Li-Hsiang Shen

Multi-functional reconfigurable intelligent surface (MF-RIS) is conceived to address the communication efficiency thanks to its extended signal coverage from its active RIS capability and self-sustainability from energy harvesting (EH). We investigate the architecture of multi-MF-RISs to assist non-orthogonal multiple access (NOMA) downlink networks. We formulate an energy efficiency (EE) maximization problem by optimizing power allocation, transmit beamforming and MF-RIS configurations of amplitudes, phase-shifts and EH ratios, as well as the position of MF-RISs, while satisfying constraints of available power, user rate requirements, and self-sustainability. We design a parametrized sharing scheme for multi-agent hybrid deep reinforcement learning (PMHRL), where the multi-agent proximal policy optimization (PPO) and deep-Q network (DQN) handle continuous and discrete variables, respectively. The simulation results have demonstrated that PMHRL has the highest EE compared to other benchmarks, including cases without parametrized sharing and pure PPO/DQN. Moreover, the proposed multi-MF-RISs-aided downlink NOMA achieves the highest EE compared to cases of no-EH/amplification, traditional RISs, and deployment without RISs/MF-RISs under different multiple access.

Low Earth orbit (LEO) satellite constellations are emerging as a core enabler of sixth-generation (6G) wireless systems, providing global coverage and high-capacity connectivity. However, dense multi-beam LEO deployments introduce severe inter-beam and inter-satellite interferences, while rapid orbital motion results in frequent handovers and highly dynamic channels that challenge the real-time optimization. To address these issues, this paper has proposed a digital twin (DT)-driven multi-LEO network architecture that integrates ray tracing-based virtual simulations with intelligent on-orbit control. Within this framework, we develop a DT-driven Efficient handover and Multi-Agent Twin delayed deep deterministic policy gradient (DEMAT) scheme, which jointly optimizes beam training, handover, power allocation, and beamwidth adaptation to maximize energy efficiency (EE) while satisfying user throughput requirements. DEMAT leverages bidirectional DT-LEO parameter exchange and federated learning-enhanced agents for cooperative and low-latency resource management. Extensive simulations validate its convergence and scalability under diverse network configurations such as various time frame intervals and user densities. Notably, DEMAT achieves up to a 59.6% EE improvement over the Deep Deterministic Policy Gradient (DDPG) baseline and more than 40% of EE compared to the other DT-based benchmarks, demonstrating superior adaptability and coordination for the next-generation non-terrestrial networks.

A space–air–ground integrated network (SAGIN) for Internet of Things (IoT) network architecture is investigated, empowered by multifunctional reconfigurable intelligent surfaces (MF-RISs) capable of simultaneously reflecting, amplifying, and harvesting wireless energy. The MF-RIS plays a pivotal role in addressing the energy shortages of low-Earth orbit (LEO) satellites operating in the shadowed regions, while accounting for both communication and computing energy consumption across the SAGIN nodes. To maximize the long-term energy efficiency (EE) of IoT devices, we formulate a joint optimization problem over the MF-RIS parameters, including signal amplification, phase-shifts, energy harvesting (EH) ratio, and active element selection, as well as the SAGIN parameters of beamforming vectors, high-altitude platform station (HAPS) deployment, IoT device association, and computing capability. The formulated problem is highly nonconvex and nonlinear, which contains mixed discrete-continuous parameters. To tackle this, we conceive a compressed hybrid twin-model enhanced multiagent deep reinforcement learning (CHIMERA) framework, which integrates semantic state-action compression and parametrized sharing under hybrid reinforcement learning to efficiently explore suitable complex actions. The simulation results have demonstrated that the proposed CHIMERA scheme substantially outperforms the conventional benchmarks, including fixed-configuration or nonharvesting MF-RIS, traditional RIS, and no-RIS cases, as well as centralized and multiagent deep reinforcement learning (DRL) baselines in terms of the highest EE. Moreover, the proposed SAGIN-MF-RIS architecture in IoT network achieves superior EE performance due to its complementary coverage, offering notable advantages over either standalone satellite, aerial, or ground-only deployments.

This letter proposes a novel six-dimensional movable metasurface (6DMM)-assisted downlink non-orthogonal multiple access (NOMA) system, in which a conventional base station (BS) equipped with fixed antennas serves multiple users with the assistance of a reconfigurable intelligent surface (RIS) with six-dimensional spatial configurability. In contrast to traditional RIS with static surface, the proposed 6DMM architecture allows each element to dynamically adjust its position and orient the whole metasurface in yaw-pitch-roll axes, enabling both spatial and electromagnetic controls. We formulate a sum-rate maximization problem that jointly optimizes the BS NOMA-based beamforming, phase-shifts, element positions, and rotation angles of metasurface under constraints of NOMA power levels, unit-modulus of phase-shifts, power budget, inter-element separation and boundaries of element position/orientation. Due to non-convexity and high-dimensionality, we employ a probabilistic cross-entropy optimization (CEO) scheme to iteratively refine the solution distribution based on maximizing likelihood and elite solution sampling. Simulation results show that the proposed CEO-based 6DMM-NOMA architecture achieves substantial rate performance gains compared to 6DMM sub-structures, conventional static RIS, and other multiple access mechanisms. It also highlights the effectiveness of CEO providing probabilistic optimization for solving high-dimensional scalable metasurface.

This paper investigates a simultaneously transmitting and reflecting reconfigurable intelligent surface (STARS)-aided integrated sensing and communication (ISAC) systems in support of full-space energy-efficient data transmissions and target sensing. We formulate an energy efficiency (EE) maximization problem that jointly optimizes a dual-functional radar-communication (DFRC)-empowered base station (BS), considering its ISAC-based active beamforming, along with the passive STARS beamforming configurations of amplitudes, phase shifts, quantization levels, and element selection. Furthermore, relaxed/independent/coupled STARS are considered to examine architectural flexibility. To tackle the non-convex and mixed-integer problem, we propose a joint active-passive beamforming, quantization and element selection (AQUES) scheme based on the alternating optimization: Lagrangian dual and Dinkelbach’s transformation tackle fractional equations, whereas successive convex approximation (SCA) convexifies the non-solvable problem; Penalty dual decomposition (PDD) framework and penalty-based convex-concave programming (PCCP) procedure solve amplitude and phase-shifts with the equality constraint; Heuristic search iteratively decides the optimal quantization level; Integer relaxation deals with the discrete element selection variables. Simulation results demonstrate that STARS-ISAC with the proposed AQUES scheme significantly enhances EE while meeting communication rates and sensing quality requirements. The coupled STARS further highlights its superior EE performance over independent and relaxed STARS thanks to its reduced hardware complexity. Moreover, AQUES outperforms existing configurations and benchmark methods in the open literature across various network parameters and deployment scenarios.