Securing IRS-SWIPT Deployment in O-RAN

The Open Radio Access Network (O-RAN) revolutionizes the telecom domain to attain open, interoperable, and software-driven network architecture. This fosters innovation, increases flexibility, and enhances cost-efficiency in next-generation networks (NGN). In NGN, two of the proposed technologies are the Intelligent Reflecting Surface (IRS) and Simultaneous Wireless Information and Power Transfer (SWIPT), which combines data transmission and energy harvesting (EH). In this study, we investigate the scenario of securing the IRS-SWIPT communication downlink scheme in O-RAN through the physical layer. We aim to maximize the secrecy sum rate for legitimate users and efficiently allocate energy to them while adhering to energy harvesting and O-RAN’s computational resource constraints. Further, we develop a deep reinforcement learning (DRL) method, specifically, deep deterministic policy gradient (DDPG) named SecureLink-DDPG, to determine the optimal optimization parameters that maximize the secure sum rate. Simulation results validate the effectiveness of the proposed DRL solution in ensuring secure communication for legitimate users while meeting the specified constraints. The SecureLink-DDPG approximates the optimum in small-scale scenarios and substantially outperforms a discretized grid-based search (GBS) technique, while operating with a fraction of the computational resources required by the GBS.