Two major trends are driving the wireless industry to develop fifth-generation cellular networks: the rapid increase in demand for wireless broadband services that require much higher data rates, and much higher capacity networks that can provide video and other resource-intensive services; and Internet of Things (IoT) services, which encourage the need to connect devices en masse, as well as the need for ultra-low latency.
At the same time, with the development of cellular networks, new and more advanced network architectures for data transmission and management appear. However, there are still a number of unresolved issues and problem areas that need to be addressed and addressed accordingly.
For example, in recent decades, the model of cloud capacity and computing has been widely used in the field of Information Technology (IT). However, despite its success, the introduction of cloud technologies must overcome several of the challenges they faced with the advent of the IoT and 5th generation networks. First and foremost, it is the rapid growth in the number of IoT devices. Second, there is a large physical distance between IoT devices and cloud data centers, leading to long delays. Third, applications deployed in the cloud find it difficult to adapt to changes in local conditions of distributed mobile devices.
To address these cloud-related issues, recent research has introduced a similar concept – Mobile Edge Computing or Multi-access Edge Computing (MEC).
It has been shown that computing power at the cell boundary is a rather promising concept in the context of the development of the Internet of Things, especially to support delay-dependent applications. The main problem is the task of placing relevant services, which concerns the decision in which place to place several applications according to their requirements for the quality of QoS services, on the one hand, and the computational availability of the resource, on the other hand.
To do this, the author developed a method for optimizing the placement of scalable services on the distributed computing resources of the cellular network, which consists in the consistent use of the boundary computation model, a generalized model of the cellular network and a heuristic solution based on genetic algorithms.
Processing downloaded tasks based on their delays requires deciding on the MEC server to load each task, determining computing resources for distribution in IoT applications that will handle the tasks, in addition to determining the order in which the downloaded tasks should be processed by each application. Solving the above three tasks greatly affects the acceptance of tasks into the network, as they directly affect some of the delays they experience.
Therefore, it is to solve the above problems that the method of dynamic unloading and scheduling of tasks for the boundary computer systems of the cellular operator was improved by forming the problem of mixed integer programming and its solution by Benders decomposition.
Further, the power control of the radiated power of radio transmitters in a cellular multiuser system under the influence of interference was investigated. In this case, the difference between this system and free from interference is that the power control in a multi-user system under interference is mainly distributed through the interaction between different users of mobile devices, which greatly complicates the task of power management.
Therefore, the method of controlling the radiated power of mobile devices during the unloading of tasks in the distributed computer system of boundary calculations of the cellular operator was improved by sequential use of the model to assess the need for unloading tasks in the mobile network and control the radiated power of radio devices in interference channels. based on game theory.
The theoretical results obtained in the dissertation research opened the possibility to identify and propose new practical ways to increase the efficiency of the subsystem of cellular base stations during their implementation in Ukraine based on the use of new methods of network management, data transmission.
