Browsing by Author "Parlak, Mehmet"

Now showing 1 - 5 of 5

Metadata only
Blockchain, aI and IoT empowered swarm drones for precision agriculture applications
(IEEE, 2022) Güven, İslam; Parlak, Mehmet; Electrical & Electronics Engineering; PARLAK, Mehmet; Güven, İslam
Recently, the usage of blockchain on swarm UAV applications has gained popularity due to the simplicity and security of the designed frameworks. Applications of drones to precision architecture have also been a prominent research topic for years. However, architectures that use drones as a service (DAAS) for such applications are still lacking some detailed analysis. This paper addresses the known problems and challenges of swarm UAV networks, such as availability, route de-confliction, confidentiality and authenticity, and energy consumption. It presents a conceptual solution and architecture for blockchain and computer vision-assisted precision agriculture applications of swarm UAVs through a combination of early designs of agriculture-specific system components and methods adapted from proposed frameworks for other fields in the industry.
Metadata only
Classifying LPI radar waveforms with time-frequency transformations using multi-stage CNN system
(IEEE, 2022) Güven, İslam; Yağmur, İsmail Can; Karadaş, Hasan Bahadır; Parlak, Mehmet; Electrical & Electronics Engineering; PARLAK, Mehmet; Güven, İslam; Yağmur, İsmail Can; Karadaş, Hasan Bahadır
As the number of radar waveforms in the cognitive electronic warfare applications increases, individual detection and classification performances of each waveform vary furthermore due to their different characteristics. To provide a supervised signal classification in an efficient framework, we propose a multi-stage waveform classification system, where multiple modular blocks are combined to classify 18 different radar waveforms. In the first stage, we transform the signals into time-frequency images (TFIs) using Fourier-based Synchrosqueezing Transform (FSST) and SqueezeNet to classify the signals into two subsets: P1-4 and others. Then, the subsets are used as inputs to two different systems. These systems use different TFI techniques such as FSST and Smoothed Pseudo Wigner Ville Distribution (SPWVD) for processing and convolutional neural network (CNN) architectures such as Squeezenet, ResNet-50, and ShuffleNet for classification. In experiments, we provide supervised classification results at different signal-to-noise ratio (SNR) levels and achieve 98.08% classification accuracy at 10-dB SNR on a diverse set of frequency and phase modulated signals.
Metadata only
Design of multiple-output flyback converter with independently controlled outputs for TV power supply
(IEEE, 2023) Düzgün, Ramazan; Parlak, Mehmet; Yilmazlar, İ.; Electrical & Electronics Engineering; PARLAK, Mehmet; Düzgün, Ramazan
This paper presents the design of a multiple-output flyback converter (MOFC) for a television power supply system. The proposed converter is capable of independently controlling the output voltage of multiple power rails, allowing for efficient and precise power delivery to various subsystems within the TV. The backlight LEDs' current and system supply voltage is controlled independently and the backlight LEDs are fed from the same flyback transformer directly, so there is no need for a buck or boost converter stage. The system voltage of 12V is regulated by the extra N-MOSFET. By eliminating the need for an additional converter stage and adopting smaller transformer sizes, this design not only reduces card sizes and cost but also leads to significant improvements in peak efficiency and power density. The proposed MOFC is explained in details together with the simulation and measurement results.
Metadata only
EV-integrated power system transient stability prediction based on imaging time series and deep neural network
(IEEE, 2021) Behdadnia, T.; Parlak, Mehmet; Electrical & Electronics Engineering; PARLAK, Mehmet
The market penetration of electric vehicles (EVs) has increased drastically. However, the high integration of EV fast-charging stations (EVFCS) into the power systems makes them more vulnerable to severe grid disturbances. In case of a disturbance driving the power system to instability, a fast prediction of stability status is vital for allowing sufficient time to take intelligent emergency control actions. Although various types of machine learning (ML) and deep learning (DL) algorithms have been developed for early detection of instability, the lack of reliable ML/DL models, trained with a realistic dataset, limits their practical application. This paper presents a reliable, accurate DL-based model for early detection of instability in power systems, and compares the results with/without coupling of the EVFCSs. For training our ML/DL models, a large set of realistic phasor measurement unit (PMU) data is generated through a new approach involving a hybrid-type simulation, as an alternative to conventional approaches in data generation. In our experiments, time-synchronized measurements of voltage signals obtained from PMUs are taken as raw input data. Through our proposed method, raw PMU data are encoded into images for developing a reliable and robust convolutional neural network (CNN) model, predicting the stability status of power systems.
Metadata only
Tamper-proof evidence via blockchain for autonomous vehicle accident monitoring
(IEEE, 2022) Parlak, Mehmet; Altunel, Nurkan Fatih; Akkaş, Utku Ayaz; Arıcı, Emir Tarık; Electrical & Electronics Engineering; PARLAK, Mehmet; Altunel, Nurkan Fatih; Akkaş, Utku Ayaz; Arıcı, Emir Tarık
In case of an accident between two autonomous vehicles equipped with emerging technologies, how do we apportion liability among the various players? A special liability regime has not even yet been established for damages that may arise due to the accidents of autonomous vehicles. Would the immutable, time-stamped sensor records of vehicles on distributed ledger help define the intertwined relations of liability subjects right through the accident? What if the synthetic media created through deepfake gets involved in the insurance claims? While integrating AI-powered anomaly or deepfake detection into automated insurance claims processing helps to prevent insurance fraud, it is only a matter of time before deepfake becomes nearly undetectable even to elaborate forensic tools. This paper proposes a blockchain-based insurtech decentralized application to check the authenticity and provenance of the accident footage and also to decentralize the loss-adjusting process through a hybrid of decentralized and centralized databases using smart contracts.