co-located with The 23rd International Conference on Pervasive Computing and Communications (PerCom 2025)
Washington DC, March 17-21, 2025.
Wireless sensing has recently attracted a great deal of attention thanks to its non-invasive and sensor-free nature. Contrary to traditional sensor-based and wearable sensing, wireless sensing does not need any sensors, but leverages the modifications induced on the wireless channel by objects and people to infer information about their position and movement within a physical environment. This leads to an unobtrusive system that can be integrated with nowadays transmission technology. Moreover, if operated, e.g., at sub-6GHz frequencies, wireless signals propagate through walls, allowing sensing to be performed even in non-line-of sight (NLOS) scenarios with a subsequent increase in the sensing coverage over camera-based systems. Different types of wireless signals have been employed for sensing including WiFi, RFID, mmWave, UWB, and acoustic signals. As wireless signals bounce off of physical objects within the environment such as static objects like walls or furniture, as well as any humans in the environment, their characteristics (e.g., amplitude, phase) change uniquely. This then provides an opportunity to sense the environment and obtain contextual information (e.g., recognizing the motion) through a finegrained analysis of signal variations. Wireless sensing has been considered in various applications including but not limited to localization, human activity and gesture recognition, gait estimation, fall detection, respiration monitoring, crowd counting, etc.
Deploying the wireless sensing systems on edge devices is also important, to reduce their costs and make them scalable. However, this comes with several challenges due to the constrained resources (e.g., memory, computation power, energy) of edge nodes. Accordingly, in this workshop we also look for solutions that develop novel, lightweight and cost-efficient techniques that can run at the network edge, providing means to train and run machine learning models in an energy efficient manner, both according to centralized and distributed training paradigms. We are also interested in the characterization of energetic aspects on currently available edge computing technology, including but not limited to empirical energy models.
The objective of this workshop is to bring together the research community utilizing different types of wireless signals for sensing purposes, and the community dealing with computing for embedded and energy efficient systems, and have them benefit from each other's findings. The workshop will also serve as a discussion platform about the standardization and industrial implications of wireless sensing and edge computing, by also targeting potential privacy issues. Toward these goals, the workshop will span topics including, but not limited to:
Submissions will be made using Easy Chair
Manuscripts submitted for consideration should not have been already published elsewhere and should not be under review or submitted for review elsewhere during the consideration period. Manuscripts must be written in English, are limited to 6 pages, single spacing, double column, and must strictly adhere to the IEEE template format available here. All accepted papers will be included and indexed in the IEEE Digital Library (IEEE Xplore), showing their affiliation with IEEE Percom 2025. At least one author of each accepted paper is required to attend and present his/her work at the workshop.
Paper submission deadline: | |
Author Notification: | |
Camera-ready submission: | February 2, 2025 |
Workshop date: | TBD |
Abstract: With more and more bandwidth readily available for the next generation of wireless applications, many intelligent products/services by leveraging the ambient radio waves unimaginable before are now possible. What impact will it bring to our lives? Many may wonder or even speculate if it is science fiction or real? In this talk, we will show that with more bandwidth, one can see many multi-paths, which can serve as hundreds of virtual sensors around us that can be readily leveraged as new degrees of freedom for our use (but we never realized that before!). We discovered for the first time that the time-reversal focusing spot is not a point but exhibiting a stationary power distribution of Bessel function which enabled accurate/reliable speed estimation under nonline- of-sight, severe multipath conditions, uncovering a new physical principle ideal for indoor applications. Together with the use of machine learning, a revolutionary wireless AI platform can be built to enable many IoT applications that have been dreamed for a long time but have never been possibly achieved. Such a technology forms the core of a wireless sensing AI platform that can be applied to many device-free, nonobtrusive applications. We will show the world’s first ever centimeter-accuracy wireless indoor positioning systems that can offer indoor tracking without any costly infrastructure, home/office monitoring and security, radio human biometrics, vital signs detection, sleep monitoring, gait recognition, and fall detection, without any wearable but sorely relying on ubiquitous commodity Wi-Fi. In essence, now and in the future, it is the wireless sensing that will forever change Wi-Fi as we know it today, as well as future 5G/6G systems, allowing us to decipher our surrounding world with a new “sixth sense”. Some products/services of Origin have been deployed worldwide and will be demonstrated to illustrate how such a fundamental discovery can make the world a better place.
![]() |
About the Speaker: K. J. Ray Liu is the founder of Origin Wireless that pioneers AI for wireless sensing and indoor tracking. The invention of wireless AI won three prestigious CES Innovation Awards, including CES Best of Innovation in 2021, and 2017 CEATEC Grand Prix. Dr. Liu was the 2022 IEEE President and CEO. He was IEEE Vice President, Technical Activities, Division IX Director of IEEE Board of Director, President of IEEE Signal Processing Society, where he has served as Vice President – Publications and Editor-in-Chief of IEEE Signal Processing Magazine. He was Distinguished University Professor and Christine Kim Eminent Professor of Information Technology of the University of Maryland, College Park, from where he retired after over three decades of career in education. His research contributions encompass broad aspects of signal processing and communications. He has trained over 70 doctoral/postdoctoral students, of which 12 are now IEEE fellows. According to Mathematics Genealogy Project, he has had over 200 doctoral descendants. Dr. Liu is a recipient of two IEEE Technical Field Awards: the 2021 IEEE Fourier Award for Signal Processing and the 2016 IEEE Leon K. Kirchmayer Graduate Teaching Award, and also IEEE Signal Processing Society 2014 Norbert Wiener Society Award, 2009 Claude Shannon-Harry Nyquist Technical Achievement Award, and more than a dozen best paper awards. Recognized by Web of Science as a Highly Cited Researcher, he is a Fellow of IEEE, AAAS, and US National Academy of Inventors. |