Miroslav Joler

Prof. Miroslav Joler

Department of Computer Engineering, University of Rijeka, Faculty of Engineering, Rijeka, Primorsko-Goranska Županija, Croatia
orcid

Research Focus & Metrics

Academic Career Length

25Years (2002-Present)

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

Representative Publications

View All Publications
18 August 2020 in IEEE Transactions on Very Large Scale Integration (VLSI) Systems
Verified
Journal Article
Closed access
YXYusheng Xie
Yusheng Xie
ARAlex Noel Joseph Raj
Alex Noel Joseph Raj
ZHZhendong Hu
Zhendong Hu
SHShaohaohan Huang
Shaohaohan Huang
Zhun FanZhun Fan
Zhun Fan
Miroslav JolerMiroslav Joler
Miroslav Joler
In this article, we propose a novel approach to reduce hardware resource consumption when neural networks (NNs) are deployed on field-programmable gate array (FPGA) boards. Rather than using a classical approach with lookup tables (LUTs) to approximate the activation functions of an NN, the proposed solution is based on a twofold LUT (t-LUT) architecture, which comprises an error-LUT (e-LUT) and a data-LUT (d-LUT), in order to achieve high precision and speed as well as low hardware resource consumption. The efficiency of the proposed approach was tested against multiple earlier approaches. Our solution showed that the compressibility of the previously referenced works, which were based on single LUTs, could be improved by up to 94.44% and those that were based on a range addressable LUT (RALUT) by up to 6.35% in the examined case of a hyperbolic tangent (tanh) activation function. Moreover, when RALUT and our architecture were combined, it improved the compressibility of the RALUT-based result by up to additional 10.21% for a tanh activation function. The designed architecture had an initial latency of 39.721 ns, when tested with a 50-MHz clock, to simultaneously retrieve data from the d-LUT and t-LUTs.

17 January 2018 in IEEE Transactions on Antennas and Propagation
Verified
Journal Article
Closed access
Miroslav JolerMiroslav Joler
Miroslav Joler
MBMaja Boljkovac
Maja Boljkovac
A circularly polarized sleeve-badge antenna is proposed for a 2.45 GHz frequency band. The proposed antenna is small and thin enough to fit within a standard jacket sleeve badge, while the assembly is based on standard off-the-shelf parts, which enables simple and costeffective mass production. A wide beamwidth and a 5.6% impedance bandwidth provide practical properties.

8 July 2004
Verified
Conference Paper
Closed access
YYYang Ye
Yang Ye
SZS.C. Zhang
S.C. Zhang
FDF. Dogan
F. Dogan
ESE. Schamiloglu
E. Schamiloglu
JGJ. Gaudet
J. Gaudet
CCC. Christodoulou
C. Christodoulou
In an effort to develop transmission lines with higher energy storage capabilities for compact pulsed power applications, the University of Missouri-Rolla (UMR) and the University of New Mexico (UNM) have undertaken a collaborative approach to developing and studying ceramic dielectrics. At UMR, the electrical breakdown strength (BDS) of TiO/sub 2/-based materials is investigated for high energy density applications. The results of research to-date show that dense titania ceramics with nanocrystalline grain size (/spl sim/200 nm) exhibit significantly higher BDS as compared to ceramics made using coarse grain materials. Processing-microstructure-property relationships in TiO/sub 2/ systems are found to play a role with respect to increasing the BDS. At UNM, a pulsed power system is being assembled to perform BDS studies of the ceramic materials produced at UMR. Electromagnetic simulations in support of this work will also presented. The long-term aim of this research is to enable the fabrication of large sizes of high energy density ceramics for use in pulsed power systems.

25 January 2022 in Sensors
Verified
Journal Article
Open access
PDF file
Miroslav JolerMiroslav Joler
Miroslav Joler
ARAlex Noel Joseph Raj
Alex Noel Joseph Raj
JBJuraj Bartolic
Juraj Bartolic
In this paper, we present a simple yet efficient method for determination of the relative permittivity of thin dielectric materials. An analysis that led to definition of the proper size and placement of a sample under test (SUT) on the surface of a microstrip ring resonator (MRR) was presented based on the full-wave simulations and measurements on benchmark materials. For completeness, the paper includes short descriptions of the design of an MRR and the variational method-based algorithm that processes the measured values. The efficiency of the proposed method is demonstrated on 12 SUT materials of different thicknesses and permittivity values, and the accuracy between 0% and 10% of the relative error was achieved for all SUTs thinner than 2 mm.

1 September 2015 in IEEE Antennas and Propagation Magazine
Verified
Journal Article
Closed access
Miroslav JolerMiroslav Joler
Miroslav Joler
JKJosko Kucan
Josko Kucan
Due to the importance that reconfigurable antennas have in modern designs of low-profile antennas, a comprehensive study of the impact of slot parameters on the resonant frequencies of a rectangular microstrip antenna (RMA) has been undertaken and the results have been presented here. The changes exhibited in the first three resonant frequencies were graphed with respect to the changes in the values of the specific slot parameter. The impact of the slot length, width, and the position was thoroughly documented. Following the computational results, a few models were selected for manufacturing and the measurements performed on them showed a good agreement with the computational results.

8 July 2004
Verified
Conference Paper
Closed access
Miroslav JolerMiroslav Joler
Miroslav Joler
CCC.G. Christodoolou
C.G. Christodoolou
ESE. Schamiloglu
E. Schamiloglu
JGJ. Gaudet
J. Gaudet
In this paper, we present a computational approach to the research of designing high-power microwave systems as compact and portable units. We discuss the motivation to run a comprehensive computational study as an alternative to the, more common, experimental set-ups. We further elaborate on the differences between a physical set-up of the experiment vs. a computational model. The differences between the ideal theoretical settings and the current computer model are contrasted as well as the current challenges encompassing the computational modeling.