IRES Track 1: Sensor Information Processing and Machine Learning for Wearable Devices

This IRES project brings together advances in sensor devices with machine learning and digital signal processing (DSP) algorithms in an international research endeavor that promises to elevate precision in mobile and wearable technologies. Arizona State University (ASU) faculty and students will collaborate with the Dublin City University (DCU) Insight Center for Data Analytics, which has a synergistic relationship with ASU in several areas including sensors, analytics, machine learning (ML), wearables and Internet of Things (IoT).

ASU brings expertise in flexible sensors, chemical and biosensors, statistical signal processing, bio-informatics, and machine learning. DCU brings expertise in data analytics, human activity monitoring, environmental monitoring, artificial intelligence, sensor analytics and big data analysis. IRES participating students will spend an immersive six-week summer program at the DCU Insight Center to actively improve their ML and sensor research skills; participants will produce and evaluate sensor analytics, and create algorithms and software for IoT, wearables and mobile health monitoring.

Programs and workshops will be established to train IRES participants to skillfully and effectively present their research in international settings. Weekly presentations at the international site and guidance by international mentors will enrich the cohort’s professional experience. Embedding students in the DCU Insight center funded by European Union (EU) and Irish Science Foundation (ISF) grants will provide knowledge on EU and international research practices, ethics, standards and policies.

The goals of this project are to: a) advance the science of integrated design of sensors and machine learning algorithms, b) train and enable a diverse cohort of students to make research contributions in integrated sensing and ML for IoT systems, c) gain knowledge on international policies/standards of deploying AI, big data systems, and sensors, and d) provide experiences that broaden understanding of global practices and career options. This project is motivated by the fact that inexpensive sensors are required for IoT, mobile health and wearable systems; to achieve the requisite precision, sensor design must be accompanied by corrective ML and SP algorithms.

IRES research therefore focuses on the overlap of new sensor device design and novel ML algorithm development. In terms of ML, one of the key objectives is to develop compact, low power algorithms adequate for integration with sensors and mobile devices. IRES project areas include flexible sensors, sensor information management, efficient deep learning, and big data analysis.

Example research project applications include biomarker detection, big data processing, gait detection, and deep neural nets for sensor and IoT systems. Work will be disseminated via collaborative publications and presentations in international conferences and refereed journals. Industry engagement at the ASU SenSIP and DCU Insight centers will provide ongoing valuable feedback, and annual external evaluation will assess progress and outcomes across all IRES activities.

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.