A Survey of Machine Learning Applications to Handover Management in 5G and Beyond

Abstract

Handover (HO) is one of the key aspects of next-generation (NG) cellular communication networks that need to be properly managed since it poses multiple threats to quality-of-service (QoS) such as the reduction in the average throughput as well as service interruptions. With the introduction of new enablers for fifth-generation (5G) networks, such as millimetre wave (mm-wave) communications, network densification, Internet of things (IoT), etc., HO management is provisioned to be more challenging as the number of base stations (BSs) per unit area, and the number of connections has been dramatically rising. Considering the stringent requirements that have been newly released in the standards of 5G networks, the level of the challenge is multiplied. To this end, intelligent HO management schemes have been proposed and tested in the literature, paving the way for tackling these challenges more efficiently and effectively. In this survey, we aim at revealing the current status of cellular networks and discussing mobility and HO management in 5G alongside the general characteristics of 5G networks. We provide an extensive tutorial on HO management in 5G networks accompanied by a discussion on machine learning (ML) applications to HO management. A novel taxonomy in terms of the source of data to be utilized in training ML algorithms is produced, where two broad categories are considered; namely, visual data and network data. The stateof-the-art on ML-aided HO management in cellular networks under each category is extensively reviewed with the most recent studies, and the challenges, as well as future research directions, are detailed. INDEX TERMS Handover, machine learning, mobility management, fifth generation.

Machine Learning in Beyond 5G/6G Networks—State-of-the-Art and Future Trends

“…RL tries to learn the best policy that would enable the agent to make an optimal decision at any given state of the environment.Figure 3 displays an example of RF. RL algorithms can be categorized to value-based (e.g., Q-learning, SARSA) and policy-based algorithms (e.g., Policy Gradient (PG), Proximal Policy Optimization (PPO) and Actor-Critic (A2C) [29]. • Q-learning: Q-learning is the most common used RL algorithm.…”

A Deep Learning Approach to Location- and Orientation-aided 3D Beam Selection for mmWave Communications

“…At each UT position and orientation, the measurements are captured by sensing the environment with all the possible combinations of precoders U and combiners V at the transceivers. Thus, the entries of R in (11) are set using the RSS measured using all beam pairs in the codebook. The entries of R are used to set the labels used in the collected datasets.…”

“…We consider {8, 8} and {4, 4} UPA at the AP and UT, respectively. After constructing the channel response for each UT location, we calculate the RSS for each beam pair using (11). According to the type of encoding scheme, M, the labels are calculated and stored in D Ξ .…”

“…Reinforcement learning (RL) and deep neural networks (DNNs) are the two areas of machine learning most frequently employed in mmWave beam alignment [11]- [13]. For instance, a RLbased method that uses historical beam management data in a specific environment to reduce the beam training overhead is presented in [14].…”

The Efficient Mobile Management Based on Metaheuristic Algorithm for Internet of Vehicle

“…Therefore, handover between BSs alone is not feasible. According to [ 9 ], we can know that there will be two handover modes in 5G, horizontal and vertical handover. If the level of the next infrastructure is the same, it is called a horizontal handover; otherwise, when the target infrastructure is of a different level, a vertical handover occurs.…”