The evolution of 6G networks demands ultra-massive connectivity and intelligent radio environments, yet existing reconfigurable intelligent surface (RIS) technologies face critical limitations in hardware efficiency, dynamic control, and scalability. This paper introduces LLM-RIMSA, a transformative framework that integrates large language models (LLMs) with a novel reconfigurable intelligent metasurface antenna (RIMSA) architecture to address these challenges. Unlike conventional RIS designs, RIMSA employs parallel coaxial feeding and 2D metasurface integration, enabling each individual metamaterial element to independently adjust both its amplitude and phase. While traditional optimization and deep learning (DL) methods struggle with high-dimensional state spaces and prohibitive training costs for RIMSA control, LLM-RIMSA leverages pre-trained LLMs cross-modal reasoning and few-shot learning capabilities to dynamically optimize RIMSA configurations. Simulations demonstrate that LLM-RIMSA achieves state-of-the-art performance, outperforming conventional DL-based methods in sum rate while reducing training overhead. The proposed framework pave the way for LLM-driven intelligent radio environments.

The utilization of radio frequency (RF) signals for wireless sensing has garnered increasing attention. However, the radio environment is unpredictable and often unfavorable, the sensing accuracy of traditional RF sensing methods is often affected by adverse propagation channels from the transmitter to the receiver, such as fading and noise. In this paper, we propose employing distributed Reconfigurable Intelligent Metasurface Antennas (RIMSA) to detect the presence and location of objects where multiple RIMSA receivers (RIMSA Rxs) are deployed on different places. By programming their beamforming patterns, RIMSA Rxs can enhance the quality of received signals. The RF sensing problem is modeled as a joint optimization problem of beamforming pattern and mapping of received signals to sensing outcomes. To address this challenge, we introduce a deep reinforcement learning (DRL) algorithm aimed at calculating the optimal beamforming patterns and a neural network aimed at converting received signals into sensing outcomes. In addition, the malicious attacker may potentially launch jamming attack to disrupt sensing process. To enable effective sensing in interference-prone environment, we devise a combined loss function that takes into account the Signal to Interference plus Noise Ratio (SINR) of the received signals. The simulation results show that the proposed distributed RIMSA system can achieve more efficient sensing performance and better overcome environmental influences than centralized implementation. Furthermore, the introduced method ensures high-accuracy sensing performance even under jamming attack.

Dynamic spectrum access (DSA) is a technology proposed to address issues, such as spectrum scarcity, inflexible spectrum management, and spectrum waste in wireless communication. This is crucial in supporting the escalating demands of spectrum particularly for Internet of Things (IoT)-based applications. Modeling the spectrum access problem as a Markov decision process (MDP) and incorporating deep reinforcement learning (DRL) have emerged as a cutting-edge approach to tackle this challenge. However, the application of DRL in spectrum access is vulnerability to malicious attacks, posing significant security threats. We introduce both glass-box adversarial attack and closed-box jamming attack over-the-air to assess the susceptibility of DRL-based spectrum access system. The simulation results demonstrate the destructive effect of these attack methods in disrupting the spectrum access process of DRL models, adversely affecting the overall performance of communication systems. Moreover, we propose effective defense mechanisms to mitigate potential threats posed by adversary on DRL-based spectrum access. By incorporating advanced defense mechanisms, we successfully enhance the robustness of the system, ensuring the secure and stable operation of the spectrum access system.

Intelligent reflection surface (IRS) has been used to assist communication by reflection and beamforming where a direct path is not available. Thus, IRS adjusts the wireless channel to enhance the data transmission efficiency with low power consumption. Recently, deep reinforcement learning (DRL) has been exploited in IRS coefficients optimization. IRS can be controlled by DRL to adapt their phase shift to the propagation environment and an expected reflection pattern can be obtained. However, due to the openness of wireless channel and the unexplainability of DRL, it is vulnerable to adversary attacks lauched through wireless channels. In this paper, we investigate a pilot contamination based backdoor attack against DRL based IRS beamforming, where an IRS controlled by an adversary attacker is used to contaminate the channel state information (CSI) during the training phase. The backdoor attack is covert because the adversary need not to know the legitimate pilot sequence and DRL performs well when the adversary IRS keeps inactive. We show that the backdoor attack can reduce the data rate significantly with the adversary IRS. At last, we propose a retraining method agaist the attack to recover the data rate.