VHVolodymyr Hryha

Volodymyr Hryha

*This is an algorithmically generated author record based on publicly available information.

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article


Volodymyr Hryha has a total of 57 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Open Access

    Development of a multichannel quasi-digital capacitance-to-frequency sensor node with a shared reference channel for an inertial measurement unit of an unmanned aerial vehicle

    by Vadym Hula, Volodymyr Hryha

    This study considers a multi-channel sensor node in the inertial measurement unit of an unmanned aerial vehicle on ring oscillators with a common reference channel in 180 nm technology. The task addressed relates to the fact that there are no quantitative relationships for quasi-digital converters between the number of oscillators, the duration of the counting window, the resolution, the update rate of the digital frequency reading, and the path calculation error. It has been established that three sensor oscillators with one common reference require four oscillators instead of six and reduce the consumption of the analog part from 3.1 to 2.1 mW. It is shown that for measurement times of 1, 10, and 100 ms the resolution of the sensor capacitance reading is 1.0, 0.1, and 0.01 fF, and the counter update rate is 1000, 100, and 10 Hz, respectively; for this purpose, a 24-bit counter would suffice. The behavioral model reproduced the law of frequency control by the number of inverters with an accuracy of one count. The error propagation model for a 10 pF/g sensor gives 1.4–1.6 m for 60 s of hovering and 35–39 m for 5 min of flight. The task has been solved due to the fact that the reference oscillator, which serves three axes, reproduces the temperature delay function common to the crystal, and the common counting window gives three simultaneous normalized signals. The results are attributed to the fact that in the normalized frequency difference, common perturbations are reduced, the quantization error decreases inversely to the window duration, and the temperature term does not depend on the window and dominates already for a deviation of 0.13°C. The results could prove useful for designing small-sized inertial measurement units with capacitive sensors with a sensitivity of a few picofarads per unit of acceleration for generator temperature matching better than 0.03°C

  • Open Access

    Development of a two-channel capacitance-to-frequency conversion interface with a reference channel and a temperature drift rejection metric

    by Vitalii Vintoniak, Vadym Hula, Volodymyr Hryha

    The object of research is the process of suppressing temperature drift at capacitance-to-frequency conversion interface based on two identical five-stage ring oscillators in 180 nm CMOS technology with a 1.8 V supply voltage, a node capacitive load of 2 pF, and a measurable capacitance range of 0–5 pF. Single-ended capacitance-to-frequency conversion interfaces based on ring oscillators have a high temperature coefficient of frequency. It reaches 2000–3000 ppm/°C. This causes a significant capacitance measurement error over the industrial temperature range from –40 to +125°C. This error exceeds the useful signal of a capacitive sensor by an order of magnitude. The problem addressed in this work is the lack of quantitative metric for evaluating temperature drift suppression efficiency in two-channel differential capacitance-to-frequency conversion interfaces based on ring oscillators in 180 nm CMOS technology. A dual ring oscillator architecture is proposed: the sensor and reference channels are built from identical subcircuits, their outputs are buffered and fed to a differential frequency unit, and the output signal is the normalized frequency difference ΔFnorm = (fsensor − fref)/fref. It is shown that the correlation of temperature drifts between the two channels (correlation coefficient r = 0.99998) yields a temperature drift rejection ratio of approximately 40–61 relative to the typical range of 2000–3000 ppm/°C; the temperature coefficient of frequency of the normalized signal is +49.5 ppm/°C, and at the balance point Csensor = Cload = 2 pF the rejection ratio tends to infinity. The combined power consumption of both channels is 1.04 mW at 27°C; the energy per oscillation cycle is stable at approximately 67 pJ across the entire temperature range. The transfer characteristic is monotonic over the 0–5 pF range. The results can be used for the design of interface circuits for capacitive microelectromechanical sensors of pressure, acceleration, and humidity operating under varying temperature conditions without active temperature compensation schemes. The normalized frequency output is directly compatible with digital frequency counters of standard microcontrollers, simplifying integration into measurement systems.

  • Open Access

    Problems and Prospects of Using Database in Modern Information Systems

    by Ruslan Zapukhlyak, Bohdan Dzundza, Volodymyr Hryha, Vitalii Martyniuk, Viktor Dovhyi, Iryna Hatala

    This paper analyzes the challenges, current limitations, and future prospects of database systems in modern information systems. Quantitative assessments indicate that NoSQL solutions can increase transaction processing speed by up to 40% compared to traditional SQL systems under high load conditions, while cloud-based architectures reduce maintenance costs by approximately 30% and improve system availability. The analysis highlights key issues such as scalability, data consistency, security vulnerabilities, and integration with emerging technologies like artificial intelligence and distributed computing. The study provides a comprehensive evaluation of performance, reliability, and cost-effectiveness, offering a foundation for implementing more efficient, scalable, and secure database management practices in both enterprise and research environments. Results demonstrate that adopting modern database architectures can significantly enhance operational efficiency, reduce downtime, and mitigate security risks, making them suitable for handling large-scale, high-velocity data in diverse application domains.

  • Open Access

    Design of a ring oscillator for direct conversion of capacitance into frequency in capacitive sensor interfaces

    by Vadym Hula, Vitalii Vintoniak, Volodymyr Hryha

    This study examines CMOS ring oscillators that are used as converters of capacitive sensor parameters. The issue with most analytical models is their assumption of symmetric stage loading, making them inaccurate for the topology where a sensor connection to a single node introduces asymmetry. The lack of a validated model for 45-nm technology complicates the design of sensitivity and energy efficiency. An analytical model for the capacity to frequency converter that accounts for asymmetric loading has been built and verified. The model is based on the physical principle of summing asymmetric stage delays and a linear approximation of inverter delay versus load capacitance. A parametric analysis was performed in LTspice (sensor capacitance Csensor is from 0 to 2.5 pF) to verify the model. It was determined that the oscillation period has a quasi-linear dependence on capacitance; therefore, the frequency dependence is hyperbolic. The proposed model predicts the frequency with a maximum relative error of no more than 1.55% over the entire simulation range (21.17–29.96 MHz) compared to SPICE data. Key metrics have been analyzed: the average sensitivity is 3.52 MHz/pF, while the instantaneous sensitivity is non-linear, decreasing from 5.57 MHz/pF to 2.15 MHz/pF. Power consumption increases slightly (151.3–155.7 µW), as the capacitance growth is compensated by the frequency drop. Energy per cycle (Ecycle), conversely, increases almost linearly (5.05–7.35 pJ) with a slope of 0.92 pJ/pF. This closely matches the theoretical value of VDD2 = 1.0 pJ/pF, confirming the dominance of dynamic power consumption. The proposed model allows engineers to accurately predict and design the capacity-to-frequency characteristics, sensitivity, as well as power consumption of compact integrated sensor interfaces.


  • Open Access

    Design of improved structures for multi-bit data addition devices in binary codes of the Rademacher theoretical-numerical basis

    by Yaroslav Nykolaychuk, Igor Pitukh, Volodymyr Hryha

    This study investigates the process of designing multi-bit adders based on an improved element base of their components in binary codes of Rademacher theoretical-numerical basis. The task addressed relates to the fact that the known element base of modern single-bit adders does not make it possible to achieve the minimax characteristics of sum formation and carry-save efficiency in their structures in 1 micro cycle. The improved element base of single-bit adders has been built on the basis of the «Exclusive AND» logic element, which has low hardware complexity and high speed of output signal formation. Improved architectural solutions for full and half single-bit binary adders with minimax characteristics of hardware and time complexity in structures of multi-bit adders of various types have been applied. The system characteristics of microelectronic structures of multi-bit adders of various types based on full and half single-bit binary combinational adders with direct, inverse, and paraphase inputs and outputs have been investigated. As a result, it has been theoretically established and practically confirmed that multi-bit adders based on an improved element base have 2 times less hardware complexity and 6 times higher speed. The structure of a cascaded n-bit fast adder has been designed, which has a speed 1.8 times higher than that of the known implementation. Using the VHDL hardware description language and Vivado CAD, a 64-bit cascade adder was modelled and synthesized on FPGA, which has twice the speed of the known adder. The designed multi-bit Rademacher basis adders are applicable to statistical, correlation, and entropy analysis tasks, video image processing, image recognition, as well as various artificial intelligence tasks


  • Open Access

    ADVANCED METHODS OF APPLYING CODING SYSTEMS IN THE DESIGN OF DIGITAL COMPONENTS FOR CYBER-PHYSICAL SYSTEMS

    by Natalia Vozna, Volodymyr Hryha, Lesya Mychuda, Lidiya Shtaiyer

    This study aims to develop highly efficient digital components for cyber-physical systems capable of self-recovery, high-speed data processing, and reliable operation in real-time environments. The paper proposes new approaches to information encoding, particularly for RGB color image representation, and the modeling of neuro-like structures using the code systems of Krestenson, Rademacher, and Haar. The methodology is based on the mathematical foundations of the residue number system (RNS), modular arithmetic, and structural analysis of digital components. Hybrid models of formal neurons, perceptrons with delay lines, and wavelet neurons are employed to solve signal classification tasks. A model of signal self-recovery in a neural bundle is developed, taking into account failures of individual elements and inhibitory effects. As a result, a fault-tolerant mechanism for information transmission in bioneural structures is implemented, along with algorithms for encoding RGB pixels in the R-C and H-C code systems. These methods ensure unambiguous decoding and allow adaptive encoding in the presence of data loss or corruption. The scientific novelty lies in the integration of biological fault-tolerance principles with digital encoding methods based on RNS, providing adaptive signal recovery without the need for complete decoding. The practical significance of the research is in the potential application of the results to digital vision devices, sensor platforms, embedded systems, and high-performance processors for intelligent computing in cyber-physical environments.

  • MOSFET transistor modeling including parasitic leakage and drain resistance

    by V.M. Hryha, V.M. Vintoniak, V.S. Hula

    The aim of the work is to develop and analyze a method for including parasitic source (RS) and drain (RD) resistances in the basic SPICE Level 1 MOSFET model. Relevance: even in the simple quadratic MOSFET model (SPICE level 1), ignoring parasitic resistances can lead to significant simulation errors, especially at high currents. This paper provides a literature review of approaches to account for RS and RD in compact MOSFET models: from the classic Shichman–Hodges model (SPICE Level 1) to modern works on the extraction of resistances and modeling their impact. Theoretical background describes a modification of the Level 1 model by introducing effective voltages VeffGS and VeffDS that account for voltage drops across RS and RD, and analytical equations for the drain current ID in linear and saturation regions with these resistances. Methodology includes a numerical iterative algorithm implemented in Python/PySpice which solves the implicit equation ID(VeffGS, VeffDS). Results show reduction of current and a shift in saturation point when adding parasitic resistances: for a typical NMOS at VGS=5 V, introducing RS=RD=50 Ω reduces ID by ≈16% and increases the saturation voltage by ≈ 0.3 V. Output characteristics graphs and tables of relative current deviation are presented. The novelty lies in the proposed simple iterative procedure to include RS, RD in the SPICE Level 1 model without resorting to more complex models, and the practical value is the applicability of this approach for educational modeling and quick evaluation of parasitic effects on MOSFET behavior. Conclusions: accounting for RS, RD significantly improves the accuracy of Level 1 modeling, bringing results closer to real devices with minimal computational complexity, which is useful for engineering practice and further research.

  • Analysis of the technology for controlling digital data transmission channels with compression in computer systems

    by V.V. Dovhij, V.M. Hryha, B.S. Dzundza, I.V. Svyd, A.I. Terletsky, M.F. Pavlyuk

    When transmitting digital data, various types of errors may occur, including in the data address, and accordingly, there is a need to regulate the data transfer rate. In this case, simple synchronization technology is no longer enough, and it is necessary to use a data link control protocol that would provide data flow control, error detection and protection against the latter. For the effective use of high-speed communication lines, signal compression is used, which allows several transmission sources to use the high bandwidth of the channel. The paper analyzes the technology for controlling digital data transmission channels with compression in computer systems. Two forms of control are considered: frequency-division multiplexing (FDM) and time-division multiplexing (TDM). It is shown that sufficient data transfer speed and bandwidth, low price and the ability to use when operating from autonomous power sources make xDSL technology relevant, especially as a backup Internet access channel.

  • Open Access

    АСПЕКТИ ВИКОРИСТАННЯ ЗМАГАЛЬНИХ ЗБУРЕНЬ ПРИ НАВЧАННІ ШТУЧНИХ НЕЙРОННИХ МЕРЕЖ НА ОСНОВІ ПЕРЦЕПТРОНІВ ЗГОРТКОВОГО ТИПУ

    by І.з. Мануляк, Ю.й. Стрілецький, А.м. Топалов, В.м. Грига

    На основі ряду емпіричних досліджень проведено аналіз впливу збурень на класифікатор нейронної мережізгорткового типу щодо коректності прогнозування, що покликано отримати практичне оцінювання ефектив-ності змагального навчання як механізму підвищення стійкості моделей. Розглянуто архітектури LeNet-5 таResNet-20 на основі набору MNIST. Для кожного зображення згенеровано змагальні приклади методами FGSM,JSMA та C&W, після чого проведно порівняння чотирьох сценаріїв тренування, а саме: оригінальні дані cNat;cNat та FGSM; cNat та JSMA; cNat та C&W, а також їхню комбінацію. Показано, що одноканальне змагальненавчання убезпечує модель лише від тієї збурення, що використовувалася під час тренування, тоді як комбінова-не сумісне навчання на FGSM, JSMA та C&W дозволяє отримати понад 90 % точності на всіх типах збурень,зберігаючи практично ≈99 % точності на тестовому наборі cNat. Додаткові геометричні аугментації покра-щують узагальнюваність, але не замінюють потреби у змагальному навчанні. Отримані результати підкреслю-ють актуальність та доцільність багатоканального змагального навчання для безпекозалежних систем а та-кож дозволяють означити напрями подальших досліджень, зокрема інтеграцію сертифікованих методів роба-стності й масштабування експериментів до складніших наборів даних.

  • Open Access

    Використання векторних інструкцій мікропроцесора x86-64 в задачах лінійної алгебри

    by В.і. Голота, В.м. Грига, Т.г. Бенько, А.в. Моргун, V.I. Holota

    У роботі показано використання векторних інструкцій AVX при розв'язанні задач лінійної алгебри. Розроблено алгоритми і бібліотеку функцій на мові асемблера NASM для векторних арифметичних операцій (додавання, множення) та розв’язування систем лінійних алгебраїчних рівнянь методом Жордана-Гауса. Порівняно швидкодію розроблених програм із використанням векторної SIMD і скалярної SISD технологій при розв’язування систем лінійних алгебраїчних рівнянь різної розмірності. За результатами експериментів отримно прискорення обчислень від 5 до 6 разів при використанні вектроних інструкцій AVX.