Boban Pavlovic

Followers

Following

Public Views

Interests

Uploads

Papers by Boban Pavlovic

Deep learning channel estimation for 5G wireless communications

Vojnotehnički Glasnik, Dec 31, 2022

Results: Our findings reveal a significant potential of CNN-based channel estimation in 5G commun... more Results: Our findings reveal a significant potential of CNN-based channel estimation in 5G communication systems, with the VDSR method demonstrating a consistent performance across all scenarios. This re-911 Laidouni, M. et al, Deep learning channel estimation for 5G wireless communications, pp.911-940 search contributes to the advancement of channel estimation techniques in 5G networks, paving the way for enhanced wireless communication systems with improved reliability. The VDSR architecture demonstrates remarkable adaptability to different types of channels, which results in achieving requested performances for all analyzed SNR values.

Download

Virtuelne privatne mreže – moguće rešenje pouzdanih komunikacija

Vojnotehnički Glasnik, 2008

Analiza tehnika upravljanja prenosom podataka

Vojnotehnički Glasnik, 2006

Analiza karakteristika MPLS mreznog simulatora

Vojnotehnički Glasnik, 2005

profesor dr Milojko Jevtovi}, dipl. in`.

Download

Koncepcija i karakteristike buduće generacije telekomunikacionih mreža

Vojnotehnički Glasnik, 2007

On Visually Lossless JPEG Image Compression

2023 Zooming Innovation in Consumer Technologies Conference (ZINC)

Prediction of the First Just Noticeable Difference Point based on Simple Image Features

2023 Zooming Innovation in Consumer Technologies Conference (ZINC)

Anomaly network intrusion detection system based on NetFlow using machine/deep learning

Vojnotehnički Glasnik, Dec 31, 2022

have emerged as a valuable tool, particularly in military fields, for protecting networks against... more have emerged as a valuable tool, particularly in military fields, for protecting networks against cyberattacks, specifically focusing on Netflow data, to identify normal and abnormal patterns. This study investigates the effectiveness of anomaly-based machine learning (ML) and deep learning (DL) models in NIDSs using the publicly available NF-UQ-NIDS dataset, which utilizes Netflow data, with the aim of enhancing network protection. Methods: The authors Sarhan, M., Layeghy, S., Moustafa, N. and Portmann, M. in the conference paper Big Data Technologies and Applications, in 2021, involve a preprocessing step where 8 features are selected for the training phase out of the 12 available features. Notably, the IP source and destination addresses, as well as their associated ports, are specifically excluded. The novelty of this paper lies in the preprocessing of the excluded features and their inclusion in the training phase, employing various classification ML and DL algorithms such as ExtraTrees, ANN, simple CNN, and VGG16 for binary classification.

Download

$Research paper thumbnail of Analiza Objektivnih Procena Kvaliteta Komprimovanih Video Sekvenci Iz Vidljivog I Infracrvenog Dela Elektromagnetnog Spektra$

Analiza Objektivnih Procena Kvaliteta Komprimovanih Video Sekvenci Iz Vidljivog I Infracrvenog Dela Elektromagnetnog Spektra

ZBORNIK RADOVA ČETRDESETOG SIMPOZIJUMA O NOVIM TEHNOLOGIJAMA U POŠTANSKOM I TELEKOMUNIKACIONOM SAOBRAĆAJU – POSTEL 2022, Nov 29, 2022

Contrast Quality Measure: Full-Reference Image Quality Assessment Metric for Infrared Images

2022 New Trends in Signal Processing (NTSP)

Just Noticeable Difference Prediction and Image Quality Assessment

2022 IEEE Zooming Innovation in Consumer Technologies Conference (ZINC)

Comments on Objective Quality Assessment of MPEG-2, H.264 and H.265 Videos

2018 New Trends in Signal Processing (NTSP), 2018

This paper presents the performance analysis of five full-reference objective quality assessment ... more This paper presents the performance analysis of five full-reference objective quality assessment measures on compressed high-resolution video sequences from two publicly available video datasets. Quantitative measures that were accepted by the international research community used in the performance analysis - the correlation coefficient of ranks and the linear correlation coefficient between subjective and objective quality scores. The performance analysis is carried out globally - the level of complete datasets and on subsets of sequences with common characteristics (compression type or source sequence). Three types of compressed video content were analyzed - H.265 and its predecessors H.264 and MPEG-2. The correlation between subjective and objective quality scores is analyzed in different scales (resolutions) of the video signal. From the analysis of the results, very important conclusions on the objective quality assessment of the compressed video signals were derived, as well as the directions for further work.

Picture-wise just noticeable difference prediction model for JPEG image quality assessment

Vojnotehnicki glasnik, 2022

Introduction/purpose: The paper presents interesting research related to the performance analysis... more Introduction/purpose: The paper presents interesting research related to the performance analysis of the picture-wise just noticeable difference (JND) prediction model and its application in the quality assessment of images with JPEG compression. Methods: The performance analysis of the JND model was conducted in an indirect way by using the publicly available results of subject-rated image datasets with the separation of images into two classes (above and below the threshold of visible differences). In the performance analysis of the JND prediction model and image quality assessment, five image datasets were used, four of which come from the visible wavelength range, and one dataset is intended for remote sensing and surveillance with images from the infrared part of the electromagnetic spectrum. Results: The pap 86 er shows that using a picture-wise JND model, subjective image quality assessment scores can be estimated with better accuracy, leading to significant performance impro...

Download

Comments on objective quality assessment of JPEG images with visible differences

2017 13th International Conference on Advanced Technologies, Systems and Services in Telecommunications (TELSIKS), 2017

We analyse the results of a comprehensive collection of objective quality metrics for high defini... more We analyse the results of a comprehensive collection of objective quality metrics for high definition compressed visual content. The investigation provides guidelines that allow a reliable JPEG image quality assessment, and therefore the development of perceptual-based image/video coding standards. We show on a comprehensive, recent subjective dataset with visible differences that visual information fidelity, most apparent distortion, and gradient-based objective measures have reached the best agreement with the subjective quality scores, where contrast is preferred according to the luminance and structure changes.

RF-based drone detection and classification system using convolutional neural network

2021 29th Telecommunications Forum (TELFOR), 2021

This paper presents an effort towards developing a detection and classification information syste... more This paper presents an effort towards developing a detection and classification information system based on the RF signature of several commercial drones. The developed application implements a Convolutional Neural Network which was trained and tested using data from a publically accessible database. The tested neural network reached an accuracy of almost 100% (4-classes), which is considered as a significant contribution to the development of a functional drone detection system. Moreover, the developed interface allows the user to supervise the spectral activity in the 2.4 GHz ISM band, notifies him about the presence and the nature of a potential threat, and stores the event log in a database for later exploitation.

An Experimental Study on the Phase Importance in Digital Processing of Speech Signal

This paper presents the results of a late large-scale subjective study of phase importance of spe... more This paper presents the results of a late large-scale subjective study of phase importance of speech quality. Present study includes a collection of speech sentences distorted by limited transmission bandwidth and phase degradations. A detailed statistical analysis of the collected subjective judgments is presented. Among the used signal distortions, subjects preferred modification with signal phase preservation. Mean opinion scores of such modification are the closest to the original, undistorted sentences. This subjective study contributes to improving the algorithms of the speech processing, and in addition provide valuable data to develop objective or automatic methods of speech quality assessment, as well as to estimate their performance.

Download

Gradient-based Image Quality Assessment

An objective measure for image quality assessment based on a direct comparison of visual gradient... more An objective measure for image quality assessment based on a direct comparison of visual gradient information in the test and reference images is proposed. A perceptual model is defined to provide local estimates of gradient preservation and investigate perceptual importance pooling of such local quality estimates by using the lowest scores. The proposed perceptual pooled measure is validated using extensive subjective test results. Results indicate that the proposed measure is perceptually meaningful in that it corresponds well with the results of subjective evaluation and can outperform actual objective metrics.

Download

Analysis of Still Image IP Transmission in the Tactical Communication Network

2019 Zooming Innovation in Consumer Technologies Conference (ZINC), 2019

This paper presents analysis of still image IP transmission in the tactical network. The analysis... more This paper presents analysis of still image IP transmission in the tactical network. The analysis is provided through the delivery time, packets sent/received/lost and the display of the images after the reconstruction on the receiving side. Image transmission is analyzed through multiple scenarios - without and with compression, without and in the presence of jamming. It has been shown that delivery time can be significantly reduced by data compression, without significant loss of visual information. It has also been shown that for analyzing the quality of transmission and usefulness of received information, it is better to use the display of reconstructed images than the degree of packet losses. Furthermore, it has been shown that in the tactical networks of limited bandwidth, it is very important to comply bit rates from source of information to the end user, and in this way to avoid buffer memory overflow and packet losses.

Efficient Prediction of the First Just Noticeable Difference Point for JPEG Compressed Images

Acta Polytechnica Hungarica, 2021

In this paper, we show that a high-level of correlation exists between a simple image featuremean... more In this paper, we show that a high-level of correlation exists between a simple image featuremean gradient magnitude and the peak signal-to-noise ratio (PSNR) of the first just noticeable difference point for JPEG image compression. On the basis of this observation, we proposed a method to estimate the JPEG quality factor which represents the effective limit between perceptually lossy and lossless coding as the PSNR of the first just noticeable difference point. The goal is optimal image/video coding, at the lowest compression bit-rate that ensures perceptually lossless output image quality. We also show that this feature can be used to predict higher PSNR just noticeable difference points.

Download

Performance analysis of full-reference objective image and video quality assessment metrics

Vojnotehnički glasnik, 2018

This paper presents the performance evaluation of image and video quality assessment metrics on t... more This paper presents the performance evaluation of image and video quality assessment metrics on two publicly available datasets with subjective quality ratings. In addition to the performance analysis at the global level-at the level of complete datasets, the paper presents the objective measures performance evaluation on subsets of signals inside them. The image dataset contains five subsets created by using different types of JPEG compression, while the video dataset contains six subsets of sequences-four created by compression of original sequences, and two subsets are with video signal transmission characteristic degradations. To determine the success of objective measures, i.e. comparison of subjective and objective quality scores, there were used measures accepted by the International Telecommunication Union-ITU (linear correlation coefficient, rank-order correlation coefficient, mean absolute error, root mean squared error and outlier ratio). It was shown that objective quality measures can reach a high level of agreement with the results of subjective tests on subsets of datasets. Objective measures performances depend on the type of degradation which significantly affects the performance at the complete dataset level. The difference in Bondžulić, B. et al., Performance analysis of full-reference objective image and video quality assessment metrics, pp.322-350 performances is more pronounced on video sequences due to considerable visual differences in sequences created by using compression, packet losses and additive Gaussian noise. Therefore, we can say that a universal objective measure, i.e. measure that is useful for different types of signal degradation, for different degradation levels, and for different applications currently does not exist.

Download

Boban Pavlovic

Uploads

Papers by Boban Pavlovic

Log In