Multi-Agent SAC Enabled Beamforming Design for Joint Secret Key Generation and Data Transmission

2026-03-13

AI Summary

This work addresses the challenge of simultaneously achieving high data rates and high physical-layer key generation rates in TDD multi-antenna systems under eavesdropping attacks. The authors propose a joint optimization framework that first models the legitimate channel using a time-correlated autoregressive process, transforming the problem into a temporally coupled non-convex optimization. To handle partial observability, they design a multi-agent soft actor-critic (SAC) algorithm augmented with LSTM modules to predict eavesdropper behavior. This approach represents the first integration of multi-agent reinforcement learning with LSTM for beamforming optimization, enabling dynamic trade-offs between secure communication and key generation. Simulation results demonstrate that the proposed scheme outperforms existing benchmarks in balancing these competing objectives and exhibits strong robustness against intelligent eavesdropping and channel observation uncertainty.

Technology Category

• Multiagent Systems: Adversarial Agents
• Search and Optimization: Learning to Search
• Planning, Routing, and Scheduling: Optimization of Spatio-temporal Systems

Application Category

• Security and Privacy: Large-scale security measurements
• Economics, Online Markets and Human Computation: Uses of LLMs and GenAI for marketplace design, bidding, and strategic interactions
• User Modeling, Personalization and Recommendation: Attacks and countermeasures in recommendation systems

Abstract

Physical layer key generation (PLKG) has emerged as a promising solution for achieving highly secured and low-latency key distribution, offering information-theoretic security that is inherently resilient to quantum attacks. However, simultaneously ensuring a high data transmission rate and a high secret key generation rate under eavesdropping attacks remains a major challenge. In time-division duplex (TDD) systems with multiple antennas, we derive closed-form expressions for both rates by modeling the legitimate channel as a time-correlated autoregressive (AR) process. This formulation leads to a highly nonconvex and time-coupled optimization problem, rendering traditional optimization methods ineffective. To address this issue, we propose a multi-agent soft actor-critic (SAC) framework equipped with a long short-term memory (LSTM) adversary prediction module to cope with the partial observability of the eavesdropper's mode. Simulation results demonstrate that the proposed approach achieves superior performance compared with other benchmark algorithms, while effectively balancing the trade-off between secret key generation rate and data transmission rate. The results also confirm the robustness of the proposed framework against intelligent eavesdropping and partial observation uncertainty.

Problem

• Physical layer key generation
• Beamforming
• Multi-agent reinforcement learning
• Eavesdropping
• Time-division duplex

Innovation

• Multi-Agent SAC
• Physical Layer Key Generation
• Beamforming Design
• LSTM-based Adversary Prediction
• Time-Correlated Channel Modeling