ETF Journal of Electrical Engineering

13

Authors: Zoran N. Milivojević, Zoran S. Veličković

This paper describes the construction of a parametric cubic interpolation kernel based on the blending method. The blending kernel is constructed by mixing two third-order interpolation kernels, the one-parameter Keys kernel (parameter α) and oMoms3 kernel. The proportionality of the participation of the oMoms3 kernel in blending kernel is represented by the blending factor (w). Created blending kernel has two parameters (α, w) which adjusting affects the accuracy of interpolation, that is, the reduction of interpolation error. The characteristics of the blending kernel are shown graphically in both, the time and spectral domains. After that, the algorithm for estimating the optimal parameters of the blending kernel is presented. The algorithm is described using a pseudo code. Subsequently, the precision of estimation of the fundamental frequency of the speech signal in the spectral domain, was tested experimentally, using an estimation algorithm. First, the speech test signal is processed in the time domain using some window (Hamming, Hann, Kaiser and Triangular). Subsequently, the speech test signal was transformed into the spectral domain using fast Fourier transformation (FFT). Then, using Peak-picking algorithm, a dominant spectral component (fundamental frequency) was determined. Most often the real fundamental frequency is between the two dominant spectral components. The real fundamental frequency is determined by applying a parametric convolution with a blending interpolation kernel. The estimation precision is represented by the mean square error (MSE) between the estimated and the real fundamental frequency. The optimal kernel parameters are determined by minimizing the mean square error, and the appropriate window is selected. The results are presented by tables and graphics.

8

Authors: Vladan Durković, Dragan Filipović

In this paper, by using separation of variables in Laplace’s equation, we derive a formula for the capacitance of a thin line conductor positioned inside a grounded square shield at one of its cross-section symmetry axes. The formula, which is in the form of a slow convergent series, is transformed into the sum of two series, one of which can be summed in a closed form, and the other one is a very rapidly convergent series and may be replaced accurately enough by its first term. As a result a very simple formula for the capacitance is obtained. Based on this formula, we give some numerical results for the capacitance versus the conductor position inside the shield. In particular, when the conductor is at the center of the shield our results are compared with the results obtained from an aviabile empirical formula for different values of the shield size.

10

Authors: Žarko Zečević, Tomo Popović, Božo Krstajić

Unmanned air vehicles (UAVs) are used increasingly often to collect images for precision agriculture. Since drones fly at low altitudes, the camera has a limited field of vision. To cover an area of interest, taking even up to a thousand photos is necessary. To provide a complete view of the area, the photos need to be merged into a single composite image. Here we describe a DroneMapper - cloud application for automatic mosaicking and georeferencing. DroneMapper uses numerous open source tools for image processing and mapping. Thanks to the powerful computing resources of the server on which the application is hosted, DroneMapper performs a multiple times faster map processing in comparison to desktop solutions.

15

Authors: Miloš Brajović, Miloš Daković, Ljubiša Stanković

The problem of the cross-terms reduction in a higher order time-frequency representation is considered in this paper. The analyzed representation is originally proposed to considerably improve the concentration of non-stationary signals in the time-frequency plane. However, in the multicomponent signal analysis, the high order of the representation leads to the appearance of undesirable cross-terms. The analyzed representation is developed based on the approximation of the first derivative of the signal phase. It ideally concentrates signals with polynomial phase up to the fourth order. The cross-terms reduction is based on the implementation using windowed frequency convolutions, a concept borrowed from S-method framework.

9

Authors: Martin Ćalasan, Boris Miljanić, Vladan Vujičić

In this paper the determination of induction machine parameters by using Particle Swarm Optimization (PSO) technique is presented. An equivalent circuit without considering the iron losses is analyzed. The circuit parameters are found as the result of the error minimization function between the estimated and maker data. In order to determine the parameters, the values of the phase current and power factor for three slip values of the machine have been used. The accuracy of the usage PSO is analyzed by determining the relative error between the optimized and real value of the phase current and power factor for all three slip values. The obtained results are also compared with results found in literature which are obtained by using Genetic Algorithm (GA). It is concluded that PSO is more efficient in solving the parameter estimation problems.

10

Authors: Aleksandra Lopušina, Enis Kočan, Milica Pejanović-Djurišić

Millimeter wave (mmWave) communications are expected to be the forerunner in achieving 5G data rate performance requirements. As mmWaves communications can provide channel bandwidths in the order of GHz, they can fulfill those demands. In this paper we have tested data rate performance of mmWave communication systems operating at 28GHz and 73GHz, in the outdoor environment, using the ns-3 simulator, in both single user and multi user scenarios. We focused on data rate, as one of the most important service quality indicators. Obtained results have confirmed that mmWave communications can meet 5G expectations and provide services to low mobile users in outdoor environment, at distances up to few hundred meters, with no significant obstacles between the transmitter and the receiver.

10

Authors: Dragan Filipović, Vladan Durković

In this paper we have derived a simple formula for the potential of an infinitely thin charged line conductor between two ground half planes at an arbitrary angle. Based on this formula the conductor capacitance per unit length is determined, followed by illustrative examples.

11

Authors: Zoran Veličković, Zoran Milivojević, Marko Veličković

In this paper, the protection of video content from copying by embedding a scrambled watermark is considered. The watermark is scrambled by the GMSAT algorithm, and the insertion is performed by a reliable algorithm in the DWT-SVD domain of each frame. The video is encoded by the H.264/AVC encoder before it is uploaded on a Web server, while variable-quality watermarks are extracted on the client side from the decoded video. Using an algorithm for quality enhancement of the extracted watermark, the excellent SSIM index of 0.8956, and NC coefficient of 0.9852 was realized. The obtained results are in rank with the best one published. Experimental results justify the use of the shown video protection technique because it provides a higher quality of the extracted watermark compared to other insertion techniques.

8

Authors: Žarko Zečević, Tomo Popović, Zdravko Uskoković, Božo Krstajić

In this paper, a new algorithm for synchrophasor and frequency estimation in power systems is proposed. The proposed approach is based on transformation of the measured signal into a complex sinusoid by using the Constant Modulus algorithm. The transformed signal and transformation coefficients are used for estimation of the parameters of interest. Simulation results show that the proposed algorithm exhibits better performances compared to the considered algorithms.

9

Authors: Vesna Rubežić, Luka Lazović, Marko Babić, Ana Jovanović

In this paper, the linear dynamic behavior of the helicopter in terms of characterization of the vibrations in a helicopter, is described. After analyzing in the time and frequency domain, vibrations are analyzed in geometrical and topological methods of chaos theory. Although the vibration in a helicopter are mainly periodic, the presence of chaotic dynamics using a systematic and detailed approach, is investigated. The aim of the research is to characterize the vibration mechanism based on a one set of measured data.

10

Authors: Stefan Vujović, Miloš Brajović, Vesna Popović-Bugarin, Nedeljko Latinović, Jelena Latinović, Milija Bajčeta

A web-based expert system for the grapevine monitoring and forecasting the appearance of downy mildew, which is a common disease of grapevine in Montenegro, is presented in the paper. The system is based on the 3 10 rule, customized for the south Montenegro region. The proposed web-based expert system is user-friendly, requiring minimal user-input data. In addition to several locations and weather stations which are initially embedded into the system, new locations can be easily added, and each location may use an arbitrary weather station. Moreover, the system is implemented with a multi-user structure, providing two types of predefined users: the administrator and the farmer. The priority levels and privileges depend on the active type of user. The purpose of the system is to predict the grapevine disease appearance based on the live data collected from weather stations in the field, combined with the expert knowledge implemented into the system. The proposed system provides the live monitoring of several relevant parameters, and alerts the user when a risk of infection is detected.

10

Authors: Jelena Šoškić, Budimir Lutovac, Dragan Filipović

This paper presents an analysis of the rectangular patch antenna. The analysis is based on the resonator model. Theoretical results are verified by using the software package MATLAB. Besides, some of the patch antenna parameters, which are very important for design and analysis of patch antennas, are presented in the same program package.

11

Authors: Zoran Veličković, Zoran Milivojević, Marko Veličković, Milojko Jovetić

Secure sharing of the digital video content over the Internet requires the use of modern techniques of protection. This paper describes the insertion of the encrypted watermark in the uncoded video content as a way of protecting copyrights. The impact of the prediction structure in the H.264 encoder to the survival of the inserted watermark is tested for two schemes. The watermark is encrypted with multistage Arnold transformation, and on the reception, an advanced algorithm for the realized by the SSIM index 0.71521 for an IBBBBBP scheme has been applied. The results obtained through the displayed example justify the application of multistage Arnold transformation for the video copy protection and raise security to a higher level.

16

Authors: Nataša Savić, Zoran Milivojević, Dejan Blagojević, Darko Brodić

First part of this paper describes Greville’s one-parametric interpolation kernel and provides its calculated spectral characteristics. Then it provides the calculation of optimal parametric values of Greville’s kernel a) by minimizing wiggles of spectral characteristics, b) by minimizing differences of the spectral characteristics in relation to kernel characteristics in the form sin(x)/x. Furthermore, this paper provides analytical form of the slope of spectral characteristics in boundary area. Second part of this paper describes an experiment in which the assessment of Greville’s algorithm efficiency in interpolation of audio signals (G tones played on piano) and speech signals (logatoms spoken in Serbian language) was performed. The results are presented in tables and graphs.

9

Authors: Vlatko Cingoski, Roman Golubovski, Saso Gelev, Goce Stefanov, Goran Klepov

The modern way of living increases the need for attractive sites for tourists to visit. Usually artistic fountains are attractive performing visual and spraying effects synchronized to musical pieces, and thus getting tourists’ attention. This paper proposes one novel method and approach to the problem of controlling and operation of an electric drive of the pumps. The proposed method is a techno-economic compromise achieving coordinated performance of the individual pump without usage of additional expensive servo valves, for the purpose of their synchronization with the choreography based on the desired music theme.

10

Authors: Slavica Tomović, Milutin Radonjić, Milica Pejanović-Djurišić, Igor Radusinović

Resource limitations of sensor nodes are the main weakness of wireless sensor networks (WSNs). This problem is difficult to alleviate in traditional ad-hoc networks, where each node participates in decentralized routing process and independently determines the next-hop to send data towards the destination. In that regard, this paper presents WSN architecture based on principles of software defined networking (SDN), which enables automatic network reconfiguration and optimal management of energy constrained sensor nodes with limited communications abilities. We discuss implementation challenges and the key benefits of software defined WSN design, with a special emphasis on advantages offered by centralized routing algorithms.

10

Authors: Žarko Zečević, Zdravko Uskoković, Božo Krstajić

A new iterative method for phasor estimation of power system is proposed. The proposed technique uses three adaptive algorithms. First two algorithms are used for frequency estimation. Third adaptive algorithm estimates amplitudes and phase angles between phasors. Simulation results confirm that the proposed algorithm exhibits better performances compared to the considered ones.

11

Authors: Miloš Brajović, Ljubiša Stanković, Miloš Daković

This paper addresses the reconstruction problem of non-stationary signals with missing samples. The reconstruction is achieved by using concentration measures of time-frequency representations in combination with a gradient-based iterative algorithm. As an example of time-frequency representation, the S-method is used in the proposed approach. The sparsity of the transform domain, needed for a successful reconstruction, is interpreted through the concept of concentration measures, and limits for successful reconstruction are discussed. Several examples with nonstationary signals which exhibit different concentrations in the time-frequency domain illustrate the presented theoretical concepts.

15

Authors: Aleksandar Sokolovski, Saso Gelev

The scope of this research paper is one very important aspects nowadays, the security and management of one of the most important services the email and all of the alike online services today. This paper attempts to investigate the possible benefits of using standard signature-driven spam detection logic in combination with algorithm for network intrusion detection system (NIDS). The primary objective is to verify that proposed solution (standard signature-driven spam detection logic and NIDS) will be an effective strategy for dealing with email spam detection. The main aim is to determine best possible integration of standard signature-driven spam detection logic in combination with algorithm for NIDS, for creating more effective solution for dealing with email spam and priority mail compared to the previous solutions available. This will be achieved by testing the effectiveness of the solution compared to other solutions until today, by using network simulators NS-3 and other programming tools.

15

Authors: Zoran Veličković, Zoran Milivojević, Milojko Jevtović

One way to protect the copyright of the video content is inserting invisible information - the watermark in the video content. The extraction of the embedded watermark is especially difficult in wireless networks, which are a subject to packet drops. Although multimedia applications are tolerant to packet losses, it cannot be said for their extraction algorithms. In this paper, it is shown that the improved iterative algorithm for watermark extraction from the encoded video can be successfully applied in the wireless environment. The efficiency of the improved algorithm is evaluated in the simulated wireless environment. In the wireless environment with 1% of dropped packets SSIM index was corrected by 24%, while in the environment with 2% of dropped packets SSIM index was corrected by 27.7%.

23

Authors: Ljubiša Stanković, Isidora Stanković

Signals sparse in a transformation domain can be recovered from a reduced set of randomly positioned samples by using compressive sensing algorithms. Simple reconstruction algorithms are presented in the first part of the paper. The missing samples manifest themselves as a noise in this reconstruction. Once the reconstruction conditions for a sparse signal are met and the reconstruction is achieved, the noise due to missing samples does not influence the results in a direct way. It influences the possibility to recover a signal only. Additive input noise will remain in the resulting reconstructed signal. The accuracy of the recovery results is related to the additive input noise. Simple derivation of this relation is presented. If a reconstruction algorithm for a sparse signal is used in the reconstruction of a nonsparse signal then the noise due to missing samples will remain and behave as an additive input noise. An exact relation for the mean square error of this error is derived for the partial DFT matrix case in this paper and presented in form of a theorem. It takes into account very important fact that if all samples are available then the error will be zero, for both sparse and nonsparse recovered signals. Theory is illustrated and checked on statistical examples.