Balkan Journal of Applied Mathematics and Informatics

SMART CITY: A REVIEW OF CURRENT DEVELOPMENTS AND IMPLEMENTATION OF SMART GRID TECHNOLOGY

Sara Kostevska — 2025-06-13

To ensure better living standards for the population, it is necessary to develop smart cities that utilize modern technology and operational efficiency to facilitate daily life. The primary goal of implementing (or creating) a smart city is to maintain, balance, and continuously upgrade economic, social, and environmental resources while managing costs. Emphasis is placed on maximizing the use of natural resources through energy conservation (utilizing renewables), efficient water usage, time savings, land optimization, environmental quality, traffic management, and the extensive use of information and communication technology systems and services for effective management and their ongoing enhancement. This work aims to create improved functionality, smart connections, and better organization among people, while maximizing the use of natural resources, minimizing environmental pollution, and improving public health.

GENERALIZATION OF APPLICATION OF FUNDAMENTAL LEMMA OF VARIATIONAL CALCULUS

Aleksandra Risteska - Kamcheski — 2025-06-13

Variational calculus studied methods for finding maximum and minimum values of functional. It has its inception in 1696 year by Johan Bernoulli with its glorious problem for the brachistochrone: to find a curve, connecting two points A and B , which does not lie in a vertical, so that heavy point descending on this curve from position A to reach position in for at least time. In functional analysis variational calculus takes the same space, as well as theory of maxima and minimum intensity in the classic analysis.

In this paper we will prove a theorem for functional where proves that necessary condition for extreme of functional is the variation of functional is equal to zero.

MONITORING OF AC MOTOR SPEED CONTROLLER PARAMETERS TO IoT NETWORK

Goce Gjorgi Stefanov — 2025-06-13

Today's development of electronics mainly takes place in two directions. One direction is based on the technological development of semiconductor elements that will work in a larger current, voltage and frequency range. The other direction is based on the development of information bases and systems that will ensure the collection, visualization, processing and transport of process parameters. The first direction will enable applications in power electronics for higher powers and process speeds, and the second local and remote transmission of process data and their availability in intranet and internet networks. Taking this into account in the paper, an electronic system is designed and implemented that measures, stores and transmits the parameters of a 1-phase AC motor speed controller in an IoT network. The solution enables the visualization of input and output parameters locally on two LCD displays and remotely on an IoT device. A data log file is also provided for the measured values on a local computer and on an IoT cloud computer.

AUTOMATED DOOR STATE DETECTION USING DEEP LEARNING: A COMPUTER VISION APPROACH WITH ROBOFLOW PLATFORM

Elena Jovanovska — 2025-06-13

This paper presents a deep learning-based approach for automated door state detection, capable of classifying doors as open, closed, or semi-open. The implementation leverages the Roboflow platform for comprehensive image processing and model development workflow. Our methodology encompasses data collection, annotation, augmentation, and model training using state-of-the-art deep learning architectures. The system demonstrates robust performance in real-world scenarios, offering potential applications in building automation, security systems, and smart home technologies. This work contributes to the growing field of automated building monitoring by providing a practical solution for door state recognition that can be integrated into existing surveillance and security infrastructures.

DEEP LEARNING-BASED DETECTION AND CLASSIFICATION OF DOCUMENT ELEMENTS USING ROBOFLOW

José Alejandro Ramón Rocha — 2025-06-13

Recent advances in deep learning have enabled significant improvements in document understanding and processing. This paper presents a workflow for detecting and classifying key document elements—paragraphs of text, tables, signatures, and stamps—using deep learning models implemented through the Roboflow platform. We address the unique challenges of document element detection, including complex spatial relationships, overlapping boundaries, and varying formats across different document types. Our methodology leverages Roboflow's comprehensive toolkit for dataset preparation, model training, and deployment, with particular attention to data augmentation techniques specific to document processing. The proposed system achieves robust detection and classification of document elements. Experimental results demonstrate the system's effectiveness in real-world applications. Our findings indicate that the proposed approach offers significant accuracy and efficiency of automated document processing, particularly in scenarios requiring the precise identification and extraction of structured and unstructured document components.

ANALYSIS OF THE MOVEMENT OF FLUCTUATIONS AND TRENDS OF THE GROSS DOMESTIC PRODUCT IN THE REPUBLIC OF NORTH MACEDONIA AND FORECASTS

Anastasija Antova — 2025-06-13

In this paper, an analysis will be made of the movement of fluctuations and trends in the gross domestic product, in the period from 2000 to 2024 by quarter with current prices expressed in millions of denars in North Macedonia, using time series. The analysis will be done using the Python program. In the end, with the help of the previously prepared program and model, a prediction is made for the movement of trends and fluctuations in GDP in the period from 2024 to 2028.

INTEGRATING AI AND CLOUD COMPUTING FOR EFFICIENT AUDIO ANALYSIS

Aleksandar Kotevski — 2025-06-13

Abstract. This paper propose a pipeline for sentiment analysis of audio inputs by using AI and techniques for speech recognition. A key aspect of our approach involves determining audio chunks based on silence detection and decibel height, allowing for more effective segmentation and analysis of the input data. These techniques will be instrumental in categorizing audio materials, enabling more organized and accessible content management. In the scope of this paper, we will compare and utilize several Python packages for audio detection, including Librosa, PyDub, and SoundFile, which offer various functionalities for audio analysis and manipulation. Additionally, we discuss the high computational demands associated with training and deploying of models, highlighting the potential cost barriers for many organizations. To address these challenges, we propose the use of cloud-based solutions, specifically AWS Spot Instances, Azure Spot VMs, and Google Cloud Preemptible VMs, which offer substantial cost savings for processing audio data. By leveraging these cloud resources, organizations can significantly reduce expenses while maintaining high performance in AI audio processing tasks.

STATE-OF-THE-ART COMPARISON OF MOBILESECURECOMM WITH MODERN SECURE COMMUNICATION PLATFORMS FOR TACTICAL OPERATIONS

Rexhep Mustafovski — 2025-06-13

Modern military and critical missions depend on secure, fast, and flexible communication systems that can perform reliably across various operational environments. Legacy platforms such as SINCGARS, EPLRS, and CERT-based emergency systems have offered foundational services, but they fall short when it comes to adapting to new threats, supporting modern data-intensive operations, or integrating with advanced technologies like artificial intelligence and quantum encryption. This paper presents a state-of-the-art comparison between existing secure communication platforms and a newly proposed system called MobileSecureComm. Designed for tactical operations, MobileSecureComm combines modular architecture, quantum-ready encryption, and AI-driven routing to meet the growing demands of modern warfare and emergency response scenarios. The paper draws from extensive literature covering secure mobile platforms, military-grade messaging systems, mobile edge computing, and real-world deployment models. Using simulation-based analysis, the study compares MobileSecureComm to 25 secure communication platforms in terms of latency, throughput, security, interoperability, and resilience. The results show that MobileSecureComm consistently achieves lower latency, higher data throughput, and stronger security safeguards than traditional platforms. It also offers seamless integration with NATO-compliant systems and the flexibility to scale across multi-domain operations. This paper highlights the importance of designing secure communication platforms that are not only technically advanced but also adaptable to changing threats and diverse mission requirements. MobileSecureComm provides a forward-looking solution that can help bridge the gap between existing systems and the future of tactical communication.