Collaborative Research: SWIFT: SMALL: Understanding and Combating Adversarial Spectrum Learning towards Spectrum-Efficient Wireless Networking

点击翻译按钮获取中文摘要

The next-generation wireless network is expected to be efficient, reliable, secure and spectrum-intelligent to maximize the efficiency of using the wireless spectrum. Dynamic spectrum access and management designs have shown their potential to substantially improve the spectrum utilization efficiency. To achieve the goal of securing efficient spectrum access and utilization, it is common in wireless network systems to collect and use spectrum reports from individual nodes to detect malicious behaviors and/or eliminate attack impacts. The goal of this project is to understand the potential strategies and impacts of a new attack, called adversarial spectrum learning, which aims to learn from the wireless spectrum data and construct specific attack models against spectrum management systems to disrupt the network performance. The project will also provide effective countermeasure against adversarial spectrum learning.The project will focus on studying important research problems associated with adversarial spectrum learning via analytical modeling, comprehensive simulations, and experimental evaluations. Specifically, the research team aims at (i) formulating the adversarial spectrum learning strategies that can be used by malicious attacks to disrupt the spectrum efficiency and evaluating their damaging impacts in varying wireless and network conditions; (ii) detecting adversarial spectrum learning attacks via new non-parametric detection methods; (iii) creating attack prevention and secure management methods for secure and efficient spectrum access in the potential presence of adversarial spectrum learning, and (iv) comprehensively evaluating the efficiency and effectiveness of the proposed detection and prevention strategies. Appropriate elements from the project will also be integrated into educational materials.This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.