Optimization of load on facilities based on intelligent control systems

DOI: 10.31673/2412-9070.2025.020614

Abstract

The main task in designing the networks of the future is to create an intelligent control system to optimize the load on the objects of the information communication network. With the increase in the spectrum and number of services, the load on the network increases dramatically. The speed and delay of service provision are the main criteria for evaluating the performance of the system. To ensure these indicators, it is necessary to optimize the network infrastructure. Currently, methods of distributing calculations of various services are being actively developed.
An important step in this direction of CDN (Content Delivery Network) technologies is the content delivery network. The advantage of this technology is that servers are placed as close as possible to clients, taking into account territorial characteristics. The most important technology that allows you to directly use the resources of base stations is the MES (Mobile Edbe Computing) technology - edge computing. MES allows you to place your nodes directly on BS base stations or BS groups.
Also, the advantage of MES is the ability to occupy the computing space of the node for a short time. Edge computing allows you to solve many problems such as network latency, low data transfer speed.
Today, the trends of automation and development of a flexible network structure are observed in communication networks. The organization of communication networks is moving from manual switching to a flexible virtualized architecture, which operators have been switching to in recent years. Also (in the context of the standards structure (3GPP), network functions virtualization (NFV) is successfully evaluated by communication operators and shows an increase in network efficiency, including from the point of view of network topology optimization.
In addition, within networks (5G), creating ultra-reliable low latency communication (URLLC) and massive machine-type communication (mMTC) may be difficult for networks with a centralized topology. Moreover, network monitoring, in particular, carrier-class equipment, shows that there is no ideal static network topology, in which there is and its elements would be uniformly loaded during a long service time.
The article provides a solution for the automatic distribution of applications on MES - nodes. An inherent advantage of such a solution is the possibility of hosting not only the operator's applications, but also third-party applications on the servers. The Rex-platform solution allows you to automate the method of downloading applications to the optimal MES node. Thanks to the proposed platform, it is possible to reduce the distance between subscribers and the application to a minimum and thus reduce the delay, reduce the load on the central cloud of the service provider, respectively, on the network of the communication operator.

Keywords: MES technology, 5G networks, URLLC, MMT, information entropy method, transport network, invariant control system, combined system.