Rastislav Kulhanek

Dr. Rastislav Kulhanek

Comenius University in Bratislava, Faculty of Management
orcid

Research Focus & Metrics

Academic Career Length

6Years (2021-Present)

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

Representative Publications

View All Publications
24 February 2025 in Scientific Reports
Verified
Journal Article
Open access
PDF file
AAAhmad Almadhor
Ahmad Almadhor
JBJamel Baili
Jamel Baili
SAShtwai Alsubai
Shtwai Alsubai
AHAbdullah Al Hejaili
Abdullah Al Hejaili
RKRastislav Kulhanek
Rastislav Kulhanek
SASidra Abbas
Sidra Abbas
GPS spoofing presents a significant threat to small Unmanned Aerial Vehicles (UAVs) by manipulating navigation systems, potentially causing safety risks, privacy violations, and mission disruptions. Effective countermeasures include secure GPS signal authentication, anti-spoofing technologies, and continuous monitoring to detect and respond to such threats. Safeguarding small UAVs from GPS spoofing is crucial for their reliable operation in applications such as surveillance, agriculture, and environmental monitoring. In this paper, we propose a compact, tiny deep learning architecture named CTDNN-Spoof for detecting and multi-label classifying GPS spoofing attacks in small UAVs. The architecture utilizes a sequential neural network with 64 neurons in the input layer (ReLU activation), 32 neurons in the hidden layer (ReLU activation), and 4 neurons in the output layer (linear activation), optimized with the Adam optimizer. We use Mean Squared Error (MSE) loss for regression and accuracy for evaluation. First, early stopping with a patience of 10 epochs is implemented to improve training efficiency and restore the best weights. Furthermore, the model is also trained for 50 epochs, and its performance is assessed using a separate validation set. Additionally, we use two other models to compare with the CTDNN-Spoof in terms of complexity, loss, and accuracy. The proposed CTDNN-Spoof demonstrates varying accuracies across different labels, with the proposed architecture achieving the highest performance and promising time complexity. These results highlight the model’s effectiveness in mitigating GPS spoofing threats in UAVs. This innovative approach provides a scalable, real-time solution to enhance UAV security, surpassing traditional methods in precision and adaptability.

24 June 2024 in IEEE Access
Verified
Journal Article
Open access
SASidra Abbas
Sidra Abbas
GSGabriel Avelino Sampedro
Gabriel Avelino Sampedro
Moez KrichenMoez Krichen
Moez Krichen
MAMeznah A. Alamro
Meznah A. Alamro
AMAlaeddine Mihoub
Alaeddine Mihoub
RKRastislav Kulhanek
Rastislav Kulhanek
The increasing occurrence of Hypertension highlights the need for advanced predictive tools in healthcare. This research proposes a novel approach that combines machine and deep learning for new feature generation and hypertension prediction. We explore machine learning-based models: Random Forest (RF), Logistic Regression (LR), Decision Tree (DT), XGBoost (XGB), and Gradient Boosting (GB) for new feature prediction (FP) and integrated these predictions with the original dataset for training deep Long short-term memory (LSTM) model. To evaluate the efficiency of the proposed approach, we compare all models, predicting new features, with those in the existing study. The results demonstrate that the GB-based FP + LSTM are standout performers. The GB-based FP + LSTM combination demonstrates the highest accuracy at 98.48%. On the contrary, the LR-based FP + LSTM combination exhibits a lower accuracy of 89.39%. The remaining combinations, including RF-based FP + LSTM, XGB-based FP + LSTM, and DT-based FP + LSTM, showcase accuracies ranging from 95.45% to 97.97%. In practical terms, the high F1-score of 98.48% is achieved by the combination of GB-based FP + LSTM, which implies a reliable tool for clinicians to aid in early hypertension detection. These findings hold deep practical implications, offering healthcare practitioners and policymakers a pathway to deploy accurate and timely hypertension identification tools.

30 December 2021 in Mathematics
Verified
Journal Article
Open access
PDF file
OCOlha Chernukha
Olha Chernukha
YBYurii Bilushchak
Yurii Bilushchak
Nataliya ShakhovskaNataliya Shakhovska
Nataliya Shakhovska
RKRastislav Kulhanek
Rastislav Kulhanek
We propose and justify a numerical method for computing the double integral with variable upper limits that leads to the variableness of the region of integration. Imposition of simple variables as functions for upper limits provides the form of triangles of integration region and variable in the external limit of integral leads to a continuous set of similar triangles. A variable grid is overlaid on the integration region. We consider three cases of changes of the grid for the division of the integration region into elementary volumes. The first is only the size of the imposed grid changes with the change of variable of the external upper limit. The second case is the number of division elements changes with the change of the external upper limit variable. In the third case, the grid size and the number of division elements change after fixing their multiplication. In these cases, the formulas for computing double integrals are obtained based on the application of cubatures in the internal region of integration and performing triangulation division along the variable boundary. The error of the method is determined by expanding the double integral into the Taylor series using Barrow’s theorem. Test of efficiency and reliability of the obtained formulas of the numerical method for three cases of ways of the division of integration region is carried out on examples of the double integration of sufficiently simple functions. Analysis of the obtained results shows that the smallest absolute and relative errors are obtained in the case of an increase of the number of division elements changes when the increase of variable of the external upper limit and the grid size is fixed.

23 September 2024 in IEEE Access
Verified
Journal Article
Open access
SASidra Abbas
Sidra Abbas
Stephen A OjoStephen A Ojo
Stephen A Ojo
IBImen Bouazzi
Imen Bouazzi
GSGabriel Avelino Sampedro
Gabriel Avelino Sampedro
AHAbdullah Al Hejaili
Abdullah Al Hejaili
RKRastislav Kulhanek
Rastislav Kulhanek
The widespread use of smartphones has brought convenience and connectivity to the fingertips of the masses. As a result, this has paved the way for potential security vulnerabilities concerning sensitive data, particularly by exploiting side-channel attacks. When typing on a smartphone’s keyboard, its vibrations can be misused to discern the entered characters, thus facilitating side-channel attacks. These smartphone hardware sensors can capture such information while users input sensitive data like personal details, names, email addresses, age, bank details and passwords. This study presents a novel Graph Neural Network (GNN) approach to predict side-channel attacks on smartphone keyboards; different GNN architectures were used, including GNN, DeepGraphNet, Gradient Boosting (GB)+DeepGraphNet, Extreme Gradient Boosting (XGB)+DeepGraphNet and K-Nearest Neighbor (KNN)+DeepGraphNet. The proposed approach detects the side channel attack using vibrations produced while typing on the smartphone soft keyboard. The data was collected from three smartphone sensors, an accelerometer, gyroscope, and magnetometer, and evaluated this data using common evaluation measures such as accuracy, precision, recall, F1-score, ROC curves, confusion matrix and accuracy and loss curves. This study demonstrated that GNN architectures can effectively capture complex relationships in data, making them well-suited for analyzing patterns in smartphone sensor data. Likewise, this research aims to fill a crucial gap by enhancing data privacy in the information entered through smartphone keyboards, shielding it from sidechannel attacks by providing an accuracy of 98.26%. Subsequently, the primary objective of this study is to assess the effectiveness of GNN architectures in this precise context. Similarly, the GNN model exhibits compelling performance, achieving accuracy, precision, recall, and f1 score metrics that showcase the model’s effectiveness, with the highest values of 0.98, 0.98, 0.98, and 0.98, respectively. Significantly, the metrics mentioned in the study outperform those documented in the previous literature. Overall, the study contributes to the detection of side-channel smartphone attacks, which advances secure data practices.

9 December 2024 in IEEE Access
Verified
Journal Article
Open access
SISalma Imtiaz
Salma Imtiaz
AAAhmad Almadhor
Ahmad Almadhor
RKRastislav Kulhanek
Rastislav Kulhanek
Context:Open-source software (OSS) development has become prominent in the software industry over the last 20 years. OSS has transformed how software is developed, distributed, and maintained. This increase in popularity has led to extensive research in various domains of OSS, such as evolution, adoption, community and development processes.With the passage of time the volume of academic studies increased in the area of OSS, including Systematic Literature Reviews (SLRs) and Systematic Mapping Studies (SMS), which have provided valuable insights into specific aspects of OSS. The OSS domain is vast, with many areas that are still under explored. The fragmented nature of the existing studies presents challenges for both researchers and practitioners in identifying comprehensive research trends, gaps, future directions, and the cohesive understanding of the broader trends within the field. The secondary studies synthesize the literature to highlight findings and gaps of an area. The fragmentation of OSS academic literature limits the ability of researchers to identify unexplored or under-researched areas. This also creates challenges for both practitioners and researchers, as they lack awareness of the best practices, tools, technologies, and methodologies in a OSS domain. Objective:This study is motivated by the need to consolidate the extensive research conducted in the OSS domain, providing a holistic view that can guide future investigations and practical applications. The rationale for conducting this research lies in the opportunity to aggregate and classify existing OSS research areas, topics, and future directions through a systematic tertiary study. By synthesizing the findings from secondary studies, this research aims to offer a meta-level understanding of the OSS field, uncovering overlooked areas and defining a research agenda. The systematic approach, guided by the established protocol of Kitchenham, ensures that the study is conducted rigorously, with a focus on comprehensiveness and reliability. Ultimately, this research seeks to contribute to the OSS community by highlighting key research areas that require further exploration, thereby advancing the field and supporting the continued growth and innovation within OSS development. Method: A systematic tertiary study is performed to cover all the systematic secondary studies in the area of OSS. The guidelines of Kitchenham are used for designing the protocol. The protocol details the research objectives, scope, search strategy, data extraction, quality assessment and synthesis. The protocol is detailed in section three for transparency. Result: We have identified seventy-four studies that consist of twenty-five SMS and forty-nine SLR. The literature is mapped to a published taxonomy of OSS by Aksulu and Wade, however, the future directions are thematically analyzed. The results of mapping show that the highest number of studies (forty-seven) are in the sub-category of “ OSS categorization/research agenda ”, whereas eight studies are mapped to the sub-category “ OSS vs Proprietary ”. Both of these sub-categories fall in the main category of “ Conceptual ”. The second major work is in the “ OSS Production ” category in the sub-categories of “ Communities ” (ten), “ Process ” (eight), “ User and Developer Motivation ” (nine), and “ Self-Organization (Product and Community Evolution) ” (six). Seven studies are also mapped to the sub-category of “ Software Quality ” in the main category of “ Performance Metrics ”. Other categories have fewer studies mapped to them. The areas identified, thematically, for future directions are “OSS contributors”, “OSS development process”, “OSS evolution and prediction”, “use of OSS in different domains”, and “OSS adoption/adaptation/integration”. Conclusion: The mapping between “key research areas” of systematic secondary studies and “taxonomy categories” shows that there is no or little research in some of the categories of taxonomy, having potential of future research. The future directions thematic analysis will also aid researchers and practitioners.

1 January 2022 in Environmental Problems
Verified
Journal Article
Closed access
LSLviv State University of Life Safety
Lviv State University of Life Safety
VKVasyl Karabyn
Vasyl Karabyn
VSVladyslav Shuryhin
Vladyslav Shuryhin
SSSofia Shutiak
Sofia Shutiak
UAUkrainian National Bar Association
Ukrainian National Bar Association
CBComenius University Bratislava
Comenius University Bratislava
Strategic environmental assessment is an effective tool for improving the level of environmental safety in various areas. The purpose of the article is to identify and classify the main tasks of strategic environmental assessment in subsoil use, taking into account its role and functions in the subsoil management system, and to suggest ways to improve environmental safety in subsoil use by means of strategic environmental assessment. The materials for the article were the results of many years of authors' research on various aspects of the methodology for assessing the environmental safety of subsoil use and scientific source analysis. The main research methods were comparison, deduction, analysis, synthesis, and classification. The authors grouped the tools of the strategic environmental assessment into five main groups: analytical, organisational, scientific and technical, control and procedural, and media. Analysis of the legal framework determines the need to finalise the draft of the Subsoil Code to take into account the requirements of the Law of Ukraine "On Strategic Environmental Assessment" in the context of carrying out state examination and assessment of reserves and resources of minerals and establishing conditions for mineral raw materials, drafting industrial development (technological scheme) of the deposit. The classification of the tools of strategic environmental assessment in subsoil use is carried out. Analysis of draft legislation in the field of subsoil use shows positive changes in achieving sustainable development goals. Simultaneously, the authors identified some inconsistencies between different legislative acts and projects.