The evolution of 5G and Beyond (B5G) networks requires innovative solutions to address the growing complexity of network management and enhance Energy Efficiency (EE). Sleep Mode (SM) has emerged as an essential strategy for minimizing energy consumption in densely deployed networks, particularly in Base Stations (BS), significantly improving Energy Saving (ES) in 5G/B5G network. By dynamically deactivating underutilized BS components, SM can be effectively applied across various 5G infrastructures. In addition, the application of Artificial Intelligence (AI) and Self Organizing Networks (SON) provides advanced tools for automatic and intelligent system reconfiguration, further enhancing ES in B5G networks. This paper explores ES approaches centered on SM, providing an overview of their application in 5G architectures and the role of AI algorithms and the SON paradigm in improving ES. Finally, we discuss potential ES advancements for future networks, including the transition to 6G, highlighting the importance of intelligent and adaptive energy management in next-generation networks.

The advent of 5G and beyond networks is envisioned to support lower latency, higher data rates, and wider connectivity than previous cellular network generations. However, given the denser deployment of base stations (BSs) to accommodate such improvements, this results inevitably in a significant and unsustainable increase in the network’s energy consumption. Sleep Control (SC), which allows switching off some BS hardware components during light-traffic time, is considered a viable solution for greener and more energy-efficient Radio Access Networks (RAN). However, the optimization of SC is a highly challenging large-scale network combinatorial problem that depends on dynamic wireless channel conditions and varying traffic demands with stringent Quality-of-Service (QoS) requirements. Driven by the benefits and efficiency of Deep Reinforcement Learning (DRL), which has been successfully applied to multiple wireless network optimization problems, this paper investigates DRL approaches addressing sleep control in 5G and beyond RAN. To this end, we propose a taxonomy to classify the related literature. Then, we provide an overview of the different components of the Markov Decision Process (MDP) modeling the sequential decision-making of sleep control and the applied DRL algorithms. Finally, we highlight the main challenges in existing works and suggest novel strategies to address them.

Considering the stringent delay requirements in some use cases of V2X networks, relying only on cloud computing to execute some tasks of vehicles is sometimes infeasible. Meantime, increasing physically the number of onboard resources will lead to an increase in the cost of vehicles during the manufacturing process. Currently, a base station (BS) equipped with an edge server is adopted to shift computing capabilities close to vehicles in order to fulfil time-sensitive application requirements. However, in bursty traffic or disaster caused by some social events, Non-Line-of-Sight (NLoS) communication will be created and the BS will not be able to serve efficiently the vehicles in such conditions. In this context, considering their flexibility and the Line-of-Sight (LoS) communication that they provide, unmanned aerial vehicles (UAVs) can be used to address the above-mentioned issue. More specifically, by dispatching edge server-mounted UAVs, they can assist the BS to fulfil the requirements of delay-sensitive use cases in V2X. Therefore, the resources equipped by UAVs and the BS need to be efficiently managed. Since using a centralized system will result in (1) high energy consumption at the UAVs level and high delay due to the communication between the edge servers and the centralized server, and (2) poor scalability, we formulate this problem as multi-agent learning and solve it with a deep deterministic policy gradient (DDPG) with the aim of maximizing the number of offloaded tasks while fulfiling the quality-of-service (QoS) of such tasks.

Internet of Things (IoT) devices face significant challenges due to their limited capabilities and battery life especially in hazardous regions. Furthermore, the availability of telecommunication infrastructures in these areas is insufficient or event non-existent. As a result, transmitting the data collected by IoT devices becomes a formidable undertaking. Moreover, the limited lifespan of IoT devices, which rely on batteries for operation, exacerbates the situation. Recent research has explored the potential of unmanned aerial vehicles (UAVs) as effective means to facilitate air-ground communications and gather data from IoT devices deployed in harsh environments. However, the ability of UAVs to provide sustainable and reliable services is hindered by their dependence on limited battery power. Therefore, the UAV mobile edge/fog computing paradigm offers low latency communication, computation and storage capabilities for offloading intensive tasks of IoT devices. In addition, the utilization of current technology, such as Wireless Power Transfer (WPT), presents itself as a viable means to facilitate the capability of unmanned aerial vehicles (UAVs) to acquire power without the need for physical connection through ground base stations (GBS). Furthermore, this technology can also be employed to wirelessly replenish the energy of Internet of Things (IoT) devices situated in distant regions. In this study, we examine a scenario where UAVs providing offloading and energy services to UEs, making binary decisions for data offloading and continuous decisions for energy allocation. Using a federated deep reinforcement learning (FDRL) approach, we address privacy concerns and data leakage. The network simulation aims to determine the optimal number of UAVs a single fog node can serve and assesses the maximum UEs each UAV can handle. Our proposed framework’s performance is compared to single-agent and traditional central training, using mean reward as a metric. We evaluate mean delay over training iterations, comparing with central learning, and assess mean energy consumption against traditional central learning.

The reduction of operational expenditure has become a major concern for telecommunication operators and Internet service providers. In this paper, we propose an energy aware routing (EAR) in Carrier Ethernet networks operating with Shortest Path Bridging (SPB) protocol with equal cost multi-path (ECMP). Since traffic load has no influence on power consumption of Carrier Ethernet network elements, the conventional solution to reduce power consumption is to find the maximal set of network elements that can be turned off on so that the network performance is not deteriorated. To tackle this optimization problem, we propose an exact method based on Mixed Integer Linear Programming (MILP) formulation, called SPB energy-aware routing (SPB-EAR). Since SPB-EAR is proved to be NP-hard, we present two heuristics algorithm suitable for large-sized networks, called Green SPB (G-SPB) and Fast Greedy SPB (FG-SPB). In this work, we consider that a connection between two nodes is represented by bundled link consisting of multiple cables. Experimentations on four realistic network topologies show that G-SPB and FG-SPB can save almost as much power consumption as SPB-EAR.

Energy optimization has become a crucial issue in the realm of ICT. This paper addresses the problem of energy consumption in a Metro Ethernet network. Ethernet technology deployments have been increasing tremendously because of their simplicity and low cost. However, much research remains to be conducted to address energy efficiency in Ethernet networks. In this paper, we propose a novel Energy Aware Forwarding Strategy for Metro Ethernet networks based on a modification of the Internet Energy Aware Routing (EAR) algorithm. Our contribution identifies the set of links to turn off and maintain links with minimum energy impact on the active state. Our proposed algorithm could be a superior choice for use in networks with low saturation, as it involves a tradeoff between maintaining good network performance and minimizing the active links in the network. Performance evaluation shows that, at medium load traffic, energy savings of 60% can be achieved. At high loads, energy savings of 40% can be achieved without affecting the network performance.

Quality of service (QoS) is still a crucial issue that deals with the IEEE 802.16 network performance. Scheduling is a key component of the MAC IEEE 802.16 layer which ensures QoS for different service classes. In this paper, we give a detailed simulation study for some scheduling algorithms by evaluating their performance in order to support the QoS classes. We propose a new WiMAX scheduling algorithm for real time applications, called minimum Delay maximum Signal to Interference Ratio (mDmSIR) which is suitable for real time Polling Service (rtPS) class. Our idea is to take into account the maximum latency required by the rtPS connections and their radio states. Our proposed scheme could be the better choice for variable-size real-time connections as it is a tradeoff between maintaining a higher throughput and minimizing the mean average delay. It gives a better results that show an improvement in term of delay. The simulation is carried out via the Network Simulator NS-2.34.