Перегляд за Автор "Popovych, P. V."
Зараз показуємо 1 - 3 з 3
Результатів на сторінці
Налаштування сортування
Документ Відкритий доступ DSP Implementation of a Video Bitrate Transcoder(НТУУ «КПІ», 2011) Abakumov, V. G.; Ádám, T.; Formanek, B.; Kane, A.; Popovych, P. V.Solutions, with which using is possible to implement video bitrate transcoder, are researched. Digital signal processors are offered for this purpose. Measurements of implemented transcoder performance parameters for selected general purpose processors and speci alized microcontroller are carried out.Документ Відкритий доступ Investigation of the Influence of the Interference Distribution Law by Different Types of Modulation for Modern Wireless Technologies on the Electromagnetic Environment(КПІ ім. Ігоря Сікорського, 2022) Bakiko, V. M.; Popovych, P. V.; Shvaichenko, V. B.Processes in electronic systems powered by an alternating current (AC) power frequency network with wireless interfaces, in particular, with the ability to transmit audio signals, which, in the presence of radio frequency interference with probabilistic characteristics corresponding to typical distribution laws, affect the parameters of electromagnetic compatibility. The influence of the electromagnetic environment created by electronic systems with a wireless interface, depending on the laws of radio frequency interference distribution and the applied wireless access technology, has been investigated. Simulation of the processes of transmission of sound information in the channels of WiFi and Bluetooth technologies under the influence of interference with probabilistic characteristics, in particular, additive white Gaussian noise, has been carried out. Simulation models of the Matlab application is presented, taking into account the peculiarities of communication and modulation channels used in these technologies, as well as the peculiarities of interference. This model contains transceiver blocks of Bluetooth devices operating in duplex mode, a block of transmission channel properties and a block for generating channel interference. This model provides for the choice of power, transmission channel and intensity of interference. An assessment of the electromagnetic environment for the situation of joint operation of electronic devices with the simultaneous operation of wireless channels of WiFi and Bluetooth technologies has been carried out. The audio fragment for assessing the transmission quality was selected from the sound composition Hard As A Rock. Simulation results are presented in frequency and time domains. The ratio of the number of received erroneous bits in the stream to the total number of received bits for the master and slave devices, depending on the distance between them, has been determined. It is shown that the quality of the transmitted content significantly depends on the features of the distance. Recommendations for improving the structures of electronic systems with several wireless interfaces have been developed, providing for the choice of technology for transmitting audio content based on the results of monitoring the electromagnetic environment.Документ Відкритий доступ Research of the Characteriscs of a Convoluonal Neural Network on the ESP32-CAM Microcontroller(КПІ ім. Ігоря Сікорського, 2023) Sharuiev, R. D.; Popovych, P. V.The paper is devoted to solving the problem of using neural networks for real-time image recognition on lowpower portable devices running on microcontrollers. The ESP-32 CAM microcontroller was used as the target device, on which an artificial neural network was deployed, written using the Python programming language and the Tensorflow library for building neural networks. The performance of the microcontroller and personal computer for object detection using a neural network and their classification were compared in the paper. The image recognition time and percentage of correctly classified objects were compared. The paper shows that the number of training epochs affects the accuracy of object classification in the image. The obtained results show that increasing the number of training epochs increases the accuracy of object recognition using the studied neural network, but a significant increase in the number of epochs does not significantly improve recognition accuracy. The difference in the obtained results for the microcontroller and personal computer image recognition accuracy ranges from 5%.