Invited Speaker Seminar - Adversarial Machine Learning Attacks on RF Signal Classifiers

Date: Thursday, July 27, 2023 at 10 a.m.  
Location: EV 1.162

Abstract

Machine learning (ML) has recently been applied for the classification of radio frequency (RF) signals. One use case of interest relates to the discernment between different wireless protocols that operate over a shared and potentially contested spectrum. Although highly accurate classifiers have been developed for various wireless scenarios, research points to the vulnerability of such classifiers to adversarial machine learning (AML) attacks. In one such attack, a surrogate deep neural network (DNN) model is trained by the attacker to produce intelligently crafted low power “perturbations” that degrade the classification accuracy of the legitimate classifier. In this talk, I will first present several novel DNN protocol classifiers that we designed for a shared spectrum environment. These classifiers performed quite well in both simulations and OTA experimentation, considering benign (non-adversarial) noise. I will then present several AML techniques that an attacker may use to generate low power perturbations. When combined with a legitimate signal, these perturbations are shown to uniformly degrade the classification accuracy, even in the very high SNR regime. Different attack models are studied, depending on how much information the attacker has about the defender’s classifier. Finally, I will discuss possible defense mechanisms as well as other research efforts related to detection of adversarial transmissions.

Biography

Dr. Marwan Krunz is a Regents Professor at the University of Arizona. He holds the Kenneth VonBehren Endowed Professorship in ECE, and is also a professor of computer science. He directs the Broadband Wireless Access and Applications Center (BWAC), a multi-university NSF/industry center that focuses on next-generation wireless technologies. He also holds a courtesy appointment as a professor at University Technology Sydney. Previously, he served as the site director for Connection One, an NSF/industry-funded center of five universities and 20+ industry affiliates. Dr. Krunz’s research is in the fields of wireless communications, networking, and security, with recent focus on applying AI and machine learning techniques for protocol adaptation, resource management, and signal intelligence. He has published more than 320 journal articles and peer-reviewed conference papers, and is a named inventor on 12 patents. His latest h-index is 60. He is an IEEE Fellow, an Arizona Engineering Faculty Fellow, and an IEEE Communications Society Distinguished Lecturer (2013-2015). He received the NSF CAREER award. He served as the Editor-in-Chief for the IEEE Transactions on Mobile Computing. He also served as editor for numerous IEEE journals. He was the TPC chair for INFOCOM’04, SECON’05, WoWMoM’06, and Hot Interconnects 9. He was the general vice-chair for WiOpt 2016 and general co-chair for WiSec’12. Dr. Krunz served as chief scientist/technologist for two startup companies that focus on 5G and beyond wireless systems.