Network-assisted full-duplex (NAFD) cell-free massive MIMO (CF-mMIMO) systems constitute a promising enabler for supporting dynamic downlink (DL) and uplink (UL) traffic demands in forthcoming sixth-generation (6G) wireless networks. In this paper, we analyze channel aging effects on NAFD CF-mMIMO systems. First, we propose an extended beamforming training scheme to relieve heavy pilot overhead for high-mobility channel estimation, significantly enhancing system performance through cross-link interference (CLI) cancellation. Then, we derive novel closed-form expressions for the UL/DL achievable SEs, enabling a comprehensive analysis of channel aging impacts on spectral efficiency (SE) and energy efficiency (EE) across diverse normalized Doppler shiftfDTsscenarios. Furthermore, we develop a joint optimization framework of transmission interval and power control to balance SE-EE tradeoffs while alleviating channel aging effects. We formulate a mixed-integer multi-objective optimization problem (MOOP), which is subsequently transformed into a tractable single-objective formulation via a weighted ℓpscalarizing method. Based on this analytical foundation, we propose a constrained deep reinforcement learning (DRL) algorithm that integrates a primal-dual optimization strategy with multi-agent deep deterministic policy gradient (MADDPG) for safe policy exploitation. Simulation results validate the accuracy of analytical expressions and demonstrate the superiority of the proposed algorithm over the conventional non-dominated sorting genetic algorithm-II (NSGA-II) in achieving Pareto-optimal SE-EE tradeoffs under channel aging.

Low-orbit satellite Internet of Things (LEO-IoT) has become a key technology for connecting remote terminals due to its wide-area coverage capability. However, the dynamic network topology, resource constraints, and service burstiness pose serious challenges to quality of service (QoS) assurance for delay-sensitive applications. In this paper, we propose an innovative approach to address traffic shaping, resource scheduling and multipath load balancing in LEO-IoT systems by fusing network calculus theory with deep reinforcement learning (DRL) techniques. We propose a QoS-oriented dynamic load balancing strategy to alleviate the local congestion and throughput degradation problems caused by time-varying topology and uneven traffic distribution. We establish an upper bound model for inter-satellite link delay based on network calculus, and determine the worst delay bound for multi-hop links through the least-additive convolution operation of the hopping-beam satellite arrival curves and the inter-satellite link service curves. On this basis, we design a load balancing algorithm for joint beam scheduling and traffic allocation, which dynamically optimizes the beam dwell time and coordinates the multipath weight allocation. In addition, we implement a DRL framework based on proximal policy optimization. Simulation results show that the proposed strategy achieves significant improvements in performance over traditional methods in LEO-IoT environments.

Fully-supervised deep neural networks have achieved remarkable progress in medical image segmentation, yet they heavily rely on extensive manually labeled data and exhibit inflexibility for unseen tasks. Few-shot segmentation (FSS) addresses these issues by predicting unseen classes from a few labeled support examples. However, most existing FSS models struggle to generalize to diverse target tasks distinct from training domains. Furthermore, designing promising network architectures for such tasks is expertise-intensive and laborious. In this paper, we introduce MOE-FewSeg, a novel automatic design method for FSS architectures. Specifically, we construct a U-shaped encoder-decoder search space that incorporates capabilities for information interaction and feature selection, thereby enabling architectures to leverage prior knowledge from publicly available datasets across diverse domains for improved prediction of various target tasks. Given the potential conflicts among disparate target tasks, we formulate the multi-task problem as a multi-objective optimization problem. We employ a multi-objective genetic algorithm to identify the Pareto-optimal architectures for these target tasks within this search space. Furthermore, to mitigate the impact of noisy labels due to dataset quality variations, we propose a noise-robust loss function named NRL, which encourages the model to de-emphasize larger loss values. Empirical results demonstrate that MOE-FewSeg outperforms manually designed architectures and other related approaches.

During Unmanned Aerial Vehicle (UAV) missions, sudden weather changes like strong winds can destabilize UAV flight, making it essential to quickly transmit weather information to improve safety and efficiency. However, these abrupt data surges can lead to network congestion, affecting transmission timeliness and reliability. Current congestion control schemes often prioritize network efficiency but overlook user fairness. This paper introduces an Adaptive Explicit Congestion Control algorithm based on Fairness (AECCF) for handling Microburst traffic, balancing congestion management with fair resource distribution. The AECCF algorithm works as follows: First, the router node monitors its outgoing packet queue and assesses local link congestion state using a double delay threshold. It then calculates and attaches congestion probability, congestion state, and sojourn time to packets, enabling complete congestion feedback to downstream nodes. In cases of burst traffic, the router labels the traffic as heavily congested and activates flow shaping, transferring some packets to a cache queue for staggered forwarding. The consumer then adjusts its interest packet rate based on congestion markings and fair bandwidth allocation, preventing throughput loss from over-adjustment and maintaining network stability and fairness. Finally, the router forwards interest packets based on the proportion of consumer request prefixes on available interfaces, improving data transmission efficiency under sudden traffic. Simulations on ndnSIM show that AECCF achieves higher throughput and fairness in different scenarios compared to existing methods. For dumbbell networks, AECCF reaches 18.64 Mbps throughput with a fairness index of 0.97, and in complex network scenarios, the throughput is 24.9% higher than that of the PCON protocol, and the fairness index is 0.86.

Confronted with the challenges of interruptions and blockages caused by dense obstacles in millimeter-wave (mmWave) communication, we propose to employ a flexible distributed reconfigurable intelligent surface (RIS) assisted massive multiple-input multiple-output (MIMO) system to improve performance in areas with poor coverage. In a scenario featuring multi-antenna user equipments (UEs), we consider the practical additive hardware impairments (AHIs) at transceivers and conduct a comprehensive analysis of their impact on the system performance. Leveraging the pronounced beam directivity inherent in mmWave MIMO, we design phase shifts of RISs and analog beamformers of transceivers to achieve beam alignment. In light of this, we explore a linear minimum mean-square error (LMMSE) equivalent channel estimation method. Furthermore, we derive the closed-form expressions for downlink achievable spectral efficiency (SE) in the presence of AHIs, utilizing statistical channel state information (CSI) and maximum ratio transmission (MRT). Based on the derived closed-form expressions, we propose an efficient power allocation strategy relying on an intelligent algorithm known as primal-dual optimization based deep deterministic policy gradient (PDO-DDPG), which can ensure safe exploration of the agent. Numerical results confirm the accuracy of the derived closed-form expressions, unveil the impact of AHIs on the achievable SE, and verify the effectiveness of the PDO-DDPG based power allocation.

High energy costs and carbon-neutral targets emphasize the economics and sustainability of mobile communications. This paper studies a long-term average energy transaction expenditure minimization problem for smart-grid powered heterogeneous cellular networks (SG-HCNs) where renewable energy is introduced. Renewable energy and wireless channel dynamics evolve over different timescales. Thus, we seek a two-timescale dynamic resource management solution in SG-HCNs, where the real-time joint issue of flow control, power allocation, and energy sharing of renewable energy, and the ahead-of-time two-way energy trading are considered. Based on a two-layer Lyapunov framework, a two-timescale dynamic optimization (TTDO) algorithm is developed for the proposed problem. Specifically, the real-time joint issue is decoupled into two subproblems, addressed by linear programming and successive convex approximation methods. An approximate solution for ahead-of-time two-way energy trading is achieved via the stochastic subgradient approach, where past data of related random events is referred to as prior knowledge that is required but difficult to acquire. Theoretically, the proposed TTDO algorithm can attain an asymptotic optimum and ensure queue stability. Simulation results verify the theoretical analysis and reveal that the proposed TTDO algorithm can obtain lower energy transaction expenditure than benchmarks. Besides, the proposed TTDO algorithm presents an energy-saving property.

The emerging architecture in the next generation of mobile networks leverages the coexistence of Low Earth Orbit (LEO) and Geostationary Orbit (GEO) satellites in a heterogeneous network. This setup not only offers seamless coverage but also enhances user rates. Nevertheless, the efficient allocation of onboard resources, particularly spectrum resources, poses a significant challenge due to their scarcity in such heterogeneous satellite coexistence networks. A practical solution is found in the use of beam hopping (BH) technology. This technology enables multi-beam satellites to serve users using fewer beams than traditional spot-beam systems. This paper proposes a resource allocation strategy for the heterogeneous LEO-GEO coexistence satellite network. We formulate this resource allocation strategy as a joint optimization problem. Due to the complexity of the system arising from the coupling of multiple variables, we break down the original problem into two manageable sub-problems. The first addresses user association, subcarrier, and power allocation and employs a standard convex optimization algorithm for a solution. The second tackles the illuminated beam selection issue, with a genetic algorithm (GA) providing a solution. The effectiveness of our proposed scheme is established through simulation experiments, demonstrating clear performance gains.

This paper proposes a two-timescale online resource management solution for smart-grid supplied heterogeneous cel-lular networks, where grid-energy pre-ordering in advance and bidirectional energy trading in real time are performed. We for-mulate a long-term average energy transaction cost minimization problem considering grid-energy pre-ordering, power allocation, and energy sharing. Leveraging the Lyapunov technique, succes-sive convex approximation method, and stochastic subgradient approach, we develop a two-timescale dynamic optimization (TTDO) algorithm to make online decisions on two time scales. It is theoretically proved that the proposed TTDO algorithm can asymptotically achieve optimality via tuning a control parameter. Numerical tests verify the theoretical results.

Vehicular edge computing is an emerging enabler to support vehicular-based computation-intensive tasks. By reason of the time-varying vehicular wireless environments and the stochastic task generation, the dynamically unbalanced task load distribution among resource-constrained edge infrastructures leads to the performance bottleneck and low efficiency of computation resource utilization. We employ an aerial relay station that can establish relay connections between vehicles and nearby heterogeneous edge infrastructures to relieve this situation. The computation offloading strategy design in the multivehicle multi-edge infrastructure scenario that is closely linked to system latency performance will be particularly complicated. To address this issue, a model-free multi-agent reinforcement learning is adopted, and we propose a practical constraint in the problem formulation. Simulation experiments show that the proposed strategy can guarantee load balancing among edge infrastructures.

This paper investigates a long-term average total energy cost minimization problem via resource management, including admission control, power allocation, and Energy Sharing (ES) of renewable energy in Heterogeneous Cellular Networks powered by Grid-connected Smart Micro Grids (GSMG-HCNs). In GSMG-HCNs, both renewable and grid energy power the base stations. Unlike existing works, we consider the cost of both renewable and grid energy and formulate the power line loss process caused by ES into our model. To solve the proposed problem, we transform it into a real-time issue by the Lyapunov technique. The proposed Cost-Aware Online Resource Management (CAORM) algorithm decouples the real-time issue into two sub-problems, one of which is linear and the other is addressed based on the successive convex approximation approach. We theoretically prove the asymptotic optimality of the CAORM algorithm and a tradeoff between the average total energy cost and the average queue length. Simulation results reveal that the CAORM algorithm outperforms benchmarks in reducing total energy cost and can make appropriate decisions according to different unit costs of renewable energy. Besides, the designed distance-related ES loss rate can help obtain better solutions with lower ES losses.

The authors introduce a novel hybrid precoding algorithm based on Gram–Schmidt orthogonalisation (GSO) in millimetre‐wave massive MIMO systems. Specifically, the columns of array response matrix orthogonalised by the GSO process are considered as a set of candidate analogue precoders, then traditional orthogonal matching pursuit (OMP) is utilised to find the optimal analogue and digital precoders. Since GSO is a recursive process that depends on the order in which the matrix columns are selected. A heuristic solution to the order of columns selection is suggested according to the array response vector along which the fully‐digital precoder has the maximum projection. The proposed algorithm, not only constrained to uniform linear arrays, can avoid the matrix inversion in designing the digital precoder compared to OMP. Simulation results show that the spectral efficiency and bit error rate of the proposed hybrid precoding solutions are close to that obtained with fully digital architectures. Furthermore, the results indicate that the proposed hybrid precoding solutions outperform the orthogonality‐based matching pursuit, which uses the columns of the DFT matrix as a set of candidate analogue precoders.

Distributed phased arrays based multiple-input multiple-output (DPA-MIMO) is a recently proposed highly reconfigurable architecture enabling both spatial multiplexing and beamforming in millimeter-wave (mmWave) systems. In this work, we focus on coping with the hybrid precoding for the wideband DPA-MIMO system with orthogonal frequency division multiplexing (OFDM) modulation. More specifically, we propose an alternating direction method of multipliers (ADMM) enabled hybrid precoding approach based on an alternating optimization framework, abbreviated to ADMM-AltMin, for such cooperative array-of-subarrays structures. Simulation results show that the proposed ADMM-AltMin method achieves favourable performance with practical quantization of phase shifters taken into account.