In green massive MIMO networks, reducing power consumption (PC) while ensuring user quality of service (QoS) is critical for sustainable operation. To this end, we propose a robust and scalable reinforcement learning framework based on independent proximal policy optimization (IPPO), enabling intelligent base station (BS) control through a three-dimensional configuration of antenna activation, sleep mode transitions, and user offloading. Compared to a non-learning simple energy-saving policy, our proposed IPPO algorithm achieves approximately a 20.3% reduction in PC and a 49% improvement in energy efficiency (EE). In addition, it demonstrates significantly faster convergence and better scalability than multi-agent PPO (MAPPO), reducing convergence time by approximately 75% with 49 BSs and by around 90% with 81 BSs.

Given the advancements in next-generation low Earth orbit (LEO) satellites, there is an expected shift from transparent architectures (acting as radio repeaters) to regenerative architectures (hosting a part or all of the gNodeB (gNB) onboard). Such regenerative architectures enable disaggregation and distribution of radio access network (RAN) functions between the ground and space. Open RAN is a promising approach for non-terrestrial networks and offers flexible function placement through open interfaces. The present study examines three open RAN-based regenerative architectures, namely, Split 7.2× (low-layer physical functions onboard), Split 2 (Layers 1 and 2 onboard), and a gNB onboard the satellite. Handover (HO) management becomes increasingly complex in this disaggregated RAN, particularly for LEO satellites, where the part of the gNB is constantly in motion. The choice of regenerative architecture and its dynamic topology influence the additional HO control signals required between the satellite and ground stations. Using a realistic dynamic LEO constellation model, we analyze the interplay among conditional handover (CHO) delay, computational complexity, and control signaling overhead under different network architectures. Our findings reveal that transitioning from a transparent architecture to Split 7.2× does not reduce CHO delay despite the introduction of additional onboard processing. The gNB onboard the satellite minimizes cumulative CHO delay but demands 55%–70% more computational resources than the Split 7.2× architecture. Conversely, although Split 7.2× is computationally more efficient, it increases the cumulative CHO delay by 25%–30%. Additionally, we observed that under limited onboard processing conditions, only the transparent and Split 7.2× architectures supported delay-sensitive services up to 100 ms. In contrast, under ample processing conditions, gNB was suitable for stringent 50 ms requirements, while Split 2 best supported delay-tolerant services with 200 ms requirements.

The increasing densification of small-cell networks substantially expands cable-based backhaul infrastructure, creating heightened vulnerability to cable link failures. This paper proposes a reconfigurable intelligent surface (RIS)-assisted backup framework that exploits a key insight: during backhaul cable failures, base station (BS) radio components remain functional, enabling wireless backhaul traffic redistribution. Our framework maintains network connectivity by redistributing disconnected BS backhaul traffic to neighboring BSs through RIS-assisted wireless links. To maximize survivability across varying traffic conditions, we formulate a joint optimization problem that maximizes total resolvable backhaul traffic by jointly deciding BS selection, RIS phase shifts, and precoding vectors. The inherent non-convexity arising from coupling and quadratic fractional term is addressed through an alternating optimization algorithm that iteratively solves tractable convex subproblems via quadratic transformation. Comprehensive numerical evaluations demonstrate that the proposed RIS-enhanced framework significantly improves survivability from 58

Electricity consumption in mobile networks is increasing with the continued 5G expansion, rising data traffic, and more complex infrastructures. However, energy management is often handled independently by each mobile network operator (MNO), leading to limited coordination and missed opportunities for collective efficiency gains. To address this gap, we propose a privacy-preserving framework for automated energy infrastructure sharing among co-located MNOs. Our framework consists of three modules: (i) a federated learning-based privacy-preserving site energy consumption forecasting module, (ii) an orchestration module in which a mixed-integer linear program is solved to schedule energy purchases from the grid, utilization of renewable sources, and shared battery charging or discharging, based on real-time prices, forecasts, and battery state, and (iii) an energy source selection module which handles the selection of cost-effective power sources and storage actions based on predicted demand across MNOs for the next control window. Using data from operational networks, our experiments confirm that the proposed solution substantially reduces operational costs and outperforms non-sharing baselines, with gains that increase as network density rises in 5G-and-beyond deployments.

Despite increasing interest in cellular-connected uncrewed aerial vehicles (UAVs), their integration into existing cellular networks poses substantial challenges, including intense interference from UAVs to terrestrial user equipments (UEs) and numerous redundant handovers. To jointly reduce the generated interference and redundant handovers of cellular-connected UAVs while keeping their low transmission delay, we define an optimization problem subject to constraints on total available bandwidth and quality of service (QoS). Then, we formulate the optimization problem as a decentralized partially observable Markov decision process (Dec-POMDP) in the context of a cooperative game. We further develop a collaborative trajectory and handover management scheme using a multi-agent deep reinforcement learning algorithm, specifically the Q-learning with a MIXer network (QMIX) algorithm, to jointly optimize the aforementioned three metrics. Simulation results demonstrate that QMIX significantly outperforms two benchmark schemes: the conventional handover management (CHM) scheme and the independent dueling double deep recurrent Q-network (ID3RQN) scheme. Compared with the CHM scheme, QMIX reduces the delay, interference, and number of handovers for UAVs by an average of 46.9%, 70.0% and 90.5%, respectively. Compared with the ID3RQN scheme, QMIX reduces the three metrics by an average of 90.0%, 43.0% and 41.7%, respectively.

Without requiring operational costs such as cabling and powering while maintaining reconfigurable phase-shift capability, self-sustainable reconfigurable intelligent surfaces (ssRISs) can be deployed in locations inaccessible to conventional relays or base stations, offering a novel approach to enhance wireless coverage. This study assesses the feasibility of ssRIS deployment by analyzing two harvest-and-reflect (HaR) schemes: element-splitting (ES) and time-splitting (TS). We examine how element requirements scale with key system parameters, transmit power, data rate demands, and outage constraints under both line-of-sight (LOS) and non-line-of-sight (NLOS) ssRIS-to-user equipment (UE) channels. Analytical and numerical results reveal distinct feasibility characteristics. The TS scheme demonstrates better channel hardening gain, maintaining stable element requirements across varying outage margins, making it advantageous for indoor deployments with favorable harvesting conditions and moderate data rates. However, TS exhibits an element requirement that exponentially scales to harvesting difficulty and data rate. Conversely, the ES scheme shows only linear growth with harvesting difficulty, providing better feasibility under challenging outdoor scenarios. These findings establish that TS excels in benign environments, prioritizing reliability, while ES is preferable for demanding conditions requiring operational robustness.

Cell-free massive multi-input multi-output (MIMO) promises uniform high performance across the network, but also brings a high energy cost due to joint transmission from distributed radio units (RUs) and centralized processing in the cloud. Leveraging the resource-sharing capabilities of Open Radio Access Network (O-RAN), we propose EARL, an energy-aware adaptive antenna control framework based on reinforcement learning. EARL dynamically configures antenna elements in RUs to minimize radio, optical fronthaul, and cloud processing power consumption while meeting user spectral efficiency demands. Numerical results show power savings of up to 81

Reliable cellular connectivity is critical during disasters, yet base stations must often operate on limited battery backup during large-scale outages. This paper develops a disaster recovery automation (DRA) module for open radio access network (O-RAN) systems that uses reinforcement learning (RL) to manage multi-layer radio access technologies autonomously. The RAT-layer control problem is formulated as a Markov decision process whose state captures battery level, service mode, neighbor signal quality, and site context, while actions activate or deactivate selected RAT layers under emergency conditions. The module is evaluated in an earthquake-calibrated simulator built from real key performance indicator (KPI) traces by comparing Monte Carlo, Q-learning, state-action-reward-state-action (SARSA), Expected SARSA, and a neural-network-based machine learning (ML) agent against All-On and All-Off baselines. Results show that Expected SARSA achieves the longest operational lifetime, 331.5min, extending the All-On baseline of 201.0min by 65%. The findings indicate that adaptive RL-based RAT management can preserve essential service availability while extending battery lifetime, outperforming both no-control operation and naive shutdown without relying on hand-crafted rules.

This paper focuses on energy savings in downlink operation of cell-free massive MIMO (CF mMIMO) networks under dynamic traffic conditions. We propose a multi-agent deep reinforcement learning (MADRL) algorithm that enables each access point (AP) to autonomously control antenna re-configuration and advanced sleep mode (ASM) selection. After the training process, the proposed framework operates in a fully distributed manner, eliminating the need for centralized control and allowing each AP to dynamically adjust to real-time traffic fluctuations. Simulation results show that the proposed algorithm reduces power consumption (PC) by 56.23

Cell-free massive MIMO promises uniformly high performance by combining densely distributed radio units, coherent transmission, and centralized processing. Unlike earlier radio generations, it depends on dense fronthaul connectivity and a virtualized cloud-RAN architecture. In this setting, energy use is no longer driven primarily by active radio components; instead, fronthaul and processing play a dominant role, calling for a fresh perspective on what defines energy efficiency. This work introduces a modular power model that captures the interplay between radios, fronthaul, and cloud processing. The analysis highlights how design choices, such as functional splits and precoding strategies, shape both fronthaul data load and total power consumption. Centralized precoding provides stronger performance with less resource utilization, while flexible activation of radios and processing elements avoids unnecessary overhead. Overall, the energy efficiency of cell-free massive MIMO grows as antennas are more densely distributed across the coverage area, particularly when combined with end-to-end resource allocation.

Densely deployed base stations are responsible for the majority of the energy consumed in Radio access network (RAN). While these deployments are crucial to deliver the required data rate in busy hours of the day, the network can save energy by switching some of them to sleep mode and maintain the coverage and quality of service with the other ones. Benefiting from the flexibility provided by the Open RAN in embedding machine learning (ML) in network operations, in this work we propose Deep Reinforcement Learning (DRL)-based energy saving solutions. Firstly we propose 3 different DRL-based methods in the form of xApps which control the Active/Sleep mode of up to 6 radio units (RUs) from Near Real time RAN Intelligent Controller (RIC). We also propose a further scalable federated DRL-based solution with an aggregator as an rApp in None Real time RIC and local agents as xApps. Our simulation results present the convergence of the proposed methods. We also compare the performance of our federated DRL across three layouts spanning 6–24 RUs and 500–1000 m regions, including a composite multi-region scenario. The results show that our proposed federated TD3 algorithm achieves up to 43.75% faster convergence, more than 50% network energy saving and 37. 4% lower training energy versus centralized baselines, while maintaining the quality of service and improving the robustness of the policy.

Cell-free massive MIMO (CF-mMIMO) combined with integrated sensing and communication (ISAC) is a promising architecture for future 6G networks, enabling new sensing-based applications. However, integrating sensing functionality increases power consumption across the radio, fronthaul, and cloud domains, which is not captured by conventional transmit power optimization approaches. In this paper, we develop a cross-layer end-to-end (E2E) optimization framework for green CF-mMIMO ISAC systems with distributed multi-target detection. We propose a distributed sensing approach in which receive access points (RX-APs) compute local test statistics and forward them to the cloud for aggregation via a weighted combination strategy. We derive maximum a posteriori ratio test (MAPRT) detectors under fully informed (FIS) and partially informed (PIS) scenarios, capturing different levels of side information available at the RX-APs. We formulate a joint optimization problem that minimizes total network power consumption by jointly optimizing transmit power allocation, AP operation modes, communication user and sensing associations, RX-AP assignments, and cloud/fronthaul resources, subject to communication and sensing constraints. The resulting mixed-integer non-convex problem is solved via a two-stage iterative algorithm based on successive convex approximation and penalty-based relaxation. Numerical results demonstrate that the proposed E2E framework significantly reduces total power consumption compared to benchmark schemes, achieving more than 50

Reconfigurable intelligent surfaces (RISs) can greatly improve the signal quality of future communication systems by reflecting transmitted signals toward the receiver. However, even when the base station (BS) has perfect channel knowledge and can compute the optimal RIS phase-shift configuration, implementing this configuration requires feedback signaling over a control channel from the BS to the RIS. This feedback must be kept minimal, as it is transmitted wirelessly every time the channel changes. In this paper, we examine how the feedback load, measured in bits, affects the performance of an RIS-aided system. Specifically, we investigate the trade-offs between codebook-based and element-wise feedback schemes, and how these influence the achievable signal-to-noise ratio (SNR). We propose a novel quantization codebook, tailored for line-of-sight scenarios, that guarantees minimal SNR loss while reducing feedback overhead from linear to logarithmic scaling with the number of RIS elements. We demonstrate the codebook's usefulness over Rician fading channels and extend it to 3D channel geometries with a uniform planar array through joint quantization of elevation and azimuth angles, including scenarios with a non-zero static path. Furthermore, we also analyze the SNR impact of discrete phase shifts and implement an efficient differential feedback scheme that leverages temporal correlation for mobility scenarios. Numerical simulations and analytical analysis are performed to quantify the performance degradation caused by reduced feedback load, shedding light on how efficiently RIS configurations can be fed back in practical systems.

This paper presents a novel symbiotic radio system for integrated sensing and backscatter communication (ISABC) technique that enables signal-domain interference-free coexistence of the primary communication signal and the backscatter communication (BC) signal within the same spectrum. The proposed system design allows simultaneous backscatter devices (BDs) sensing and data transmission without mutual interference by exploiting waveform-domain orthogonality between orthogonal frequency division multiplexing (OFDM) and affine frequency domain multiplexing (AFDM) signals. Specifically, a chirp-based AFDM waveform is adopted due to its inherent processing gain, which enhances the detectability and reliability of the weak backscatter signal while simultaneously supporting high-resolution sensing. Unlike conventional methods that attempt to suppress direct-link interference (DLI), this approach embeds the backscatter transmission within the affine domain while maintaining reliable OFDM-based primary communication. Furthermore, by assigning distinct affine-domain shifts to each backscatter device, the proposed framework inherently suppresses inter-backscatter device interference (IBDI). Comprehensive simulation results demonstrate that the proposed coexistence scheme effectively mitigates interference without affecting the error rate of the primary link and improves the miss-detection probability performance of the BC, making it a promising candidate for future low-power and interferenceresilient systems.

The proliferation of civilian and commercial unmanned aerial vehicles (UAVs) has heightened the demand for reliable radio frequency (RF)-based drone identification systems that can operate under dynamic and uncertain airspace conditions. Most existing RF-based recognition methods adopt a closed-set assumption, where all UAV types are known during training. Such an assumption becomes unrealistic in practical deployments, as new or unknown UAVs frequently emerge, leading to overconfident misclassifications and inefficient retraining cycles. To address these challenges, this paper proposes a unified incremental open-set learning framework for RF-based UAV recognition that enables both novel class discovery and incremental adaptation. The framework first performs open-set recognition to separate unknown signals from known classes in the semantic feature space, followed by an unsupervised clustering module that discovers new UAV categories by selecting between K-Means and Gaussian Mixture Models (GMM) based on composite validity scores. Subsequently, a lightweight incremental learning module integrates the newly discovered classes through a memory-bounded replay mechanism that mitigates catastrophic forgetting. Experiments on a real-world UAV RF dataset comprising 24 classes (18 known and 6 unknown) show effective open-set detection, promising clustering performance under the evaluated noise settings, and stable incremental adaptation with minimal storage cost, supporting the potential of the proposed framework for open-world UAV recognition.

Future uncrewed aerial vehicle (UAV) systems increasingly combine heterogeneous communication technologies, such as low-latency aerial mesh, terrestrial cellular, and satellite links, to improve robustness and coverage. Multipath transport is a natural mechanism for aggregating these links, yet its ability to support real-time UAV services in highly heterogeneous environments remains insufficiently characterized. We present a measurement-driven study based on UAV flight experiments in an integrated network comprising UAV-to-UAV aerial mesh, private cellular, and low Earth orbit (LEO) satellite connectivity. Using Multipath TCP (MPTCP) as a representative lossless, in-order multipath transport framework, we find that aggregation can preserve end-to-end connectivity under severe link outages. However, large round-trip time (RTT) heterogeneity amplifies packet reordering, leading to substantial receiver-side buffering and bursty delivery. In addition, when the available links do not provide sufficient capacity for the offered load, pronounced sender-side buffering emerges. These effects cause real-time streaming to violate delay constraints, including cases where aggregate capacity is sufficient. To interpret these results, we formalize the distinction between connectivity continuity and service continuity and show empirically that maintaining connectivity is necessary but not sufficient for timely real-time delivery in multi-technology UAV networks. The findings motivate multipath designs that explicitly account for delay constraints, rather than optimizing for connectivity alone.

Telecommunications play a pivotal role in shaping today’s interconnected world by fostering global development, supporting seamless information exchange across vast distances, and revolutionizing the interactions between individuals, businesses, and governments. Accessible and reliable communication networks transcend geographical barriers, promoting economic growth, the dissemination of knowledge, and societal connectivity. The integration of artificial intelligence into telecommunications has been transformative, revolutionizing the entire industry by improving system efficiency, allowing new services, and reducing complexity. By leveraging machine learning algorithms, telecommunication operators analyze vast data sets to gain insights into customer behavior, network performance, and market trends. This data-driven approach enhances service efficiency, leading to optimized network deployment, improved customer experience, and targeted marketing strategies. Machine learning’s impact extends to resource allocation optimization. Intelligent management of network resources reduces latency, congestion, and downtime, ensuring enhanced user experiences and increased overall network capacity. This optimization is vital for integrating emerging technologies like the Internet of Things and future generations of mobile systems and promoting sustainability by reducing energy consumption, contributing to a greener future. As technology evolves, the synergy between telecommunications and artificial intelligence will pave the way for a more connected, intelligent, and prosperous future. Given the relevance of this research topic, this paper presents a comprehensive survey of machine learning techniques applied to resource allocation in wireless communication systems. The objective is to guide the scientific and industrial community in the optimized selection of machine learning techniques according to network demands and network resource allocation to be refined. Additionally, it aims to encourage an in-depth discussion regarding the limitations presented in the current literature and future challenges for researchers.

This paper investigates a cell-free massive multiple-input multiple-output enabled multi-access edge computing (termed CF-MEC) system, where multiple users are served by multiple central processing units (CPUs) and their connected access points (APs), both of which are equipped with computation resources. For this system, a dynamic user-centric task offloading scheme is designed to provide seamless and efficient computation services for users. Based on this scheme, the joint optimization of user-centric AP clustering, edge server selection, communication and computation resources is formulated as a long-term problem to minimize the average energy consumption. The formulated problem is complicated non-convex due to the highly coupled time-varying discrete and continuous variables, resulting in high complexity and non-real-time to obtain the optimal solution. To tackle this challenging problem, we propose a multi-layer hierarchical multi-agent deep reinforcement learning (ML-HMADRL) based resource allocation algorithm. Specifically, the proposed algorithm incorporates a hierarchical structure with high, middle, and low-level agents that iteratively train the actor-critic networks to obtain discrete and continuous variables of the formulated problem. To further enhance the training effectiveness by leveraging the CF-MEC system, we design distinct actor-critic networks for the agents at different levels to facilitate centralized training and distributed execution. Simulation results validate the training stability of the proposed algorithm at each level, and demonstrate the superiority of the proposed algorithm over benchmark schemes in terms of the average energy consumption, providing a stable distributed framework for practical implementation in dynamic environments.

As 6G wireless networks transition toward sub-Terahertz (sub-THz) frequencies to satisfy extreme capacity demands, managing the trade-off between massive bandwidth and power consumption becomes a critical design challenge. In this paper, we investigate the fundamental energy efficiency (EE) limits of a dual-band base station site combining a coverage-oriented sub-6 GHz carrier with a capacity-oriented sub-THz carrier. By jointly optimizing hardware parameters and advanced sleep modes via activity factors, we identify four distinct operational regions that govern the EE-optimal behavior across the complete range of data rates. We derive closed-form analytical thresholds that dictate precisely when the sub-THz band should awaken from sleep and how to allocate traffic between the bands in that case. Our results demonstrate that utilizing the sub-THz band is EE-optimal when the power cost of the bandwidth-limited sub-6 GHz band surpasses the static power penalty of activating the sub-THz circuitry. Ultimately, this framework provides mathematically rigorous guidelines for power consumption minimization and sleep-mode management in future green networks.

Wireless fronthaul is a key enabler of flexible and scalable cell-free massive MIMO systems, but its limited capacity poses significant challenges for maintaining high and uniform user performance. In this work, we analyze the performance of a cell-free massive MIMO network with wireless fronthaul under realistic low physical layer functional splits. We propose a joint access and fronthaul resource allocation algorithm that maximizes the minimum user equipment (UE) spectral efficiency while satisfying fronthaul load constraints. Our analysis reveals that power allocation over the wireless fronthaul follows a modified water-filling structure, where the water level is jointly determined by the access and fronthaul channel gains. Furthermore, we show that severe fronthaul limitations not only reduce UE rates but also introduce spatial performance disparities depending on the cloud location. Finally, we demonstrate that split option 8 is impractical under wireless fronthaul constraints, underscoring the importance of dynamic fronthaul bit allocation to reduce fronthaul load and enable efficient system operation.