On the Robustness of RSMA to Adversarial BD-RIS-Induced Interference

TL;DR

This work addresses the vulnerability of downlink RSMA in multi-user MISO to adversarial BD-RIS-induced interference during channel acquisition. It develops two attack schemes (random and aligned) for fully and group-connected BD-RIS architectures and proposes a Takagi-factorization-based projection and a QCQP-based optimization to generate valid BD-RIS reflection matrices. By introducing rate-degradation and robustness-index metrics, the study shows that RSMA is highly susceptible under perfect CSI yet substantially more robust than SDMA when CSI is imperfect, with robustness improving as transmit power grows. The BD-RIS concept markedly increases adversarial effectiveness relative to diagonal RIS, especially when the uplink training is manipulated, highlighting the need for defense mechanisms such as attack detection and adaptive RSMA strategies in future 6G systems.

Abstract

This article investigates the robustness of rate-splitting multiple access (RSMA) in multi-user multiple-input single-output (MISO) systems to interference attacks against channel acquisition induced by beyond-diagonal RISs (BD-RISs). Two primary attack strategies, random and aligned interference, are proposed for fully connected and group-connected reconfigurable intelligent surface (RIS) architectures. Valid random reflection coefficients are generated exploiting the Takagi factorization, while potent aligned interference attacks are achieved through optimization strategies based on a quadratically constrained quadratic program (QCQP) reformulation followed by projections onto the unitary manifold. Our numerical findings reveal that, when perfect channel state information (CSI) is available, RSMA behaves similarly to space-division multiple access (SDMA) and thus is highly susceptible to the attack, with BD-RIS inducing severe performance loss and significantly outperforming diagonal RIS. However, under imperfect CSI, RSMA consistently demonstrates significantly greater robustness than SDMA, particularly as the system's transmit power increases.

Figures (10)

• Figure 1: An adversarial entity configures a BD-RIS to launch interference attacks on multiple RSMA users during downlink data transmission.
• Figure 2: Illustrative comparison of different RIS architectures with four elements: (a) single-connected; (b) group-connected; and (c) fully connected. $Z_i$ represents the reconfigurable self-impedance of element $i$, while $Z_{ij}$ denotes the mutual impedance between elements $i$ and $j$, with $i, j \in \{1, \ldots, 4\}$.
• Figure 3: Simplified diagram of the RIS-induced attack strategy.
• Figure 4: Impact of the RIS mode in the uplink training phase: (a) average sum rate of RSMA across different RIS architectures; and (b) comparison between RSMA and SDMA under the fully connected RIS. Results are shown for the aligned interference attack.
• Figure 5: Performance degradation and robustness analysis: (a) average robustness index comparing RSMA and SDMA under the fully connected RIS; and (b) average sum-rate degradation of RSMA under random and aligned attacks across different RIS architectures.