（1）主题：Identity-Based Attack Detection Utilizing Reciprocal RSS Variations in Mobile Wireless Networks
In this talk, we briefly discuss Identity-based attacks (IBAs), which are one of the most serious threats to wireless networks. An attacker can masquerade as a legitimate user in the network and launch various attacks. Recently, there is an increasing interest in using the received signal strength (RSS) to detect IBAs in wireless networks. However, current schemes tend to generate excessive false alarms in the mobile scenario. We propose a novel RSS based technique, Reciprocal Channel Variation-based Identification (RCVIC) for the mobile wireless networks. RCVIC exploits the reciprocity of the wireless fading channel and RSS variations naturally incurred by mobility. It can both detect IBAs and clarify the genuine and attacking frames. Different from current schemes which partition the original RSS and then detect IBAs, RCVIC first detects IBAs by utilizing well-constructed RSS variation lists. We numerically evaluate RCVIC performance through theoretical analysis and simulations, and validate it through experiments using off-the-shelf 802.11 devices under different attacking patterns in real indoor and outdoor mobile scenarios. Our results show that RCVIC can achieve desirable detection performance.
Jie Tang was born in Chengdu, P.R.China. He achieved his Ph.D. degree in Communication and information system at the University of Electronic Science and Technology of China at 2018. He studied in the National Key Laboratory of Science and Technology on Communications at UESTC, P.R.China. His major interests focus on wireless communication, wireless physical layer security and information security. He was a visiting scholar at George Mason University, U.S.A.
（1）主题：Trajectory Tracking Control of Air-Breathing Hypersonic Vehicles
Air-breathing Hypersonic vehicles (AHVs) have attracted lots of attention for decades because of their prospects for high speed, large payload transportation and excellent cost-effectiveness to access the space routine. Over the past few years, numerous works have been done by US AirForce and NASA to further their development and designs. Although X-43A and X-51A trial vehicles have achieved success in recent years, flight control design for AHVs is still a challenging task. Because of very high flight speed, the aerodynamic properties are very difficult to be measured and estimated. Furthermore, high requirements of flight stability and high speed response, the strong couplings and various random uncertainties make it more difficult for controller design. Therefore, strong robust controllers are required exactly. Accordingly, linearized-model based control methods have been employed to deal with the flight control problem of AHV, including techniques of small perturbation and linear parameter-varying (LPV). Based on linearized model of AHV, linear control methods are employed to design the closed-loop control systems. However, linearized-model based control methods could not reflect the AHVs’ high-order nonlinearities. Therefore, advanced nonlinear control approaches for flight control of AHVs has been investigated in my researches. Based on the elastic longitudinal dynamics of AHVs, the trajectory tracking problem will be introduced in this report, while the typical nonlinear control methods based on nonlinear dynamical inversion will be presented.
Lin Cao is a post-doc majored in flight vehicle design. He has obtained bachelor’s degree in Haerbin Engineering University and Master’s degree in Haerbin Institute of Technology, then performed PhD research in Northwestern Polytechnical University. His research interests is mainly focus on flight control of hypersonic vehicles.
（1）主题：Stabilization of Time-delayed Power System with Time-domain Integral Quadratic Constraints
Time delay is prejudicial to power system stability and generally worsens the performance of control system. In this talk, we focus on the delays particularly due to the measurement and actuator of exciter, and power system stabilizers of synchronous machines. Nowadays, the main approach is mostly via Lyapunov stability theory and linear matrix inequality (LMI). However, the drawbacks of Lyapunov stability theory are that the stability condition greatly depends on the definition of the Lyapunov function and the computational burden is troublesome. Thus, we propose a novel approach, in which FD-IQC(Frequency-domain Integral Quadratic Constraints) and TD(Time-domain) dissipation inequality are combined, to handle the stabilization of power system including time-varying delays. Firstly, a time-delayed power system is modeled based on time-delay differential algebraic equations. Then, an extended state-feedback system defined by the IQC method is proposed to design a delay-dependent controller. The proposed controller is proved to guarantee the time-delayed power system exponential stability. Finally, the stability issue is taken as the KYP (Kalman-Yakubovich-Popov) LMI, and the controller parameters are solved by optimal L2 gain. Simulation results show that the proposed method is better than the Lyapunov stability theory not only in computational time-saving but also in performance.
Dongsheng Cai received the M.S. and Ph.D. degrees from the University of Electronic Science and Technology of China (UESTC) in 2011 and 2018, respectively. He visited the University of Tennessee as a joint doctoral student in 2014-2016. He currently works as a Postdoctoral Fellow in the School of Mechanical and Electrical Engineering of UESTC. His current research and academic interests include power system stability control and electromechanical wave theory.
（1）主题：Planning, Learning and Control of a Lower Exoskeleton System
Exoskeleton systems have gained considerable interests in both human augmentation, rehabilitation and elder-aided related applications. As a human-coupled robotic system, the main challenge of exoskeleton system design is how to make the exoskeleton system adapt different pilots and complex environment. In this talk, I will briefly introduce some of my work in planning, learning, and control of lower exoskeleton systems. This talk will only focus on rehabilitation related scenarios. Firstly, I will give a brief introduction of the AIDER lower exoskeleton system, as well as its history. Then some related work in planning, learning and control of the ADIER system will be introduced in detail. This part is separated into two episodes: the first is gait planning of the AIDER system, which include some online learning algorithm for gait model adaptation; the second is controller design of the AIDER system, which also include reinforcement learning algorithm implementation for adapting different pilots. Finally I will introduce some future work on the AIDER system.
R. Huang received the Bachelor degree, Master degree and Ph.D. degree from the University of Electronic Science and Technology of China, in 2010, 2013 and 2018, respectively. He currently works as a Postdoctoral Fellow in the School of Automation Engineering of UESTC. Dr. Huang is working in the field of Learning and Control of Human-Exoskeleton Systems. His main research interests include: Reinforcement Learning, Modeling of Human-Exoskeleton Systems, Controller design of Human-in-Loop Systems, Adaptive Dynamic Programming.
编辑：庄志东 / 审核：罗莎 / 发布者：陈伟