A Multi-Agent Reinforcement Learning Approach for Capacity Sharing in Multi-Tenant Scenarios

Applications of Machine Learning in Resource Management for RAN-Slicing in 5G and Beyond Networks: A Survey

“…In Table 10, we categorize the reviewed proposals based on the ML technique and the type and number of used algorithms. According to our analysis of the literature review on resource management in RAN slicing, as shown in T-NN 1 Evolutionary Algorithms [65], [66], [68], [69], [67], [70], [72], [71] GA 8 [35], [79], [80], [81], [129], [130], [95] AC 7 [78] MAB 1 [111], [85], [117], [86], [87], [88], [136], [90], [96], [97], [98], [151], [126], [99], [123], [131], [114], [118], [94], [100], [103], [119], [104], [105] DQN 25 [109], [108] Q-Learning 2 Reinforcement Learning [115], [112], [84] Double DQN 3 …”

“…In addition, the problem of frequency interference in inter-cells, intra-cells, inter-RANs, or intra-RANs should be considered. Although some proposals such as [40], [68], [69], [67], [72], [80], [81], [115], [112], [85], [86], [87], [88], [136], [98], [113], [153], [144], [155], [114], [118], [109], [119], [105], [157] have addressed the dynamic power allocation issue, presenting an efficient dynamic power allocation method with a low computational and time complexity is a vital requirement. Possible Solution: Dynamic power allocation in existing proposals has been done in two manners: 1) power allocation with an iterative method using Lagrangian coefficients taking into account channel conditions and interference (e.g., [81], [117], [119]) 2) considering power as an action in online algorithms and selecting the desired power by the agent for each user, evaluating the selected power in accordance with the channel conditions and interference such as [118].…”

“…Few proposals have considered mobility on resource management in RAN slicing (e.g., [71], [110], [111], [136], [153]), but examining the impact of different mobility models on resource management algorithms is an open issue.…”

“…In addition to resource sharing, [136] addresses algorithmic aspects of the resource allocation method. In [136], a multiagent RL (MARL) method based on DQN is proposed for allocating radio and power resources to different tenants in multi-cell environments with aiming to guarantee users' SLAs and maximize resource utilization. In this method, each RL agent is assigned to a tenant, and each tenant centrally decides from observation of the environment.…”

“…In recent years, the use of distributed methods (such as transfer learning and federated learning) in ML has been considered due to the common challenges in centralized methods [79], [80], [81], [82], [84], [85], [86], [87], [88], [89], [90], [91], [92], [83], [93], [94], [95], [110], [111], [112], [113], [114], [115], [116], [117], [118], [136] ✓ ✓ ---- [96], [97], [98], [99], [100], [103], [104], [105], [123], [124], [135] [107], [108], [109], [126], [127], [128], [35], [129], [130] such as centralization, security, privacy, and time and computational complexity. In distributed methods, a ML mode...…”