loading...
logo

   jliu AT utk.edu

   865-974-4389

   865-974-5483

   307 Min H. Kao Bldg

UTK

MoSIS Lab

Follow me on:

Teaching in Fall'23: ECE 469/569

Looking for self-motivated Ph.D. students (Spring/Fall 2024): preferred areas of interest and expertise include trustworthy ML & AI, audio processing, mobile sensing, and human-computer interaction. [Ads]
Looking for undergraduate researchers (hourly pay): preferred experience in machine learning, mobile app development, and embedded system development, etc.
About

I am an Assistant Professor in the Min H. Kao Department of Electrical Engineering and Computer Science at The University of Tennessee, Knoxville. I lead Mobile Sensing and Intelligence Security (MoSIS) Lab @ UTK. Before joining UTK, I received my Ph.D. degree from Rutgers, The State University of New Jersey. My research interests span Robust and Trustworthy AI and ML, Human-Computer Interaction, Computational Sensing, Mobile Security, and Smart Healthcare. My research work has been published at top-tier mobile-computing/security/HCI/AI venues (e.g., S&P/Oakland, ACM CCS, ACM MobiCom, CVPR, AAAI, ECCV, ACM MobiSys, ACM SenSys, ACM UbiComp, and IEEE ICASSP) and has been regularly featured in the media including BBC News, Yahoo News, MIT Technology Review, NBC New York, IEEE Spectrum, WCBS TV, and Voice of America TV, etc. I also hold four U.S. patents, two of which have been licensed to industrial companies.
My research has been generously supported by:

NSF  UTK  Science Alliance  UT-ORII  Google  Nvidia  CHECCS
Recent Research Highlights

[S&P/Oakland'23] Privacy Leakage in the Age of Virtual Reality
Paper Project Website

[ICASSP'23] Speech Privacy Leakage in Federated Learning
Paper

UbiComp 2023

[UbiComp'23] Smart Seat Pad for Cycling via Conductive Fabric Sensors
Paper [Provisional Patent]

[CCS'20] Adversarial ML Attacks Against Voice Assistants via Subsecond Perturbations
Paper Project Website

[MobiCom'21] 3D Facial Reconstruction via Lightweight Single-ear Biosensors
[ACM SigMobile Research Highlights 2022]
Paper Project Website [Non-Provisional Patent]

CVPR 2022

[CVPR'22] Auditing Privacy Defenses in Federated Learning via Gradient Leakage
Paper Source Code

Latest News
  • [December 2023] Our collaborative work "Pathogen Watch in Food Manufacturing" won the 2023 IChemE Biochemical Engineering Award! GLOBAwards
  • [November 2023] Invited talk at the 2nd workshop on Federated Learning and Privacy Preserving AI for HPC at SC'23 in November in Denver, CO.
  • [November 2023] Invited talk at L&N STEM Academy.
  • [August 2023] The website for my Fall'23 course, ECE 469/569: Mobile and Embedded System Security, is now Live!
  • [July 2023] We have filed a provisional patent for our smart seat pad technology!
  • [July 2023] Our paper "SmarCyPad: A Smart Seat Pad for Cycling Fitness Tracking Leveraging Low-cost Conductive Fabric Sensors" has been accepted to ACM UbiComp'23! Kudos to the team, with three undergraduate researchers involved!
  • [May 2023] Congrats to Yi for receiving UTK EECS Gonzalez Outstanding GRA Award!
  • [May 2023] Our paper "DroneChase: A Mobile and Automated Cross-Modality System for Continuous Drone Tracking" has been accepted to ACM DroNet IX!
  • [March 2023] Our paper "Privacy Leakage via Unrestricted Motion-Position Sensors in the Age of Virtual Reality: A Study of Snooping Typed Input on Virtual Keyboards" has been accepted to IEEE S&P/Oakland 2023! Congrats to Yi!
  • [March 2023] Invited Talk at Brigham Young University, in Provo, UT!
  • [February 2023] We received an award from the "CHECCS" center.
  • [February 2023] Our paper "Speech Privacy Leakage from Shared Gradients in Distributed Learning" has been accepted to IEEE ICASSP 2023! Congrats to Zhuohang!
  • [January 2023] Our paper "RecUP-FL: Reconciling Utility and Privacy in Federated Learning via User-configurable Privacy Defense" has been accepted to ACM AsiaCCS 2023 (Acceptance rate: 16%)! Congrats to Yue!
  • [December 2022] Our project has been selected for the 2023 UTRF Maturation Grant Award!
  • [August 2022] Our paper "HeatDeCam: Detecting Hidden Spy Cameras via Thermal Emissions" has been accepted to ACM CCS 2022!
  • [July 2022] Our paper "RIBAC: Towards Robust and Imperceptible Backdoor Attack against Compact DNN" has been accepted to ECCV 2022!
  • [June 2022] Our paper "Audio-domain Position-independent Backdoor Attack via Unnoticeable Triggers" has been conditionally accepted to ACM MobiCom 2022!
  • [June 2022] Our MobiCom'21 paper BioFace-3D was highlighted in GetMobile!
  • [May 2022] We received an REU supplement for the NSF CCSS project.
  • [May 2022] The Science Alliance renewed our research project on federated learning for the 2nd year.
  • [April 2022] We received an award from the UT-ORII Science Alliance Faculty Incentive Program!
  • [April 2022] Congrats to Zhuohang for receiving UTK EECS Gonzalez Outstanding GRA Award!
  • [April 2022] Our MobiCom'21 paper BioFace-3D has been selected as the ACM SigMobile Research Highlights 2022!
  • [April 2022] Our paper "Fair and Privacy-Preserving Alzheimer's Disease Diagnosis Based on Spontaneous Speech Analysis via Federated Learning" has been accepted to IEEE EMBC 2022! Congrats to Irfan!
  • [April 2022] Our paper "Privacy-preserving Speech-based Depression Diagnosis via Federated Learning" has been accepted to IEEE EMBC 2022! Congrats to Yue!
  • [March 2022] Our paper "Auditing Privacy Defenses in Federated Learning via Generative Gradient Leakage" has been accepted to CVPR 2022 (Acceptance rate: 25.3%)! Congrats to Zhuohang!
  • [February 2022] Our paper "Invisible and Efficient Backdoor Attacks for Compressed Deep Neural Networks" has been accepted to ICASSP 2022!
  • [December 2021] Our NSF BIOSENS proposal on "High-precision Monitoring of Foodborne Pathogens in Food Manufacturing Facilities" is awarded (as PI).
  • [December 2021] Our paper "Robust Continuous Authentication Using Cardiac Biometrics from Wrist-worn Wearables" has been accepted to IEEE Internet of Things Journal (IEEE IoT)!
  • [November 2021] Our paper "mPose: Environment- and Subject-Agnostic 3D Skeleton Posture Reconstruction Leveraging a Single mmWave Device" has been accepted to IEEE/ACM CHASE 2021!
  • [September 2021] Our paper "Byzantine-robust Federated Learning through Spatial-temporal Analysis of Local Model Updates" has been accepted to IEEE ICPADS 2021!
  • [August 2021] Our paper "Robust Detection of Machine-induced Audio Attacks in Intelligent Audio Systems with Microphone Array" has been accepted to ACM CCS 2021! Congrats to Zhuohang!
  • [August 2021] Our paper "Time to Rethink the Design of Qi Standard? Security and Privacy Vulnerability Analysis of Qi Wireless Charging" has been accepted to ACSAC 2021! Congrats to Yi, Zhuohang and Nicholas!
  • [August 2021] Our NSF SaTC proposal on "Securing IoT and Edge Devices under Audio Adversarial Attacks" is awarded (as PI).
  • [August 2021] Our NSF CCSS proposal on "Continuous Facial Sensing and 3D Reconstruction via Single-ear Wearable Biosensors" is awarded (as Lead PI).
  • [August 2021] Our paper "BioFace-3D: Continuous 3D Facial Reconstruction Through Lightweight Single-ear Biosensors" has been accepted to MobiCom 2021! Congrats to Yi and Zhuohang!
  • [August 2021] Our paper "Face-Mic: Inferring Live Speech and Speaker Identity via Subtle Facial Dynamics Captured by AR/VR Motion Sensors" has been accepted to MobiCom 2021!
  • [August 2021] NVIDIA is donating us ten Jetson AI Developer Kits!
  • [July 2021] We received Google Cloud Research Award. Thanks, Google!
  • [May 2021] Our paper "Spearphone: A Lightweight Speech Privacy Exploit via Accelerometer-Sensed Reverberations from Smartphone Loudspeakers" has been accepted to WiSec 2021 (Acceptance rate: 22.3%)!
  • [February 2021] The Science Alliance awarded our research project "Towards Robust and Trustworthy Federated Learning for Ubiquitous Cyber-Physical Systems: Security, Privacy, and Scalability".
  • [January 2021] Our journal paper "WiFi-enabled User Authentication through Deep Learning in Daily Activities" has been accepted to ACM Transactions on Internet of Things.
  • [January 2021] Our journal paper "Enabling Finger-touch-based Mobile User Authentication via Physical Vibrations on IoT Devices" has been accepted to IEEE Transactions on Mobile Computing (TMC).
  • [Dectember 2020] Invited to serve on the TPC of ACM WiSec 2021.
  • [Dectember 2020] Our audio adversarial attack paper has been accepted to Journal of Signal Processing Systems (Springer JSPS).
  • [Dectember 2020] Our paper on human motion recognition via radar radios has been accepted to IEEE Sensors Journal.
  • [November 2020] Our paper "Enabling Fast and Universal Audio Adversarial Attack Using Generative Model" has been accepted to AAAI 2021 (Acceptance rate: 21%)!
  • [October 2020] Our paper "EchoVib: Exploring Voice Authentication via Unique Non-Linear Vibrations of Short Replayed Speech" has been accepted to ACM AsiaCCS 2021 (Acceptance rate: 18.5%)!
  • [October 2020] Our paper "HVAC: Evading Classifier-based Defenses in Hidden Voice Attacks" has been accepted to ACM AsiaCCS 2021 (Acceptance rate: 18.5%)! Congrats to Yi! Good job!
  • [September 2020] Please check our AdvPulse Webpage (CCS'20 work), on which we released a few audio/video samples to showcase how powerful our attack is.
  • [September 2020] Our paper "BatComm: Enabling Inaudible Acoustic Communication with High-throughput for Mobile Devices" has been accepted to ACM SenSys 2020 (Acceptance rate: 20.6%)! We got 12× throughput improvement than contemporary state-of-the-art systems while keeping the whole communication process inaudible!
  • [September 2020] Our paper "AdvPulse: Universal, Synchronization-free, and Targeted Audio Adversarial Attacks via Subsecond Perturbations" has been accepted to ACM CCS 2020 (Acceptance rate: 16.9%)! Congrats to my students Zhuohang and Yi! Good job! (Audio and video samples will be released soon)
  • [September 2020] Our paper "Towards Environment-independent Behavior-based User Authentication Using WiFi" has been accepted to IEEE MASS 2020.
  • [September 2020] Invited to serve on the TPC of IEEE VNC 2020.
  • [August 2020] Invited to be a Keynote Speaker at the 8th Annual IEEE North Jersey Advanced Communications Symposium (NJACS-2020) (Theme: AI and Deep Learning).
  • [July 2020] Our survey paper "Acoustic-based Sensing and Applications: a Survey" has been accepted to Computer Networks.
  • [June 2020] Received "ECE Graduate Program Academic Achievement Award" at Rutgers University.
  • [May 2020] Our paper "Mobile Device Usage Recommendation based on User Context Inference Using Embedded Sensors" has been accepted to IEEE ICCCN 2020.
  • [February 2020] Congratulations to Zhuohang, who received "One Time" UTK EECS Fellowship Award!
  • [February 2020] Invited to serve as a TPC member for the 16th EAI International Conference on Security and Privacy in Communication Networks (SecureComm 2020).
  • [February 2020] Our journal paper "When Your Wearables Become Your Fitness Mate" has been accepted to Smart Health (Elsevier).
  • [January 2020] Our paper "Real-time, Universal, and Robust Adversarial Attacks Against Speaker Recognition Systems" has been accepted to ICASSP 2020.
  • [January 2020] Our survey paper "User Authentication on Mobile Devices: Approaches, Threats and Trends" has been accepted to Computer Networks.
  • [December 2019] Our work "Towards Low-cost Sign Language Gesture Recognition Leveraging Wearables" has been accepted to IEEE Transactions on Mobile Computing (IEEE TMC).
  • [December 2019] Our work "Practical Adversarial Attacks Against Speaker Recognition Systems" has been accepted to ACM HotMobile'20 (Acceptance rate: 16/48 = 33.3%). Congratulations to Zhuohang!
  • [December 2019] Three full papers ("Continuous User Verification via Respiratory Biometrics", "MU-ID: Multi-user Identification Through Gaits Using Millimeter Wave Radios" and "TrueHeart: Continuous Authentication on Wrist-worn Wearables Using PPG-based Biometrics") have been accepted to IEEE INFOCOM'20 (Acceptance rate: 268/1,354 = 19.8%).
  • [November 2019] Our work "Enable Traditional Laptops with Virtual Writing Capability Leveraging Acoustic Signals" has been accepted to The Computer Journal.
  • [October 2019] Our work "Inaudible High-throughput Communication Through Acoustic Signals" has received Best Poster Award Runner-up at ACM MobiCom'19. This work also is the Winner of Student Research Competition (SRC). Congrats to Yang!
  • [October 2019] Invited to serve as a TPC member for the 16th International Conference on Wireless and Mobile Communications (ICWMC 2020).
  • [October 2019] Three demos (robot car dynamics monitoring, WiFi-based human activity real-time analysis, and human pose tracking using mm-wave sensors) have been accepted to IEEE DySPAN'19.
  • [October 2019] Invited to serve as the primary organization advisor of The Church of Jesus Christ Student Association at UTK.
  • [September 2019] Invited to serve as a TPC member for IEEE Vehicular Networking Conference (VNC) 2019.
  • [August 2019] Our defense system against hidden audio channel attacks on voice assistants has been accepted to ACSAC'19 (Acceptance rate: 60/226 = 22.6%).
  • [August 2019] Our new attack against speech recognition systems has been accepted to IEEE DySPAN'19.
  • [July 2019] Please check our recent work Spearphone, which has been reported in The Hacker News: New Attack Lets Android Apps Capture Loudspeaker Data Without Any Permission.
  • [July 2019] One survey paper on wireless sensing has been accepted to IEEE Communications Surveys and Tutorials (IF=22.97)!
Selected Publications
*H-index = 34, Citation = 4,456 according to Google Scholar
*19 papers published in top-tier mobile-computing/security/HCI/AI venues chosen by CSRankings
View Full Publication List
Teaching
  • [Spring'22/23] ECE 315: Signals and Systems.
  • [Fall'20] ECE 599/692: Special Topics: Mobile Sensing Systems & Applications.
  • [Summer'20] COSC 593: Independent Study (Security & Privacy on Smart Speakers).
  • [Spring'20/21/23; Fall'21/22] ECE 469/569: Mobile and Embedded System Security.
Selected Honors & Awards
  • IChemE Biochemical Engineering Award, IChemE, 2023.
  • SigMobile Research Highlights, ACM SigMobile, 2022.
  • ECE Graduate Program Academic Achievement Award, Rutgers University, 2019-2020.
  • Winner of Student Research Competition, ACM MobiCom, 2019.
  • Best Poster Award Runner-up, ACM MobiCom, 2019.
  • Best Teaching Assistant Award, Rutgers University, 2019.
  • Best Poster Award Runner-up, ACM MobiCom, 2018.
  • ECE Research Excellence Award, Rutgers University, 2018.
  • Best Paper Award, IEEE CNS 2018.
  • Best Paper Award, IEEE SECON 2017.
  • Best-in-session Presentation Award, IEEE INFOCOM 2017.
  • App Contest Finalist, ACM MobiCom, 2016.
  • Best Poster Award Runner-up, ACM MobiCom, 2016.
Misc.
  • I love music (rock, blues) and play bass guitar during leisure time. I also enjoy running (1:42 half-marathon time).