Abstract: The subject of this research is the adaptive management of access to Random Access Channels (RACH) in Narrowband Internet of Things (NB-IoT) networks, which frequently face congestion due to high device density and limited channel capacity. The study focuses on the practical application of Reinforcement Learning algorithms, specifically Q-learning and Deep Q-Network (DQN), to address this issue. The authors thoroughly examine the problem of RACH overload and the resulting collisions that cause delays in data transmission and increased energy consumption in connected devices. The article analyzes the limitations and inefficiency of traditional static slot allocation methods and justifies the necessity of implementing a dynamic, learning-based approach capable of adapting to constantly changing network conditions. The research aims to significantly minimize collision rates, improve connection success rates, and reduce the overall energy consumption of NB-IoT devices. The research methodology involved the use of advanced machine learning methods, including Q-learning and DQN, together with simulation modeling conducted in the NS-3 environment, integrating a dedicated RL-agent for dynamic and intelligent RACH slot allocation. The main conclusions of the study highlight the demonstrated effectiveness of the adaptive RL-based approach for optimizing access to communication slots in NB-IoT networks. The scientific novelty lies in the development and integration of a specialized RL-agent capable of dynamically managing slot distribution based on real-time network conditions. As a result of implementing the proposed approach, the number of collisions was reduced by 74%, the number of successful connections increased by 16%, and the energy efficiency of the devices improved by 15% in comparison with traditional static methods. These results clearly demonstrate the practical applicability, and scalability of adaptive RL-based management techniques for enhancing both the performance and reliability of real-world NB-IoT networks.

Keywords: Internet of Things, IoT, RACH, reinforcement learning, NS-3, collisions, DQN, Q-learning, Reinforcement Learning, NB-IoT

Abstract: Stochastic packet generation models are models that are used to generate traffic in computer networks with certain characteristics. These models can be used to simulate network activity and test network performance. Standard data transmission on the network is packet generation with delays, in which packets are sent at certain intervals. Various stochastic models can be used to generate delayed packets, including uniform distribution, exponential distribution, and Erlang distribution. In this work, an experimental setup was assembled and a client-server application was developed to conduct research and analyze the performance of the data transmission channel. An algorithm has been proposed that allows to restore the moment characteristics of a random value of the interval between packets for further use of queuing models. The analysis of the distribution laws on the performance of the experimental network sample was performed and estimates of the efficiency of channel use and the average packet generation time in network segments, as well as histograms of delays according to the distribution laws, were obtained. An experimental setup was created, and a client-server application was developed to analyze the performance of the data transmission channel. An algorithm for restoring the moment characteristics of the time intervals between packets is proposed. The analysis of the distribution laws on network performance was carried out, estimates of the efficiency of channel use and the average packet generation time in network segments were obtained, as well as histograms of delays according to the distribution laws. The generation of packets with delays according to stochastic distribution laws (uniform, exponential, Erlang) is of great importance in modeling and analyzing the operation of network systems. Also, the generation of packets with delays according to the above-mentioned distribution laws allows testing and debugging of network applications and devices in conditions close to real ones. This allows to identify possible problems and improve the operation of network systems. As a result of the experiment, an algorithm was proposed that allows to restore the moment characteristics of a random value of the interval between packets for further use of queuing models. Also, the analysis of the influence of distribution laws on the performance of the experimental network sample was performed and estimates of the efficiency of channel use and the average packet generation time in network segments, as well as histograms of delays according to distribution laws, were obtained.

Keywords: generating packages, data transmission, traffic analysis, client-server application, delays, packet switching, Erlang distribution, exponential distribution, uniform distribution, Stochastic models

Abstract: The object of the study is information transmission systems. Improving the quality of information on the receiving side is considered as the subject of the study. The authors consider in detail such aspects of the topic as the formation of a procedure that significantly reduces the influence of intersymbol interference caused by the interaction of the signal itself and the sampling window. This source of errors, as a rule, is dominant at information transfer rates close to the bandwidth of the communication channel. Studies on reducing the influence of intersymbol distortions are important and relevant, and in recent years a significant number of works have been devoted to them. A signal model simulating the transmission of a sequence of non-modulated pulses of the meander type was considered. It was assumed that the processing of the incoming pulse stream is carried out in real time based on the procedures of periodic evaluation of each element. The methodological basis of the research was the methods of mathematical modeling of information transmission systems and linearization methods using the generalized superposition principle. The main result of the conducted research is the proposed method of forming a homomorphic filter for processing the incoming pulse stream in real time based on the procedures of periodic evaluation of each element. The algorithm of its functioning ensures the transformation of the multiplicative interaction of the signal and the sampling windows into an additive one and ensures the separation of the multiplicatively interacting transmitted information signal and the sampling window in the communication channel. The resulting procedure, which reduces the effect of intersymbol interference on the receiving side, is an implementation of an optimal filter based on a homomorphic transformation. An estimate of the magnitude of intersymbol interference is obtained when using the proposed processing method. The effectiveness of the method in signal streaming processing is demonstrated. The expressions are obtained in the most general form and can be detailed within the framework of the described information transfer model, which is the subject of further research.

Keywords: linearization, homomorphic filter, periodic evaluation, interaction, data selection window, pulses of the meander type, intersymbol interference, communication channels, information transmission systems, the principle of superposition

Abstract: The object of this study is a service for managing virtual machines in a cloud environment. When developing and operating such a service, it becomes necessary to assess its availability and reliability for compliance with the selected quality level that the client can count on. This paper presents a developed system that allows testing the availability of a cloud service for managing virtual machines. The method of integration with the existing monitoring system at the enterprise using open source software in order to reduce the cost of development and operation is considered. A test case for deploying and removing a virtual machine using a graphical user interface has been developed and implemented, and triggering criteria have been defined. The requirements for the architecture and implementation of the system based on the production statistics of the virtual machine creation service using the Prometheus monitoring system are collected and analyzed. The novelty of the research lies in the development of a new method of testing a cloud service for managing virtual machines in order to increase its reliability and availability. Based on this method, a system for testing virtual machines is described and implemented, as well as a method for integration into the monitoring system of the Intel cloud service. During the operation of cloud environments with the help of this system, problem areas were identified in the architecture of the virtual machine creation service, which made it possible to optimize the system operation in a timely manner. The described method is an effective way to test cloud services, and can also be used to analyze and improve reliability and availability.

Keywords: application testing, python, kubernetes, selenium, prometheus, monitoring, information technology, metrics, data processing, threshold

Abstract: The subject of this research is the new methods for improving the efficiency of Zabbix monitoring systems for the computational infrastructure of the universal infocommunication commutator, built upon the principle of globally distributed computational complex using the information cloud technologies. The object of this research is the globally distributed computational system of RingCentral corporation (USA), which is based on the technology of hybrid information cloud deployed in over a hundred  data storage and processing centers in all large economic and geographical zones, which provides universal infocommunication services, including video and audio conferences, virtual stationary automated telephone, and organization of call centers. Design and improvement of infocommunication cloud platforms, which provide universal communication services is a complex scientific, technical and economic problem; without its solution it is impossible to create information infrastructure with the service quality indicators that meet the demands of the society. This problem is of particular relevance in the context of worldwide sanitary-epidemiological pandemic of coronavirus, since part of the production processes was switched to remote work, and is executed without physical presence of administrative and technical personnel at the enterprises of national economy. This entailed a decline in the efficiency of networks and quality of customer service in certain infocommunication companies due to excessive nominal loads on the information network, and this, the failure of crucial nodes of user traffic processing.

Keywords: info communication, IT, maintenance, distributed IT systems, change management, cloud computing, automation, monitoring systems, Zabbix, quality of service

Abstract: One of the topical areas of the digital economy is the development of services based on the use of systems for local positioning of objects. The article presents the results of analytical modeling of a positioning system built on the equipment of the IEEE 802.15.4 standard. The purpose of this study is to evaluate the efficiency of data transmission in a ZigBee network. As a modeling tool, the mathematical apparatus of queuing networks (CeMO) was selected, which has proven its effectiveness for calculating the characteristics of various infocommunication systems. The method of analysis of average values was chosen by the authors to assess the quality of service indicators due to its computational simplicity and the adequacy of the results obtained. The authors have developed a structural diagram of the investigated positioning system. On the basis of the network architecture, a conceptual, algorithmic and software model in the form of a closed homogeneous queuing network has been developed and its characteristics have been calculated. The parameters obtained as a result of modeling make it possible to evaluate and analyze the quality of service indicators in the telecommunications segment of the local positioning system under study, such as latency, performance, network load factor and the probability of losses, which can be used for efficient traffic management in the telecommunications segment of the geolocation system based on IEEE 802.15.4 standard.

Keywords: throughput, mean value analysis, queueing networks, ZigBee, geopositioning, time-probability characteristics, QoS, utilization, delay, queueing system

Abstract: Improving the performance of various types of communication networks in the modern world remains an urgent task, in connection with which research is underway to create hardware that expands the throughput of physical channels, new network protocols are being developed and existing network protocols are being modified, mathematical and computer modeling of data transmission mechanisms in communication networks is being carried out. The speed and reliability of data transmission over networks also depends on a number of factors, the nature of the influence of which is random. The combination of such factors is called a random environment. If the change in the states of the medium is continuous, then we speak of a diffusion medium. The object of the research is communication networks controlled by multiple access protocols and functioning in a random (diffusion) environment. The research tool for multiple access networks is the mathematical apparatus of the theory of finite-difference and differential equations, the theory of random processes and the theory of queuing. The proposed mathematical model of communication networks in a diffusion environment is investigated by an asymptotic method. The scientific novelty of the work lies in the fact that for the first time a mathematical model of a multiple access network operating in a diffusion environment was proposed and an asymptotic study was carried out. The asymptotic average of the normalized number of claims in the orbit (the source of repeated calls) and the deviation from this average are found, and the probability density of the values of the process of changing the number of claims in the orbit is obtained.

Keywords: Markov chain, asymptotic analysis method, diffusion approximation, diffusion process, random environment, orbit, multiple access protocol, communication network, variable parameters, Poisson flow of applications

Abstract: The article presents the results of the analysis of the error in calculating the coordinates of a mobile subscriber using three Wi-Fi access points, based on measuring the received signal power. The purpose of the study was to develop a geographic system for large storage facilities and a preliminary assessment of the accuracy of calculations in conditions close to real ones. The principles of local and global positioning based on the measurement of various physical characteristics of signal propagation were considered, a method was selected for the implementation of a positioning system. The authors developed a test bench layout of a map on the IEEE 802.11 standard mobile network and software for solving this problem using the trilateration mechanism. To simulate the surrounding electromagnetic environment, a parameter imitating additive white Gaussian noise was added to the calculation formulas. In the course of the work, graphical dependences of the error of measuring the coordinates of the object on the signal / noise ratio in the channel and on the distance from the reference access point are obtained. The results showed that even in the worst operating conditions, the developed system surpasses most systems on the market in terms of measurement accuracy.

Keywords: location, RSSI, distance, trilateration, geopositioning, access point, Wi-Fi, coordinates, power, accuracy

Abstract: The subject of the research is the new principles of building globally distributed computer and computing systems for creating a software infocommunication switch providing telephony services, sending facsimile, short and multimedia messages services, via packet data transmission channels, and also storing user files. The object of scientific research is the globally distributed computing system of RingCentral (USA), which began to be created from the beginning of the company's founding in 1999 and is technologically constantly evolving at the present time. The author reveals in detail the main criteria for testing the hardware environment of the implemented computer infrastructure during the transition to the distributed computing model. The creation of a globally distributed computing system architecture for an information and communication software switch was a decisive factor in the development of computer systems in the provision of universal information and communication services. The exchange of the obtained scientific knowledge at specialized conferences and seminars showed that the constructed model of information and communication services is the most modern from the technological point of view of the proposed architecture.

Keywords: maintenance, disaster recovery, network architecture, resilience, outage prevention, data store, high availability, distributed IT systems, performance testing, backup

Abstract: The subject of the study are algorithms for the interaction of mobile stations for sharing a wireless channel with a wireless access network with token access. The object of the study is a wireless network with token access. The author considers in detail such aspects of the topic as combining the technologies of local computer networks: the IEEE 802.11 wireless technology standard (Wi-Fi technology) and Token Ring Token ring technology, describing the interaction of mobile stations for sharing a common broadcast channel. The research methodology combines methods of system analysis, informatics, reliability theory, queuing theory theory, information-logical design, mathematical modeling. The main conclusions of the study are the algorithms of wireless interaction of mobile stations for sharing a broadcast channel, providing a reliable (by the criteria of limited delay and reserved bandwidth) communication, reducing the number of retransmissions due to the fact that each station in turn uses the channel for equal periods and is forced to abandon the right to transfer after the expiration of the time allocated to it and provide resistance to node failure and automatic recovery from failures.

Keywords: communication reliability, computer network, broadcast channel, network ring technology, wireless communication, marker access, mobile station, wireless network, local area network, wireless interaction

Abstract: The article presents the results of analytical modeling of mobile communication systems 4G. The purpose of this study is to evaluate QoS performance in an LTE network channel. As a simulation tool, the mathematical apparatus of queuing networks (SEMO) was chosen, which proved its effectiveness for calculating the characteristics of infocommunication systems of various purposes and of any dimension. The author examined various research methods, further on the basis of the evolutionary system architecture of LTE, a conceptual, algorithmic and programming model was developed in the form of a closed homogeneous SemOI and its characteristics were calculated. The mean value analysis method was adapted to obtain network quality parameters with LTE technology. During the simulation, the probabilistic-temporal characteristics of the service process in the fourth-generation mobile communication systems were obtained: the transmission delay and the network capacity for a different number of active users, represented in the form of dependency curves from the input load. The results can be used both for designing new ones and for efficient traffic management in existing mobile networks.

Keywords: LTE, delay, mobile systems, QoS, quality of service, queuing systems, queuing networks, modeling, Mean Value Analysis, throughput