Near-field simultaneously transmitting and reflecting reconfigurable intelligent surface (STAR-RIS) systems enable full-space coverage, but dense multiuser operation requires active-user grouping under practical stream or RF-chain constraints. Gain-based or proportional-fair scheduling may over-serve strong users, while random selection improves service regularity but ignores channel and region information. This work proposes a randomized deficit-aware user grouping (RD-FRUG) scheme for energy-splitting STAR-RIS-aided near-field multiuser systems. RD-FRUG jointly accounts for service deficit, long-term rate imbalance, transmission/reflection region balance, and inter-user channel correlation. Given the selected group, the STAR-RIS energy-splitting coefficients, passive phase profile, and BS precoder are updated with low computational overhead. Simulation results under near-field channels with distance-dependent pathloss show that the proposed heuristic provides a favorable sum-rate–fairness tradeoff under the considered settings. It approaches the Jain’s fairness index of random selection, achieves higher sum-rate than random selection, and improves the fifth-percentile user rate over gain-greedy, proportional-fair greedy, and region-balanced baselines.
This paper studies coverage imbalance in reconfigurable intelligent surface (RIS)-assisted plateau LoRa multi-user downlink transmission. Terrain blockage, propagation distance, and scattering diversity cause heterogeneous channel qualities, which may leave weak users with high bit error rate (BER) and outage risk even when the average link quality is improved. To address this issue, a weak-coverage user compensation method is proposed. Users are first classified according to initial channel quality, and inverse-quality weights are constructed to increase the influence of weak users. A weighted multi-user contribution metric is then used to select dominant RIS elements, and a local candidate phase codebook is generated around a weighted reference phase. Finally, a coverage-fairness-aware score selects the discrete RIS phase vector. Simulations show that the proposed method reduces average BER, outage probability, weak-user BER, and inter-user performance variation while reducing the candidate search space from QN to ${L^{{K_d}}}$.
Simultaneous transmitting and reflecting reconfigurable intelligent surface (STAR-RIS) is crucial to achieve full-space coverage in next-generation wireless networks. However, optimizing resource allocation in STAR-RIS-assisted systems to balance the system sum rate with user fairness, especially in the presence of imperfect channel state information (CSI), remains a significant challenge. To address this issue, this work investigates resource allocation in an STAR-RIS-assisted multiple-input single-output system under imperfect CSI and proposes a novel method based on the deep reinforcement learning (DRL) framework to solve this problem. Specifically, the DRL framework is utilized to solve the maximization problem of the weighted sum of Jain’s fairness index and the normalized system sum rate, and a segmented training strategy is employed to decouple the complexity of the original joint optimization problem. The simulation results demonstrate that the proposed solution achieves a flexible trade-off between the system sum rate and user fairness. Moreover, it effectively mitigates the performance degradation caused by imperfect CSI, thereby ensuring robust system performance.
Lifan Zeng, Yuyang Peng, Mohammad Meraj Mirza et al.· IEEE Wireless Communications...· 0 citations
In this paper, we propose a CSI-free distributed user scheduling scheme for intelligent reflecting surface (IRS)-aided multi-user systems. Building upon adaptive conditional sample mean (A-CSM), originally developed for blind beamforming in IRS-aided systems, we exploit the first-stage A-CSM output as a local scheduling metric without explicit channel state information (CSI). The obtained metric exhibits a non-negative and rightskewed distribution, which can be effectively approximated by a Gamma distribution. Based on this observation, the proposed CSI-free-D-Gamma scheme first maps the local metric into a normalized access variable using a moment-matched Gamma CDF. Then, unlike LUT- and Uniform-based baselines that terminate with equal-width slot-region mapping, the proposed Gamma scheme further adjusts the discrete slot-index regions in the normalized domain by considering both the Gammainduced rate contribution and the earliest-singleton collision behavior. Under the considered symmetric small-scale fading setting, numerical results show that the proposed CSI-free-D-Gamma scheme reduces the collision probability and improves the average achievable rate compared with LUT-based, fixed minmax uniform mapping, and Random Scheduling baselines.
Jaeheon Park, Junsu Kim, Su Min Kim· International Conference on...· 0 citations
This paper considers priority-aware partial computation offloading in an uplink mobile edge computing (MEC) network. Devices assigned to different groups occupy orthogonal subbands, whereas devices within each group use power-domain non-orthogonal multiple access (NOMA) with successive interference cancellation. Task-input size determines the transmitted and processed workload, while queue backlog and application urgency determine the service weight. The Gaussian multiple-access-channel rate region is convex, but the complete allocation problem is not jointly convex in the adopted variables because the offloaded workload is coupled with reciprocal transmission rate and reciprocal edge-CPU allocation. A structure-exploiting block-coordinate projected-gradient method is developed. It combines exact finite-candidate offloading updates, an exact edge-CPU allocation bounded below by deadline feasibility and above by local-path saturation, and an analytical projected power step with Armijo backtracking. For eight users at 23 dBm, pairwise group-based NOMA reduces the weighted delay–energy cost and device energy by 6.18% and 23.44%, respectively, relative to orthogonal access. Queue-aware weighting reduces upper-backlog-quartile delay by 2.69 ms (95% confidence half-width: 0.78 ms) while increasing lower-quartile delay by 8.34 ms (half-width: 2.07 ms). In a paired 15-iteration ablation, generic projected block-coordinate updates have a cost ratio of 1.0098 (half-width: 0.0086) relative to the structured method. A hybrid deep deterministic policy-gradient policy, evaluated over five training seeds, has an 11.77% higher cost while requiring 0.84% of the median online decision time. Of 432 allocations, 392 satisfy the residual-qualified stopping tests and 40 are explicitly reported as iteration-safeguard terminations.
Jamil K. J. Bataineh, Ahlam Jawarneh, K. Hayajneh et al.· Italian National Conference...· 0 citations
This letter proposes an adaptive element grouping method for discrete-phase reconfigurable intelligent surfaces (RISs) serving a high-mobility user equipment (UE) to maximize the average net spectral efficiency. A drift-aware channel state information (CSI) aging model is derived for the RIS-to-UE link, enabling pilot refreshing, CSI prediction, and error covariance characterization. With pilot-training and RIS-reconfiguration overhead included, a two-timescale design is developed: slow-timescale grouping exploits spatial correlation, whereas fast-timescale beamforming and discrete-phase optimization use predicted CSI while retaining residual inter-group coupling. Simulations show higher net spectral efficiency than existing grouping schemes, especially under high mobility and large-scale RISs.
Yilin Wu, Shuqi Tang, Kun Chai et al.· IEEE Communications Letters· 0 citations
Millimeter-wave (mmWave) communication systems are vulnerable to severe attenuation, blockage-induced LOS/NLOS transitions, and time-varying co-channel interference. This paper develops a lightweight distributed power-allocation framework in which each base station independently updates a tabular Q-learning policy using locally observable blockage-ratio, serving-distance, and aggregate-interference information. The proposed state-dependent dynamic reward is recalculated at every decision step, and its coefficients vary explicitly with the instantaneous blockage ratio, QoS satisfaction ratio, and normalized interference level. All learning-based and non-learning baselines are evaluated using the same topology realizations, blockage and mobility traces, and random seeds. Under the reconstructed simulation settings, the proposed method achieves performance comparable to fixed Q-learning while retaining a transparent blockage-aware state and low-complexity distributed implementation. DQN, greedy, and uniform power achieve higher raw capacity or QoS in the considered small-scale network. Results from 30 paired runs with 95% confidence intervals, together with ablation, sensitivity, beam-misalignment, and overhead analyses, clarify the empirical benefits, limitations, and deployment scope of the proposed method. The study focuses on power control after beam establishment; joint beam tracking and power allocation remain outside the present scope.
Zhuoning Yang, Ziwei Chen· Italian National Conference...· 0 citations