Connectivity

Title

2025-03-10T21:26:31+00:00

Title

Content

2025-03-10T21:34:10+00:00

Content

Leveraging vivit transformers and foresight pruning for scalable scene change detection on distributed architecture

2025-03-11T23:34:40+00:00

Nowadays, the amount of video content is increasing rapidly and requires a more efficient way to analyze it. Scene change detection is a crucial step in video processing because it allows to group of results by context and provides more detailed analyses. This research focuses on the application of Video Vision Transformer (ViViT) architecture to overcome the following challenges - significant computational power requirements and lack of context capturing by convolutional and recurrent-based architectures. We also focus on applying foresight pruning for ViViT to reduce resource requirements even more.
By training the ViViT model we achieved a 5.5% improvement in the F1 score for scene detection methods over existing approaches. We also optimized the ViViT model size by 43% and inference time by 10% by applying foresight pruning while maintaining state-of-the-art accuracy.
We also propose a pipeline based on a shot detection algorithm that significantly reduces computational complexity by analyzing only key frames for scene and attribute detection. Applying parallelized processing architecture enables simultaneous scene and attribute detection that leads to a 24.21x speed up against the classic approach of analyzing every frame. This study presents a robust, efficient, and scalable solution for scene and attribute detection that allows a further improvement of methods for scene and attribute detection applications with the potential for realtime analytics.

Keywords: Scene Change Detection, Video Vision Transformer (ViViT), Video Analysis, Parallel Processing, Scalability, Neural Networks, Artificial intelligence, Software Engineering.

Potential problems of applying the model-oriented paradigm in mobile network design

2025-03-12T00:20:23+00:00

In this paper, we address the limitations of the traditional model-based paradigm in mobile wireless communication networks, such as the complexity of defining accurate models, obtaining system parameters, high computational demands, and the inability to create lossless block decompositions. These challenges hinder the effective use of model-based approaches in dynamic, evolving mobile networks.
We analyze the data-driven paradigm, supported by advanced machine learning techniques, as a solution. Unlike model-based approaches, which depend on predefined network models and parameters, the data-driven paradigm builds networks directly on data generated by the network itself. This approach efficiently handles real-time, dynamic data, bypassing the need for static models. We investigate one of a typical use case where this paradigm is applied to implement proactive load balancing. A key feature of this approach is online learning, which enables networks to predict traffic spikes before they occur and adjust network parameters accordingly. This proactive strategy minimizes congestion and improves efficiency by anticipating traffic surges, such as those caused by big groups of users which are moving between base stations. The paper describes how online learning methods are used to predict and avoid packet overloads due to rapid traffic changes, especially in mobile networks with varying data rates. We also investigated load balancing using online learning, where access points independently predict traffic based on neighboring cells' data, and a central load balancer adjusts cell configurations to reduce congestion. Proactive adjustments are made before traffic spikes, ensuring minimized disruptions.
In conclusion, the data-driven paradigm, combined with machine learning, offers significant advantages over traditional approaches, particularly in scalability, flexibility, and real-time adaptability. As mobile networks evolve, further research into these methods will be crucial for more efficient, intelligent communication systems.

Keywords: 5G, model-based approach, data, mobile wireless networks, machine learning, traffic prediction, load balancing, computational complexity.

Lawful interception of information in IP multimedia communication subsystems

2025-03-12T22:33:53+00:00

The article examines lawful interception of information in an IP multimedia communication subsystems electronic communication network (IMS). The architecture of IMS is completely based on the standard open IP protocol. It interacts with fixed, mobile and wireless communication networks for voice and data transmission. It is noted that each IMS function does not necessary correspond to a separate module (hardware). Two functions can be combined in a single module, and one function can be implemented in several modules. Certain functional modules in the specified subsystem, interaction with which is necessary for authorized structures to lawful intercept information, are given. It is established the type and content of the information, that generated by the specified functional modules. It is noted, that during the lawful interception of information the authorized bodies will receive the communication sessions of the subjects of interception, information about their location and additional information about service profile attached to the end (terminal) equipment of the subjects of interception. Attention is drawn to need to modernize the gateways of network sets of technical means for their adaptation to specifics of the application of IP subsystems of multimedia communication in electronic communication networks. The gateway of the network set of the technical means of the interception system on the internal interfaces must provide to functional modules a surveillance table with identification objects and must receive from them informational messages and metadata of interception objects, related with interception subjects (surveillance subscribers). Suggested offers for changes to the normative document on lawful interception in Ukraine. The proposals are recommended to be used during the planning of operative-search, counter-intelligence, reconnaissance measures and covert investigative (search) actions in IP subsystems of multimedia communication of mobile communication networks of public use in Ukraine.

Keywords: lawful interception of information, electronic communication network, multimedia communication, IP subsystem, technical means, functional modules.

The role of Edge computing in computer networks

2025-03-13T00:00:44+00:00

In modern computer networks, edge computing is becoming increasingly relevant, particularly due to its impact on the efficiency of data processing amidst the growing volume of data, the operation of network systems, real-time requirements, and the high demands for speed and security in information transmission.
The main challenges faced by traditional centralized computing systems include data processing delays, network congestion, rising costs for maintaining data centers, and risks to data confidentiality. The article highlights that edge computing can address these issues by processing data directly at the network edge, closer to the source of data generation. This approach reduces delays, optimizes network bandwidth usage, and enhances security by keeping data local.
With the rapid development of 5G and artificial intelligence (AI) technology, more and more intelligent devices are connected to the Internet, and the demand for computing services is also increasing. The traditional combination of cloud computing and the Internet of Things (IoT) has gradually revealed some drawbacks, such as transmission delay, greatly reduced transmission rate and network bandwidth pressure. Edge computing is a new computing paradigm. It puts the service near the physical edge of the end users. Due to its close proximity, dense distribution and low latency, edge computing can effectively reduce latency, improve transmission speed and relieve bandwidth pressure.
At present, there are more and more researches on edge computing. It is necessary to discuss edge computing in general. The system mapping study uses the visualization method to summarize a specific field. This paper aims to introduce the development of edge computation in detail by means of a systematic mapping study.
This paper examines the concept, characteristics, and key aspects of edge computing, its role in computer networks, and its practical applications. Modern hardware solutions for implementing edge computing are explored. A comparison of edge, fog, and cloud computing is presented, highlighting their advantages, disadvantages, and areas of application. Practical examples of edge computing usage in industries such as healthcare, autonomous transport, energy, and gaming are provided.
Recommendations are offered for selecting hardware and implementing edge computing based on the specific needs of users and organizations.

Keywords: computer network, edge computing, data center, cloud computing, Internet of Things,
IoT, enterprise networks, computer network management.

FTTBusiness technology and ways to implement it

2025-03-13T00:39:06+00:00

Fiber optic technologies continue to dominate the communications industry, as their speed and reliability make them a very popular choice for business processes. In a dynamic business environment, speed and flexibility are key success factors, and to stay competitive, companies are actively implementing innovative solutions that take business process efficiency to the next level through cloud services, the Internet of Things, ERP systems, and other modern tools. These changes and implementations allow companies to respond quickly to market changes, improve the quality of customer service and optimize the use of resources.
This article is devoted to the FTTBusiness technology and how it can be implemented to optimize business processes using high-speed optical networks. The general concept of FTTx and its variants, including FTTN, FTTC, FTTB/O and FTTH, were considered. Special attention was paid to the analysis of basic architectures, their advantages and disadvantages.
In addition, the article analyzes the benefits of using FTTBusiness for business. High bandwidth, low latency, and immunity to interference make optical networks an ideal solution for large data transfers, video conferencing, cloud computing, and other resource-intensive applications. With FTTBusiness, enterprises can increase operational efficiency, improve customer service, and gain a competitive advantage in the market.
Interoperability with other technologies was also considered, such as the use of WDM (Wavelength Division Multiplexing) technologies in combination with FTTBusiness to further increase network capacity and efficient use of optical spectrum. This allows transmitting several optical signals of different wavelengths over a single fiber, which significantly increases the efficiency of network resources and improves network scaling.
The study presented options for implementing FTTBusiness using P2P architecture and C/DWDM technologies and determined that the choice of the optimal architecture depends on specific business needs, network size, and budget.

Keywords: FTTBusiness, fiber optic network, high-speed Internet, access technologies, WDM.

Improvement of loadsharing algorithms in the common channel No. 7 signalling network

2025-03-13T01:25:06+00:00

The quality of telecommunication services depends on performance characteristics, which determine ability of telecommunication networks to transfer information between users. One of the main influencing factors to the quality of services is signalling network performance characteristics. The common channel signalling No. 7 (SS7) is used as the main signalling system in the channel switching digital telephone networks, integration services digital networks, intelligent networks. One of the ways to ensure high-quality, reliable and efficient operation of the SS7 network is to optimize the loadsharing in it. The standardized loadsharing algorithm in the SS7, based on using bits of signalling link selecting (SLS) field of route label, is analyzed. The following shortcomings of this algorithm are determined: limited number signaling links in the linkset and the choice of SLS bits for loadsharing at signalling points depends on the SLS bits used for loadsharing on previous sections of the signalling route. The improved signalling loadsharing algorithms, based on using, additionally to SLS field, information from another fields of route label: originating point code, destination point code, circuit identification code, service indicator, are studied. Comparative analysis of these algorithms is carried out, the advantages and disadvantages of them are determined. These algorithms eliminate limitations of the standardized algorithm. However, the implementation of these algorithms depends on the functionality of the telecommunications equipment of the SS7 network. Ways to improve loadsharing in the SS7 network are proposed, which are not related to additional functions of telecommunication equipment. Firstly, it is not recommended to use for loadsharing the first (the youngest) bit of SLS in signalling ended points. Secondly, it is proposed to separate signalling ended points of the netwok into clusters and use the same SLS bit in each cluster for load sharing from ended points to signalling transit points, and different SLS bits use in different clusters. Thirdly, it is recommended to use: multiple signalling point codes in one physical point, high-speed signalling links, IP network resources using SIGTRAN protocol.

Keywords: telecommunication network, common channel signalling No.7, signalling load, loadsharing.

Analysis of the current stage of development of artificial intelligence in telecommunications

2025-03-14T00:20:11+00:00

As you know, the basis of the telecommunications industry is network planning. It is the guarantor of the reliability of network connections and the effectiveness of communication depends on it, which is the basis of everything, from interactions in social networks to planning and conducting business operations. In this way, network planning ensures reliable and uninterrupted operation of the internal interactions of the network itself. And this, in turn, accelerates innovation processes within the telecommunications industry.
The appearance of 5G networks, as well as sixth-generation networks not on the horizon and the complexity of network infrastructures, once again emphasizes the importance of using the latest algorithms of artificial intelligence and machine learning. This makes it possible to solve the issue of network performance optimization. AI/ML have already become indispensable elements in networks today, capable of raising various aspects of communications to significant heights, as well as effectively solving the problems of network planning, diagnostics and optimization.
For the effective application of artificial intelligence (AI) in future generations of telecommunication networks, a wide range of neural networks is being researched quite intensively in this direction.
These are feedforward neural networks, deep neural networks, feedback networks, and convolutional neural networks. All of them use machine learning (ML) methods to model multiple relationships between system input and output and identify patterns in the data.
AI brings operational efficiency and network management to telecommunications. The level of interaction with customers is significantly increased, and the ability of artificial intelligence to analyze and forecast enables telecommunications companies to anticipate the urgent needs of their customers. Artificial intelligence algorithms allow you to predict and effectively fight various cyber threats, DDoS attacks, etc.

Keywords: artificial intelligence, telecommunications, 5G networks, machine learning, neural networks.

Comparison of algorithms for building a cluster model based on a dataset obtained from bigdata

2025-03-14T01:13:57+00:00

MeanShift is a popular clustering algorithm widely used in a range of machine learning applications. A major drawback is the slow speed of the algorithm, as it requires quadratic time for one iteration. By enhancing the MeanShift algorithm with a mode-merging method based on mean-shift clustering, we justify this approach by showing that it allows probabilistic clustering interpretation based on the affinity of kernel density weights. This type of integration also optimizes the weight kernels and enables the use of variable-sized kernels according to local data structures. As a result, we achieved a significant speed improvement. Unlike classical MeanShift, this combined approach is based on linear time with respect to the number of points and exponential with respect to size. The aim of this article is to provide an overview of how mean-shift clustering can be applied to model building and to highlight the advantages of using a non-classical approach to mean-shift methodology compared to traditional methods. We will attempt to create a generalized list of crypto transactions to provide users with risk analytics for a crypto wallet or an individual crypto transaction. We will also compare the influence of different parameters and functions on cluster composition. The proposed method reduces computational costs while maintaining an acceptable level of clustering accuracy, similar to the standard mean-shift procedure. We will demonstrate the method’s effectiveness on a sequence of vectors that are non-constant and change over time. This experiment shows that the mean-shift values obtained through our distance calculation method outperform those obtained using classical methods when dealing with non-obvious and unstructured data values. To clarify the relationnnships between clusters and improve sorting accuracy, parameters such as market capitalization and other fiat indicators were used, which can be applied in future studies.

Keywords: clustering, machine learning, Big Data, blockchain, crypto transfer, Mean Shift Clustering.

Research on the impact of artificial intelligence on the development of video surveillance systems

2025-03-14T01:38:42+00:00

This paper presents the analysis on the impact of artificial intelligence on the development of video surveillance systems, focusing on key technologies and algorithms that significantly enhance the efficiency and functionality of such systems. The main aspects are examined, including automatic object detection, facial recognition, and integration with other technologies, particularly IoT. The advantages of using artificial intelligence are highlighted, including increased data analysis accuracy and reduced human intervention. Ethical and legal challenges associated with the implementation of AI in video surveillance are also discussed. The article contributes to understanding how cutting-edge technologies are transforming security and management systems, opening new opportunities for businesses and society.
Intelligent video analytics, smart facial recognition, asset monitoring, task automation tools, and biometric recognition are just a few examples of AI applications in the field of security and surveillance. AI plays a crucial role in all these aspects, helping the surveillance industry achieve optimal operational efficiency and a high level of security.
The implementation of artificial intelligence in video surveillance systems is a result of technological advancements and the transition of society into the era of machine learning and AI technologies. This process is characterized by the integration of artificial intelligence, the Internet of Things (IoT), cloud services, video surveillance systems, and a range of other technologies to improve operational outcomes. Such a transition promises several advantages, including: increased flexibility and responsiveness in obtaining video data through the computerization of surveillance systems; enhanced efficiency of systems due to the Internet of Things, cloud computing, and automation; and improved accuracy of output data through the use of artificial intelligence.

Keywords: digital video surveillance, machine learning, Internet of Things, artificial intelligence, CCTV.

Analysis of the effectiveness of using artificial intelligence to control enemy forces in tower defense games

2025-03-16T01:27:54+00:00

This research paper is devoted to analyzing the effectiveness of using artificial intelligence to control enemy forces in Tower Defense games. In today's gaming world, Tower Defense games play an important role, and the use of artificial intelligence can greatly improve the gameplay and provide a more refined experience for players.
The topic of this article is devoted to the analysis of the effectiveness of using artificial intelligence to control enemy forces in tower defense games. The study examines the general characterristics and features of artificial intelligence. The results of the analysis show that the existing methods of shaping enemy behavior have both advantages and disadvantages.
A deep analysis of existing methodologies in this field revealed several key problems related to the predictability of enemy artificial intelligence. These problems include frequent errors due to incomplete or inaccurate information, limited opportunities to analyze different decision-making options, and the complexity of managing a large set of potential actions.
After a detailed review of both the theoretical foundations and empirical studies, it became clear that the application of the fundamental principles of the partial enumeration optimization method (PEOM) has significant potential for further development. The main goal of future research is to improve the current algorithms developed for creating scenarios of the behavior of opponents in turn-based strategy games. This is aimed at eliminating the shortcomings of previous methodologies and improving the overall efficiency and quality of the gaming experience.

Keywords: Artificial Intelligence; Tower Defense; management of enemy forces; machine learning algorithm; Partial Enumeration Optimization Method (PEOM); analysis of the effectiveness of the use of artificial intelligence.

Application of game structure generation methods for an open world game in the RPG genre

2025-03-16T02:38:14+00:00

The article considers important aspects of using game structure generation methods to create a unique game process. In connection with the spread of 3D world generation technologies in modern games, there is a need to improve the processes of settings and control over world generation. The authors investigate the impact of using algorithms for generating game structures on improving the quality of the game process, as well as on game optimization.
The article considers the possibilities of using game structure generation algorithms to automate world creation processes. This includes selection of optimal generation parameters, such as speed, processor load, and others. The authors consider the possibility of creating prediction models that allow predicting the optimal conditions for the generation of game structures for specific computers and users.
The main focus is on eliminating generation defects using object creation algorithms and analyzing the large amount of data collected during the game discovery process in order to select the optimal algorithm. The use of methods for generating game structures allows you to identify possible problems during generation and find optimal ways to correct them. This not only increases the quality of the generation of game structures, but also reduces the percentage of the load on the operating system, but also ensures the stability of these processes.
The general conclusion of the article is that the application of game structure generation methods to the creation of the world in a 3D game has great potential for improving the game's efficiency and optimizing the processes of its creation. Further research in this direction could open up new opportunities for the game industry and game design, contributing to the development of digital production.

Keywords: 3D games, 3D models, game structures, algorithm, NPC, RPG, Open world, information systems, information technologies, software modelin.

Improving the efficiency of authorization and authentification using blockchain technology

2025-03-16T02:57:22+00:00

In the contemporary research landscape, blockchain technology is garnering recognition as an innovative solution that enhances the security and transparency of operations across various sectors.
Its application is particularly significant in the realms of authorization and authentication, where blockchain introduces innovative approaches to managing these critical processes. Research in this domain indicates a trend towards increasing adoption of blockchain-based solutions due to their ability to enhance security, ensure transparency, improve audit capabilities, optimize user interaction, and expand user control over processes.
Given the widespread use of online services such as online banking, which enables users to conduct transactions directly and efficiently, there is a growing need for robust authentication mechanisms to prevent fraud. Traditional methods like passwords and PIN codes are proving insufficient in the face of modern cybersecurity challenges. Blockchain offers alternative solutions that allow users to have independent and secure ownership of their information and facilitate safe data exchange among various service providers.
This article examines how blockchain can address data accuracy issues and provide a robust infrastructure for handling incidents and scenarios related to identification and authorization. It also considers the prospects of decentralization as a key element in all digital solutions provided by governments and private enterprises, with an emphasis on information confidentiality.
Ultimately, blockchain technology emerges as a potentially revolutionary tool for ensuring the security and reliability of digital transactions, considering the necessity for scalability, integration, and compliance with regulatory requirements for its effective implementation and widespread application.

Keywords: blockchain, authorization, authentication, smart contracts.

Using chatbots as an alternative to standart services for improving the efficiency of medium-sized enterprise

2025-03-16T03:21:18+00:00

The article explores the implementation of chatbots as an effective alternative to traditional software solutions for optimizing and automating internal processes in medium-sized enterprises. Particular attention is paid to analyzing the main challenges of existing solutions, such as high licensing costs, complexity of integration with corporate systems, and insufficient flexibility to adapt to specific business needs. The study highlights the advantages of chatbots, including their modularity, seamless integration capabilities through APIs with necessary systems, and accessibility for every employee on corporate platforms such as Microsoft Teams.
The primary goal is to develop flexible and scalable solutions that reduce costs and enhance operational efficiency. The technological aspects of chatbot development and implementation are analyzed, particularly their ability to integrate with various systems, such as databases, automate routine processes, and provide access to internal services through a unified interface. Special emphasis is placed on the economic justification: developing an in-house chatbot by internal specialists signifycantly reduces costs compared to ready-made solutions such as Jira Service Management. Moreover, it allows the company to maintain and dynamically adapt the product according to business needs and changes in the system infrastructure.
Practical results of the study demonstrate significant improvements in the performance of internal services. For instance, the automation of parking management now enables one-click reservations, proactive notifications in Microsoft Teams have reduced response times to queries, and a transparent request tracking system has optimized the workload of HR and IT departments. The chatbot not only simplifies task execution but also provides a foundation for further personalization of its functions and integration of new modules.
The research results emphasize the importance of business process automation for medium-sized enterprises, demonstrating the effectiveness of using chatbots as a key element of digital transformation. The adoption of custom solutions opens up new opportunities for businesses, allowing them to focus on strategic development and innovation.

Keywords: automation, chatbot, medium-sized enterprises, integration, cost-effectiveness, digital transformation.