Veljko Lončarević

Veljko Lončarević

Auto-generated profile, based on publicly available information

Veljko Lončarević

University of Kragujevac

Research Focus & Metrics

Academic Career Length

3Years (2024-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
25 November 2025
Verified
Conference Paper
Closed access
BMBranko Marković
Branko Marković
VLVeljko Lončarević
Veljko Lončarević
JGJovan Galić
Jovan Galić
This paper presents the implementation of 1DCNN for whispered speech recognition. At the front end, MFCCs were used to construct feature vectors, while the back end employed a 1D-CNN for training, validation, and testing. Speech patterns were taken from the Whi-Spe database, considering two speakers (one male and one female). Four types of feature vectors were examined through different scenarios: Normal/Normal, Whisper/Whisper, Normal/Whisper, and Whisper/Normal. The results confirm the effectiveness of 1DCNN for speech recognition.

9 June 2025
Verified
Conference Paper
Closed access
VLVeljko Lončarević
Veljko Lončarević
ZJZeljko Ljubinko Jovanovic
Zeljko Ljubinko Jovanovic
Smartphones equipped with GPS and accelerometer sensors enable large-scale collection of spatiotemporal data, which can be leveraged for transportation research and urban analysis. In this study, the clustering of geospatial data collected along the route from Čačak to Kragujevac, Serbia, using an Android-based mobile application is investigated. The dataset comprises longitude, latitude, time, and speed attributes, enabling the identification of movement patterns through clustering techniques. DBSCAN and K-Means algorithms are applied to group data points based on location and visualize the resulting clusters using Folium maps. The findings provide insights into travel behavior and variations in movement dynamics.

1 January 2025 in Serbian Journal of Electrical Engineering
Verified
Journal Article
Open access
PDF file
BMBranko Marković
Branko Marković
VLVeljko Loncarevic
Veljko Loncarevic
JGJovan Galić
Jovan Galić
Discrete Wavelet Transforms (DWTs) provide time?frequency representations that are well suited for nonstationary signals such as speech. This study presents a comparison of four wavelet families (Daubechies, Symlets, Coiflets, and Biorthogonal) for bimodal automatic speech recognition across two speech modes (normal and whispered). Experiments use the Whi-Spe database comprising ten speakers (five female and five male). A Dynamic Time Warping (DTW) back-end performs sequence alignment and recognition. Results are reported via summary tables, histograms, and confusion matrices and reveal systematic differences among the wavelet families, identifying the most effective transform for bimodal recognition. These findings provide practical guidance for selecting wavelet-based front ends in whisper-robust automatic speech recognition (ASR) systems.


1 January 2024
Verified
Conference Paper
Closed access
VLVeljko Lončarević
Veljko Lončarević
ZJZeljko Ljubinko Jovanovic
Zeljko Ljubinko Jovanovic
VLVanja Lukovic
Vanja Lukovic
MMMarina Milošević
Marina Milošević
SŠSavo Šućurović
Savo Šućurović
AIAleksa Iričanin
Aleksa Iričanin
This paper explores horizontal scaling of PHP MVC applications with session preservation for enhanced availability and resource efficiency. It covers theoretical aspects of MVC architecture, PHP in web development, session handling, and horizontal scaling methods, including load balancers, Docker, and Kubernetes. The practical methodology details environment setup, application development, session management, Dockerization, Kubernetes integration, and horizontal scaling configuration. Performance testing reveals significant improvements, showing a response time decrease from an unresponsive state at 1000 RPS (5111 ms) to 32 ms at 2500 RPS with horizontal scaling. The study contributes insights and practical guidance for highly available and scalable web applications.

1 January 2026
Verified
Conference Paper
Closed access
VLVeljko Lončarević
Veljko Lončarević
ZJZeljko Ljubinko Jovanovic
Zeljko Ljubinko Jovanovic
Geolocation-based applications increasingly rely on multiple external services to enrich location information with administrative, environmental, and contextual data. As the volume of location requests grows, traditional synchronous service architectures face challenges related to scalability, latency, and database performance. This paper presents an event-driven geolocation data enrichment framework that combines multi-service API orchestration, Apache Kafka event streaming, hierarchical database partitioning, indexing strategies, and caching mechanisms. The proposed architecture processes incoming geolocation requests through a distributed enrichment pipeline, where independent microservices retrieve complementary location information from multiple free geolocation and open-data APIs. Enriched events are persisted within a hierarchically partitioned PostgreSQL database organized according to geographical structures (continent, region, country, and city). The study evaluates the impact of partitioning, indexing, and caching on query performance, response latency, and system throughput. Experimental results demonstrate that combining event-driven processing with database optimization techniques improves scalability and significantly reduces query execution times compared to a non-partitioned baseline architecture. The findings provide practical guidance for the design of efficient geolocation information systems and other event-driven spatial data processing platforms.