and reviewer for a number of high-esteem journals including: IET Radar, Sonar and Navigation; IEEE Geoscience & Remote Sensing; and IEEE Communications magazine.
Nic Lane: Nic is an Associate Professor in the听Computer Science department听at the听University of Oxford, where he leads the听Machine Learning Systems lab. This lab is part of the听Cyber-Physical Systems group. At Oxford, he is also a Fellow of听Kellogg College, and teaches machine learning on the Professional Masters Program. Along side his academic roles, he is a Programme Director at听the recently announced听Samsung AI Center听in Cambridge. Broadly, Nic鈥檚 research interests revolve around the systems and modeling challenges that arise when computers collect and reason about sensor data. In the past, he has developed computational models of听human behavior and context (using audio, image, location and inertial data), in addition to algorithms and systems software for mobile and embedded platforms. More recently, his work has focused on high efficiency deep learning and enabling training and inference to occur within limited compute, memory and energy budgets.听听Previously, Nic held dual academic and industrial appointments as a Senior Lecturer (Associate Professor) in the Computer Science department at 果冻影院 (果冻影院), and a Principal Scientist at Nokia Bell Labs. At 果冻影院 he was part of the Digital Health Institute and 果冻影院 Interaction Center, while at the Bell Labs he led DeepX 鈥� an embedded focused deep learning unit at the Cambridge location.听Before moving to England, Nic spent four years at Microsoft Research based in Beijing. There he was a Lead Researcher within the Mobile and Sensing Systems group (MASS). In March 2011, he received a Ph.D. from Dartmouth College.
Ian Craddock: Ian is a Professor in the Faculty of Engineering at the University of Bristol. He is Institutional Lead for Digital Health, leading a programme of investment and research development across six faculties of his University. He leads the EPSRC funded "SPHERE" IRC programme, one of the UK's largest digital health research projects. His track record includes the successful commercialisation of a medical device. He was employed by Toshiba corporation as Managing Director of their research laboratory in Bristol from 2011 to late 2018. Ian's research interests including embedded AI for constrained devices such as wearables, the use of AI in clinical decision support, the analysis and visualisation of time-domain streaming data and the interface between health, AI and society.
Karl Woodbridge: Karl听is Emeritus Professor of Electronic and Electrical Engineering at 果冻影院. Recent research interests include multi-static and software-defined radar systems, passive wireless surveillance and Doppler classification using machine learning methods. Current research is focussed on the development of passive wireless based sensors for activity detection and classification with application to听Healthcare, IoT and Security. He is a Fellow of the IET, a Fellow of the UK Institute of Physics and a Senior Member of the IEEE. He has published or presented over 200 journal and conference papers in the areas of semiconductors, photonics and RF sensor systems.
Raul Santos-Rodriguez: Raul is a Senior Lecturer in Data Science and AI at the听University of Bristol. His research interests lie in machine learning, data science, artificial intelligence and their application to healthcare, media analysis, music information retrieval and more.听After a year as a data scientist developing large-scale recommender systems for a content provider company, in 2014 Raul moved to IPL, Universitat de Valencia, where he was a Research Fellow working with听Prof Gustau Camps-Valls. Before, he was a post-doctoral researcher at the听Intelligent Systems Laboratory, University of Bristol, working with听Prof Tijl de Bie. He completed my PhD in Machine Learning in 2011, under the supervision of听Prof Jesus Cid-Sueiro. He spent research visits at听果冻影院,听University of Bristol,听University of 罢谤辞尘蝉酶听补苍诲听狈滨颁罢础.
Jonas Paulavicius: Jonas works on deep learning for indoor sensors, mostly focusing on active non-intrusive sensing with radar. His main interests are in practicalities of model deployment - model trust and continuous learning. He is听also interested in model-based reconstruction and functional priors in deep learning.
Shelly Vishwakarma: Shelly received the B.Tech and M.Tech degrees in electronics and communication engineering from Guru Gobind Singh Indraprastha University, Delhi, India, in 2011 and 2013, respectively, and the Ph.D. degree from Indraprastha Institute of Information Technology Delhi, New Delhi, India, in 2020, in micro-Doppler radar based detection, classification and imaging of indoor targets. She is currently a research fellow with the Department of Security and Crime Sciences, 果冻影院. Her interests include micro-Doppler based passive sensing of indoor occupants.
Mohammud Junaid Bocus: Mohammud received the B.Eng. degree (first-class honors) in Electronic and Communication Engineering from the University of Mauritius, Mauritius in 2012, the M.Sc. (Distinction) degree in Wireless Communications and Signal Processing from the University of Bristol, Bristol, U.K, in 2015 and PhD degree in Electrical and Electronic Engineering from the University of Bristol, Bristol, U.K, in 2020. His research interests include terrestrial and underwater wireless communication based on orthogonal frequency-division multiplexing (OFDM), filter-bank multicarrier (FBMC) modulation, multiple-input multiple-output (MIMO) and massive MIMO systems as well as video coding. He is currently working on the OPERA project as a research associate, focusing on activity recognition and localisation using commodity WiFi and ultra-wideband (UWB) systems.
Wenda Li: Wenda is currently a research fellow in the Department of Security and Crime Science at 果冻影院 (果冻影院), London, UK. He听received the M.Eng and Ph.D. degree in Electrical and Electronic Engineering from the University of Bristol, Bristol, UK in 2013 and 2018, respectively. He has worked as a research fellow at University of Birmingham. He has also been an honorary research fellow at University of Bristol. His research focuses on signal processing for radar and wireless communications system, digital system design and wireless sensing applications in healthcare, security and positioning.
Chong Tang: Chong is a PhD student in the 果冻影院 Department of Security and Crime Science, where he is investigating the applicability of passive WiFi radar technology for occupancy detection and human-skeletal model reconstruction using only Doppler data. He has received a Bachelors degree in Automation and Electrical & Electronic Engineering from University of Nottingham in 2018, and a Masters degree in Robotics from 果冻影院 in 2019.

Publications

2020

K. Ton-Tran, L.Griffin, S. Vishwakarma and K. Chetty, "Transfer Learning from Audio Deep Learning Models for Micro-Doppler Activity Recognition", 2020 IEEE Radar Conference (RadarConf), Washington, USA.

2019