ECE Researchers Win ICASSP Best Student Paper Award
Department of Electrical and Computer Engineering graduate students Chen Chen and Yi Han, alumni Yan Chen (Ph.D. EE ’11) and Hung-Quoc Lai (Ph.D. EE ’11), and their advisor Christine Yurie Kim Eminent Professor of Information Technology K.J. Ray Liu (ECE) have received the 2016 Best Student Paper Award at the IEEE International Conference on Acoustics, Speech, and Signal Processing (ICASSP). The paper, entitled “High Accuracy Indoor Localization: A WiFi-Based Approach,” presents the authors’ work to create a positioning system which can achieve centimeter-level localization precision under non-line-of-sight conditions.
“I am privileged to receive this award at ICASSP 2016. I would like to thank all the coauthors for their hard work. This Best Paper award strengthens my belief that centimeter-level indoor localization precision is feasible and of high value among the Internet of Things applications which we have envisioned for so many years,” said Chen Chen, first author.
Co-author Yi Han also notes the recognition of their work as an illustration of the new future of indoor positioning. “This award further confirms that Time-Reversal technique has the greatest potential for indoor positioning.”
ECE graduate students Chen Chen and Yi Han are currently working on their Ph.D. degrees under the advisement of Professor Liu. Yan Chen and Hung-Quoc Lai graduated with their Ph.D. in electrical engineering in 2011 and are now alumni of Professor Liu’s research group. Yan Chen is a Professor at the University of Electronic Science and Technology of China. Hung-Quoc Lai is now the Vice President of Engineering at Origin Wireless, Inc.
The paper was selected from over 1200 accepted papers. The details of the selection process can be found at http://www.icassp2016.org/Awards.asp. The ICASSP 2016 Best Student Paper Award was sponsored by Starkey Hearing Technologies, iFLYTEK, and IBM. The award was presented at the opening ceremony of the conference held during March in Shanghai, China at the 41st meeting of ICASSP. ICASSP is the world's largest and most comprehensive technical conference on signal processing and its applications.
More about the research
Indoor positioning systems (IPS) based on Wi-Fi signals have recently gained popularity. IPS based on Received Signal Strength Indicator (RSSI) could only achieve a precision of several meters due to the strong temporal and special variation of indoor environments. On the other hand, IPS based on Channel State Information (CSI) drive the precision into the sub-meter regime with several access points (AP). However, the performance degrades with fewer APs mainly due to the limit of bandwidth. In this paper, the authors propose a Wi-Fi-based time-reversal indoor positioning system (WiFi-TRIPS) using the location-specific fingerprints generated by CSIs with a total bandwidth of 1 GHz. WiFi-TRIPS consists of an offline phase and an online phase. In the offline phase, CSIs are collected in different 10 MHz bands from each location of interest and the timing and frequency synchronization errors are compensated. The authors perform a bandwidth concatenation to combine CSIs in different bands into a single fingerprint of 1 GHz. In the online phase, the resesarchers evaluate the time-reversal resonating strength using the fingerprint from an unknown location and those in the database for location estimation. Extensive experiment results demonstrate a perfect 5 cm precision in an 20 cm x 70 cm area in a non-line-of-sight office environment with one link measurement.
April 6, 2016