Distinguished Lectures

ICNC 2023 features 3 Distinguished Lectures, which are OPEN to ALL attendees of the conference and workshops.

Shiwen Mao (IEEE Fellow)
Professor and Earle C. Williams Eminent Scholar, Auburn University, USA.

Title: RFID-based 3D Human Pose Tracking: Design, Generalization, and Data Augmentation
Time: 13:3:00-15:30, Wednesday, February 22, 2023

Abstract:
In recent years, 3D human activity recognition (HAR) has become an important topic in human-computer interaction (HCI). It is a more challenging problem than classification, and is the enabler of many HCI applications. To preserve the privacy of users, there is considerable interest in techniques without using a video camera. To this end, RFID tags, as a low-cost wearable sensor, provide an effective solution for 3D human pose tracking. In this talk, we first present RFID-Pose, a vision-aided realtime 3D human pose estimation system based on deep learning. The RFID phase data is calibrated first. The system then estimates the spatial rotation angle of each human limb, and utilizes the rotation angles to reconstruct human pose in realtime with the forward kinematic technique. The second part of this talk addresses the generalization problem, when a well-trained system is applied to a new environment. Meta-learning, as an effective technique to improve the model adaptability, is leveraged as a promising solution to the generalization problem. In the third part, we present a technology-agnostic approach for RF-based human activity recognition (HAR), termed TARF, which works with several different RF sensing technologies in different frequency bands. The proposed HAR solution works with different types of RF technologies, such that the cost and the barrier of wide deployment can both be greatly reduced, and more robust performance can be achieved by utilizing the complementary RF sensory data. Finally, in order to mitigate the cost of collecting training data, we propose a data augmentation method based on Generative Adversarial Network (GAN), named RFPose-GAN, to generate synthesized RFID data. Our experiments demonstrate that the synthesized data is useful for training the RFID-Pose model to achieve high pose estimation performance.

Biography:
SHIWEN MAO is a professor and Earle C. Williams Eminent Scholar Chair, and Director of the Wireless Engineering Research and Education Center (WEREC) at Auburn University, Auburn, AL, USA. His research interest includes wireless networks, multimedia communications, and smart grid. He is a Distinguished Lecturer of IEEE Communications Society and the IEEE Council of RFID, and is on the Editorial Board of several IEEE and ACM journals. He received the IEEE ComSoc TC-CSR Distinguished Technical Achievement Award in 2019 and NSF CAREER Award in 2010. He is a co-recipient of the 2021 Best Paper Award of Elsevier/KeAi Digital Communications and Networks Journal, the 2021 IEEE Communications Society Outstanding Paper Award, the IEEE Vehicular Technology Society 2020 Jack Neubauer Memorial Award, the 2004 IEEE Communications Society Leonard G. Abraham Prize in the Field of Communications Systems, and several conference best paper/demo awards. He is a Fellow of the IEEE.


Yi Qian (IEEE Fellow)
Professor, University of Nebraska-Lincoln, USA.

Title: Machine Learning and Security for Vehicular Communication Networks
Time: 16:00-18:00, Wednesday, February 22, 2023

Abstract:
Vehicular networks have been considered as a promising solution to achieve better traffic management and to improve the driving experience of drivers. However, vehicular networks are susceptible to various security attacks. Due to the wireless nature of vehicular communications, how to secure vehicular networks are great challenges that have hampered the implementation of vehicular network services. Many solutions have been proposed by researchers and industry in the recent years. In this tutorial, we first present an overview of security issues for vehicular networks, followed by a survey on the state-of-the-art solutions on security for vehicular networks. After that, we present two case studies on misbehavior detections in vehicular communication networks, one by introducing machine learning and reputation-based misbehavior detection systems to enhance the detection accuracy as well as to ensure the reliability of both vehicles and messages, another by introducing a deep reinforcement learning based dynamic reputation policy for misbehavior detections in vehicular networks. These types of misbehavior detection systems are trained using datasets generated through extensive simulations based on the realistic vehicular network environment. We show that various machine learning schemes can be exploited in accurately identifying several misbehaviors in vehicular networks.

Biography:
Yi Qian is a professor in the Department of Electrical and Computer Engineering, University of Nebraska-Lincoln (UNL). Prior to joining UNL, he worked in the telecommunications industry, academia, and government. His research interests include wireless communication networks and systems, and information and communication network security. Prof. Yi Qian is a Fellow of IEEE. He was previously Chair of the IEEE Technical Committee for Communications and Information Security. He was the Technical Program Chair for IEEE International Conference on Communications 2018. He serves or has served on the Editorial Boards of several international journals and magazines, including as the Editor-in-Chief for IEEE Wireless Communications between July 2018 and June 2022. He was a Distinguished Lecturer for IEEE Vehicular Technology Society and a Distinguished Lecturer for IEEE Communications Society.


Junshan Zhang (IEEE Fellow)
Professor, University of California, Davis, USA.

Title: Adaptive Ensemble Q-learning: Minimizing Estimation Bias via Error Feedback.
Time: 10:00-12:00, Wednesday, February 22, 2023

Abstract:
The ensemble method is a promising way to mitigate the overestimation issue in Q-learning, where multiple function approximators are used to estimate the action values. It is known that the estimation bias hinges heavily on the ensemble size (i.e., the number of Q-function approximators used in the target), and that determining the `right' ensemble size is highly nontrivial, because of the time-varying nature of the function approximation errors during the learning process. To tackle this challenge, we first derive an upper bound and a lower bound on the estimation bias, based on which the ensemble size is adapted to drive the bias to be nearly zero, thereby coping with the impact of the time-varying approximation errors accordingly. Motivated by the theoretic findings, we advocate that the ensemble method can be combined with Model Identification Adaptive Control (MIAC) for effective ensemble size adaptation. Specifically, we devise Adaptive Ensemble Q-learning (AdaEQ), a generalized ensemble method with two key steps: (a) approximation error characterization which serves as the feedback for flexibly controlling the ensemble size, and (b) ensemble size adaptation tailored towards minimizing the estimation bias.

Biography:
Junshan Zhang is a professor in the ECE Department at University of California Davis. He received his Ph.D. degree from the School of ECE at Purdue University in Aug. 2000, and was on the faculty of the School of ECEE at Arizona State University from 2000 to 2021. His research interests fall in the general field of information networks and data science, including edge intelligence, reinforcement learning, continual learning, network optimization and control, game theory, with applications in connected and automated vehicles, 5G and beyond, wireless networks, IoT data privacy/security, and smart grid.

Prof. Zhang is a Fellow of the IEEE, and a recipient of the ONR Young Investigator Award in 2005 and the NSF CAREER award in 2003. He received the IEEE Wireless Communication Technical Committee Recognition Award in 2016. His papers have won a few awards, including the Best Student Paper award at WiOPT 2018, the Kenneth C. Sevcik Outstanding Student Paper Award of ACM SIGMETRICS/IFIP Performance 2016, the Best Paper Runner-up Award of IEEE INFOCOM 2009 and IEEE INFOCOM 2014, and the Best Paper Award at IEEE ICC 2008 and ICC 2017. Building on his research findings, he co-founded Smartiply Inc, a Fog Computing startup company delivering boosted network connectivity and embedded artificial intelligence.

Prof. Zhang is currently serving as Editor-in-chief for IEEE Transactions on Wireless Communications and a senior editor for IEEE/ACM Transactions on Networking. He was TPC co-chair for a number of major conferences in communication networks, including IEEE INFOCOM 2012 and ACM MOBIHOC 2015. He was the general chair for ACM/IEEE SEC 2017, WiOPT 2016, and IEEE Communication Theory Workshop 2007. He was a Distinguished Lecturer of the IEEE Communications Society.