Abstract:Abstract Information security is of paramount importance in the modern society, and it is crucial that communication systems are conceived and implemented in order to be inherently resilient in this respect. In current wireless communication systems, some of the most sophisticated attack strategies aim at the physical layer, i.e., at the electromagnetic wave signal carrying the information, but the implied physical interaction with the target inherently leaves traces in the physical environment, which render these attacks typically detectable. However, this may not be the case for future (the 6th generation and beyond) wireless networks, whose current vision relies on the concept of “smart radio environment”, empowered by suitably engineering reflecting devices (also called as metasurfaces) that can manipulate the wave signals in unconventional fashions (e.g., non-specular reflections) and that can be reconfigured at will. These metasurface elements, which should be pervasively deployed and suitably disguised in the indoor and outdoor environment, potentially introduce new vulnerabilities to the physical-layer attacks that should be fully understood and addressed. To this aim, here, we put forward the concept of smart wireless attacks at the physical layer by exploiting the unique capabilities of programmable metasurfaces in the joint manipulations of radio waves and digital information in a wireless scenario. Specifically, we illustrate both passive and active operational modes. In the passive mode, an attacker is capable of eavesdropping and breaking the target wireless information transfer by controlling the programmable metasurface, without radiating any signal actively. In the active mode, an attacker can not only eavesdrop but also furtively falsify the target wireless communications by sending some deceptive information to the target. In both operational modes, the detectability of the attacker can be minimized. As a proof of concept, we design and realize an attacker prototype working in the Wi-Fi band around 2.4GHz, and demonstrate experimentally its ability to hack wireless data streams. Our results raise awareness on the new types of security threats and challenges that the next-generation wireless networks will likely have to face, and indicate that suitable mitigation strategies and specific security protocols need to be conceived and developed at the present stage, while the smart-radio-environment concept is still in its infancy.

Practical Adversarial Attack on WiFi Sensing Through Unnoticeable Communication Packet Perturbation

Security Analysis of WiFi-based Sensing Systems: Threats from Perturbation Attacks

Echo: Reverberation-based Fast Black-Box Adversarial Attacks on Intelligent Audio Systems.

Physical-World Attack towards WiFi-based Behavior Recognition

Detecting Rogue Access Points Using Client-agnostic Wireless Fingerprints.

Adversarial WiFi Sensing Using a Single Smartphone

Time to Think the Security of WiFi-Based Behavior Recognition Systems

An Imperceptible Eavesdropping Attack on WiFi Sensing Systems

Manipulating Semantic Communication by Adding Adversarial Perturbations to Wireless Channel

IoT Network Security from the Perspective of Adversarial Deep Learning

Spectrum Data Poisoning with Adversarial Deep Learning

ActListener: Imperceptible Activity Surveillance by Pervasive Wireless Infrastructures

Frequency-selective Adversarial Attack Against Deep Learning-Based Wireless Signal Classifiers

Real-time Identification of Rogue WiFi Connections Using Environment-Independent Physical Features

Adversarial Attacks Against Network Intrusion Detection in IoT Systems

Real-Time Identification of Rogue WiFi Connections in the Wild

Wireless Universal Adversarial Attack and Defense for Deep Learning-Based Modulation Classification

Metasurface-enabled smart wireless attacks at the physical layer

Adversarial Deep Learning for Over-the-Air Spectrum Poisoning Attacks

Transfer-based Adversarial Poisoning Attacks for Online (MIMO-)Deep Receviers

Practical Traffic-space Adversarial Attacks on Learning-based NIDSs.