Ústav telekomunikací

Browse

Now showing 1 - 5 of 391

Local Communication in Small-Scale PV Systems: Study on Inverter - Smart Meter PLC Communication
(IEEE, 2024-09-17) Musil, Petr; Mlýnek, Petr; Benešl, Lukáš; Mahút, Michal; Rusz, Martin
This study investigates communication technologies and protocols for small-scale photovoltaic (PV) systems, focusing on the interaction between inverters and smart meters. The research evaluates the performance of Power Line Communication (PLC) technologies, comparing both narrowband (NB-PLC) and broadband (BB-PLC) options. The analysis identifies MODBUS protocol limitations and highlights the benefits of advanced protocols like DLMS/COSEM and DNP3 for enhanced efficiency and reliability. Field tests demonstrate the viability of PLC for residential PV systems, with narrowband PLC showing better performance over longer distances. Future work aims to optimize PLC communication, digitize ripple control signals, and develop a Multi-Radio and Cable Access Technology (Multi-RCAT) module. This module will integrate various communication technologies, enabling flexible and redundant local communication behind utility sub-meters. These advancements will support real-time production and consumption control, contributing to the efficient and sustainable operation of decentralized energy systems.
Revolutionizing Visuals: The Role of Generative AI in Modern Image Generation
(ACM, 2024-11-14) Bansal, Gaurang; Nawal, Aditya; Chamola, Vinay; Herencsár, Norbert
Traditional multimedia experiences are undergoing a transformation as Generative AI integration fosters enhanced creative workflows, streamlines content creation processes, and unlocks the potential for entirely new forms of multimedia storytelling. It has potential to generate captivating visuals to accompany a documentary based solely on historical text descriptions, or creating personalized and interactive multimedia experiences tailored to individual user preferences. From the high-resolution cameras in our smartphones to the immersive experiences offered by the latest technologies, the impact of generative imaging undeniable. This study delves into the burgeoning field of Generative AI, with a focus on its revolutionary impact on image generation. It explores the background of traditional imaging in consumer electronics and the motivations for integrating AI, leading to enhanced capabilities in various applications. The research critically examines current advancements in state-of-the-art technologies like DALL-E 2, Craiyon, Stable Diffusion, Imagen, Jasper, NightCafe, and Deep AI, assessing their performance on parameters such as image quality, diversity, and efficiency. It also addresses the limitations and ethical challenges posed by this integration, balancing creative autonomy with AI automation. The novelty of this work lies in its comprehensive analysis and comparison of these AI systems, providing insightful results that highlight both their strengths and areas for improvement. The conclusion underscores the transformative potential of Generative AI in image generation, paving the way for future research and development to further enhance and refine these technologies. This paper serves as a critical guide for understanding the current landscape and future prospects of AI-driven image creation, offering a glimpse into the evolving synergy between human creativity and artificial intelligence.
Enhanced Quantum Convolutional Neural Network for Signature Authentication in Consumer Products
(IEEE, 2024-11-29) Raghupathy, Bala Krishnan; Vairavasundram, Subramaniyaswamy; Ganesan, Manikandan; Namachivayam, Rajesh Kumar; Kotecha, Ketan; Herencsár, Norbert
Product tracking applications utilize the Internet of Things and cyber-physical systems to identify permitted or unauthorized user intrusions into the system. Classical machine learning algorithms cannot detect every risk in an environment that evolves constantly and where new abnormalities are visible. This article investigates the potential of quantum machine learning (QML) for real-time product purchase monitoring and intrusion detection using an enhanced quantum convolutional neural network (EQCNN) with signature-based detection over a massive volume of search space data (qubits). We suggest a three-stage technique to effectively handle the sensitive content: Pre-processing, EQCNN-based feature extraction, and syntactic pattern recognition. Signature-based identification is a feature of the EQCNN architecture that helps detect particular patterns linked to goods purchases or invasions. The model can minimize product tracking mistakes by utilizing the QML-based EQCNN with signature-based detection, resulting in a more efficient supply chain.
Sustainable Electronics: A Blockchain-Empowered Digital Twin Based Governance System for Consumer Electronic Products
(IEEE, 2024-04-29) Sasikumar, A.; Ravi, Logesh; Devarajan, Malathi; Vairavasundram, Subramaniyaswamy; Kotecha, Ketan; Herencsár, Norbert
Industry 4.0 requires digital twins (DTs), a potential technology in the transition. By offering current operational statistics visualizations of physical assets, assisting in decision-making, and mitigating possible hazards in electronic manufacturing environments, DTs are essential to the improvement of distributed consumer electronics. DTs must also work with blockchain technology in dispersed consumer electronics systems to anticipate data privacy and access control. Unfortunately, the single failure of DT collaboration is caused by centralized attacks and data interoperability, authentication, scalability, and connectivity issues. This paper provides the secure architecture for the DTs’ collaboration in consumer electronic devices. Next, we have created an access control mechanism based on a consensus model that combines blockchain, digital twin, and real-time monitoring of consumer devices. The proposed framework uses blockchain technology to empower digital twin-based consumer electronic devices. Specifically, we created a collaborative blockchain architecture based on manufacturer, retailer, and device users, emphasizing a distributed chain code model and real-time access control. The experimental results show that the proposed system integration performs better than the baseline digital twin architecture.
Short-term effects of transcranial direct current stimulation on motor speech in Parkinson's disease: a pilot study
(SPRINGER WIEN, 2024-04-09) Brabenec, Luboš; Kováč, Daniel; Mekyska, Jiří; Řehulková, Lenka; Kábrtová, Veronika; Rektorová, Irena
Introduction: Hypokinetic dysarthria (HD) is a common motor speech symptom of Parkinson's disease (PD) which does not respond well to PD treatments. We investigated short-term effects of transcranial direct current stimulation (tDCS) on HD in PD using acoustic analysis of speech. Based on our previous studies we focused on stimulation of the right superior temporal gyrus (STG) - an auditory feedback area. Methods: In 14 PD patients with HD, we applied anodal, cathodal and sham tDCS to the right STG using a cross-over design. A protocol consisting of speech tasks was performed prior to and immediately after each stimulation session. Linear mixed models were used for the evaluation of the effects of each stimulation condition on the relative change of acoustic parameters. We also performed a simulation of the mean electric field induced by tDCS. Results: Linear mixed model showed a statistically significant effect of the stimulation condition on the relative change of median duration of silences longer than 50 ms (p = 0.015). The relative change after the anodal stimulation (mean = -5.9) was significantly lower as compared to the relative change after the sham stimulation (mean = 12.8), p = 0.014. We also found a correlation between the mean electric field magnitude in the right STG and improvement of articulation precision after anodal tDCS (R = 0.637; p = 0.019). Conclusions: The exploratory study showed that anodal tDCS applied over the auditory feedback area may lead to shorter pauses in a speech of PD patients.

Browse

Recent Submissions