Multi-access edge computing provides localized resources within mobile networks to address the requirements of emerging latency-sensitive and computing-intensive applications. At the edge, dynamic requests necessitate sophisticated resource management for adaptive network slicing. This involves optimizing resource allocations, scaling functions, and load balancing to utilize only essential resources under constrained network scenarios. However, existing solutions largely assume static slice counts, ignoring the re-optimization overhead associated with management algorithms when slices fluctuate. Moreover, many approaches rely on simplified energy models that overlook intertemporal resource scheduling and are predominantly evaluated through simulations, neglecting critical practical considerations. This paper presents an incremental cooperative Multi-Agent Deep Deterministic Policy Gradient (MADDPG) algorithm for resource management in dynamic edge slicing. The proposed approach optimizes long-term slicing benefits by reducing delay and energy consumption while minimizing retraining overhead in response to slice variations. Furthermore, we implement an urban-wide edge computing testbed based on OpenStack and Kubernetes to validate the algorithm’s performance. Experimental results demonstrate that our incremental MADDPG method outperforms benchmark strategies in aggregated slicing utility and reduces training energy consumption by up to 50% compared to the re-optimization approach.

The disaggregated architecture of advanced Radio Access Networks (RANs) with diverse X-haul latencies, in conjunction with resource-limited multi-access edge computing networks, presents significant challenges in designing a general model in placing baseband and user plane functions to accommodate versatile 5G services. This paper proposes a novel approach, NetMind+, which leverages Deep Reinforcement Learning (DRL) to determine the function placement strategies in diverse and evolving RAN topologies, aiming at minimizing power consumption. NetMind+ resolves the problem with a maze-solving strategy, enabling a Markov Decision Process with standardized action space scales across different networks. Additionally, a Graph Convolutional Network (GCN) based encoding and an incremental learning mechanism are introduced, allowing features from different and dynamic networks to be aggregated into a single DRL agent. This facilitates the generalization capability of DRL and minimizes the negative retraining impact. In an example with three sub-networks, NetMind+ demonstrates a substantial 32.76% improvement in power savings and a 41.67% increase in service stability compared to benchmarks from the existing literature. Compared to traditional methods necessitating a dedicated DRL agent for each network, NetMind+ attains comparable performance with 70% of the training cost savings. Furthermore, it demonstrates robust adaptability during network variations, accelerating training speed by 50%.

The dis aggregated and hierarchical architecture of advanced RAN presents significant challenges in efficiently placing baseband functions and user plane functions in conjunction with Multi-Access Edge Computing (MEC) to accommodate diverse 5G services. Therefore, this paper proposes a novel approach NetMind, which leverages Deep Reinforcement Learning (DRL) to determine the function placement strategies in RANs with diverse topologies, aiming at minimizing power consumption. NetMind formulates the function placement problem as a maze-solving task, enabling a Markov Decision Process with standardized action space scales across different networks. Additionally, a Graph Convolutional Network (GCN) based encoding mechanism is introduced, allowing features from different networks to be aggregated into a single RL agent. That facilitates the RL agent's generalization capability and minimizes the negative impact of retraining on power consumption. In an example with three sub-networks, NetMind achieves comparable performance to traditional methods that require a dedicated DRL agent for each network, resulting in a 70 % reduction in training costs. Furthermore, it demonstrates a substantial 32.76% improvement in power savings and a 41.67 % increase in service stability compared to benchmarks from the existing literature.

Integrating mobile networks with Non-3GPP networks provides a promising solution to mitigate the wireless RF spectrum scarcity. Despite the maturity of integration technologies, a comprehensive approach for radio resource allocation in highly dynamic and complex multiple Radio Access Technologies (multi-RAT) networks is still lacking. To tackle this challenge, this paper proposes an Access Traffic Management (ATM) system that enhances radio resource allocation during access, transmission, and handover processes. The system features a scalable and concise ATM-supported multi-RAT network architecture, supported by a Deep Deterministic Policy Gradient (DDPG) based Intelligent ATM (IATM) algorithm. To evaluate the proposed system, a Network Simulator 3 (NS3) based network simulation is built with realistic 5G and WiFi modules, interacting with the IATM algorithm in real time for decision making and policy improvement. Numerical improvements of our solution demonstrate its superiority over conventional steering modes. Our solution achieves an increase in resource utilization efficiency by 45% and 70% compared to the Active-Standby and Load-Balance steering modes, respectively. Moreover, it enhances link quality by a factor of three and doubles throughput without incurring any additional costs. Additionally, our solution significantly enhances session stability under conditions involving network size dynamics and UE mobility.

Multi-access edge computing (MEC) has been regarded as one of the essential technologies for mobile networks, by providing computing resources and services close to users, thereby, avoiding extra energy consumption and fitting the low-latency ultra-reliable requirements for emerging 5G applications. Task offloading policy plays a pivotal role in handling offloading requests and maximizing the network computing performance. Most recently developed offloading solutions are designed for instant rewards, therefore, neglecting the long-term computing resource optimization at the edge, which fail to deliver optimized network performance when a significant increase of computing requests appears. In this paper, with the objective of maximizing long-term offloading benefits on delay and energy consumption, task offloading policies are proposed to firstly avoid resource over-distribution through deep reinforcement learning (DRL) based resource reservation and server cooperation, and secondly maximize the average instant reward and the utilization of reserved resources by an optimization-based joint policy consisting of offloading decision, transmission power allocation and resource distribution. The DRL-based joint policy is evaluated in a simulated multi-server edge computing network. Compared to previous solutions, the DRL-based algorithms achieve higher and more reliable overall rewards. Of the implemented three DRL-based algorithms, fully cooperative multi-agent DRL accounts for cooperation between servers, achieving a 70.5% reduction in reward variance and a 13.4% increase in average rewards over 500 continuous operations. Resource balanced policies on long-term rewards help edge networks handle the explosive growth of 5G computing-intensive applications in the future.