Advances in Computer Vision and Pattern Recognition

Generators based on unijunction transistors (UJT) are used as sensors of light and magnetic fields, the output parameter of which is generation frequency depending on a measured value. This paper shows experimentally the possibility of a significant increase in sensitivity of such sensors by connecting a bipolar transistor (BT) in a connection variant with an open base in the UJT emitter circuit; the current of this BT is dependent on a light flow and magnetic field. Such a generator based on UJT and BT can be used as a radiation and magnetic field sensor. The UJT can also be applied in industrial optical couplers, where it is an input element with contactless control by the alternation of LED. The possibility of additional contactless control of the optic coupler with an external magnetic field extends their functional possibilities.

The paper presents models that allow us to determine the influence of the primary parameters of a moving radar on the features of secondary signals that occur in the receiver when calculating the range of a moving target. This makes it possible to model various signal processing methods, determine the limitations of these methods, and formulate recommendations for the construction and the hardware and software solutions that should be included in the scheme of a moving radar that tracks a moving target. It has been shown that applying an equilateral triangular LFM law with period 20 μs in a probing signal with carrier frequency 24 GHz and frequency deviation 250 MHz makes it possible to determine the range of a moving target relative to a moving radar. The distance between them is defined uniquely using a coherent processing of received signals. The ideal case of signal propagation has been considered.

The paper presents analytical relationships obtained for calculating the vacuum conductivity of a nonlinear electron beam transport channel, the geometry of which is described by the power law dependence of the channel radius on the longitudinal coordinate. The dependences of the length of the transport channel on the pressure in the process chamber and also on the inlet and outlet radii of the channel are derived and analyzed. The obtained theoretical results are of practical importance for designing sputtering and welding electron beam equipment based on high-voltage glow discharge electron guns. The results of calculated data have been verified experimentally for a cylindrical transportation channel with an input diaphragm. To harmonize the theoretical and experimental data, a correction factor was introduced into the calculation formula to determine the length of the channel, and the value of this factor was adjusted based on the results of the experimental studies. The resulting relative difference between the calculated and experimental data does not exceed 15%. The calculated results obtained in this study can be used to optimize the geometry of the transport channel to achieve minimal losses of electron beam current in the channel.

Radioelectronic equipment (REE) for military purposes is used periodically during exercises; at other times, it is stored on a short-term basis. As is known, the failure rate of elements during storage is reduced by 10–100 times compared to the operation of REE, which leads to the accumulation of hidden defects. Therefore, the maintenance and, if necessary, routine repairs are performed before the field deployment. Depending on the storage conditions (open area, under a canopy, unheated or heated storage) and its duration, the number of hidden defects will vary, which affects the time of preparation (repair) of REE before its use. However, this circumstance is not considered in the known methods of quantitative assessment of partial and complex reliability indicators that do not allow us to predict in advance the time required to restore the operability of REE after short-term storage. The paper proposes a method for estimating the average recovery time of REE after short-term storage, and the functional dependencies of reliability indicators on the initial data and controlled variables depending on the quality of metrological and diagnostic support are obtained and studied. The possibility of practical implementation of conditional diagnostic algorithms (CDA) for diagnosing multiple defects with a quantitative assessment of the total number of checks is investigated. The proposed method is formalized as a block diagram of the algorithm, which allows us to predict the average recovery time of REE after the short-term storage, depending on the initial data. Examples of the use and evaluation of the effect of the implementation of the obtained recommendations in the practice of technical operation of REE under actual conditions are presented. This makes it possible to schedule the work of military repair bodies and effectively use their forces and facilities.

Modern cellular systems increasingly rely on concurrent communication across several discontinuous bands due to broader bandwidth and macro-diversity. Multi-frequency communication is crucial in millimeter wave (mmWave) and terahertz (THz) frequencies, frequently paired with lower frequencies for resilience. Statistical models capable of representing the combined distribution of channel routes over many frequencies are needed to assess these systems. This research presents a broad neural network-based training approach for multi-frequency double-directional statistical channel models. The suggested method involves representing every channel as a multi-clustered set and training a generative adversarial network (GAN) to generate random multi-cluster profiles. The resulting cluster data consists of vectors distributed at various frequencies with random received powers, angles, and delays. Urban micro-cellular connections at 28 and 140 GHz are modeled using ray tracing data to demonstrate the methodology. The model is readily adaptable for multi-frequency link or network layer simulation. As studies show, the model may capture intriguing statistical correlations between frequencies, and the technique involves minimal statistical assumptions.

In the field of circuit design, there is a growing trend toward the design of high-speed circuits with a minimum amount of faults on a nanoscale level. In this way, quantum-dot cellular automata (QCA) is a nanoscale-based paradigm that uses a quantum cell with four dots and two electrons to compute logic bits, comparable to transistor-based CMOS architecture. This article focuses on the low-energy compressor design employing an XOR-XNOR gate and a 2:1 multiplexer. Furthermore, a compressor design provides 152 cells employing a coplanar arrangement in QCA with eight majority gates (MG). The compressor energy dissipation is examined using the QCAPro tool, which has various tunneling energy values. Furthermore, the compressor thermal and polarisation layouts are presented. The novel circuit performance is compared with the best existing circuits on QCA regarding cell count, entire area, MG, and latency to assess the newly designed compressor performance. The proposed compressor is tested using the missing cells in the QCADesigner tool. This design has only 5 test vectors, 100% fault coverage, and is best suited for design for testability (DFT). The proposed compressor can be used with various multipliers, including the Wallace tree multiplier, DADDA multiplier, and higher order 7:3 compressor.

In orthogonal frequency division multiplexing with index modulation (OFDM-IM), data is mapped to subcarrier activation patterns (SAP) using look-up tables or combinatorial schemes. The complexity of these mapping schemes increases with the number of subcarriers. Therefore, they are applied to subgroups of subcarriers, but unfortunately, at the expense of reducing the amount of data loaded to an OFDM block. In this paper, low-complexity direct data to SAP mapping is proposed, the necessary analysis to determine the analytical bit error rate (BER) is presented, and the computer simulations are performed to find and compare theoretical and simulated BER results under conditions of the frequency selective Rayleigh fading channel. The proposed scheme can load more data bits to an OFDM block than conventional schemes. The results from computer simulations demonstrate the superiority of the OFDM-IM system using the proposed scheme over conventional OFDM-IM in terms of BER performance.

The paper presents a quantitative analysis of the directivity characteristics and the operating frequency band of a resonant dipole antenna parallel to a perfectly conducting square-shaped screen. This analysis considers the effect of in-phase diffracted fields at all four edges of the screen with a geometrooptical field in the direction normal to the screen, the dimensions of which ensure the in-phase operation of these fields. The paper uses the solution of the three-dimensional vector problem of diffraction of dipole radiation on a rectangular screen obtained within the framework of the method of uniform geometric theory of diffraction (UGTD). The programs developed in the FORTRAN environment were used to calculate radiation patterns (RP), directed action coefficient (DAC) in the direction normal to the screen and the direction of the main maximum of RP, radiation resistance of the resonant half-wave dipole, and optimal dimensions of a square screen in terms of the DAC maximum for varying the operating wavelengths and screen dimensions in a wide range. It is shown that the optimal dimensions of the square screen amount to 0.7–1.3 of dipole resonant wavelength, and the resonance of the antenna radiation resistance occurs at the electrical dimensions of the square screen in the range of 0.7–0.9. The width of the main lobe of antenna RP in the plane of the electric field vector is about 60° and about 100° in the plane of the magnetic field vector, and it is slightly dependent on the screen size. The band of operating wavelengths is limited by the radiation resistance and the identity of the main lobe of antenna RP; this band varies from 0.75 to 1.5 of the resonant wavelength depending on the screen size. The bandwidth index of the dipole antenna with a square screen can reach unity.

This paper presents a simple technique for finding the length of evanescent zones near the ends of the biconical cavity, where sample insertion openings can be made without significantly affecting the resonance frequencies or the field structure. The technique combines the overlapping domain decomposition method with the collocation method. The interior of the cavity is divided into two overlapping regions, on the boundaries of which collocation points are specified. In each region, the electric field is represented as an eigenfunction expansion. A homogeneous system of linear equations in the expansion coefficients is derived from the equality of the electric fields in both regions at the collocation points. The expansion coefficients are determined as an eigenvector of the system matrix associated with its zero eigenvalue. Using the found expansion coefficients, the distribution of the electric field normalized to the maximum electric field in the cavity is calculated to give the length of evanescent zones, which are defined as regions where the electric field does not exceed a specified small fraction of the maximum electric field in the cavity. Because as few as four pairs of collocation points were proved to be sufficient, the computational effort of the proposed technique is relatively moderate, thus allowing it to be used in the development of instruments based on open-ended biconical cavities.

The paper briefly presents the results of preliminary studies, during which it was determined that the probable reason for the appearance of a significant error in calculating the losses of a single-layer winding of inductive element without a core by using the Ferreira analytical technique in the ultrasonic frequency range is the neglect of the influence of some geometric parameters of the winding, such as its diameter and the number of turns, on the current distribution in the turns. Thus, the calculation expression of this technique needs to be refined by including the neglected geometric parameters in this expression. It has been determined that solving the refinement problem analytically is difficult. Therefore, we propose to solve it empirically. Using computer simulation, the losses of a single-layer winding without a core were determined for the most common in-practice values of winding diameters, conductor diameters, the number of turns, and the operating frequencies. Several graphs of relationships are presented, based on which the types of refining expressions and the method of their introduction into the calculation expression as complementary factors are determined; in addition, the corresponding correction factors are calculated. The results of calculating the values of losses by the refined expression and the initial one are compared. It has been established that the discrepancy between the theoretical and experimental results changed from 59.5 to 11% when the number of turns was changed and from 63 to 4% when the influence of the winding diameter was considered. An alternative method for calculating the value of losses of single-layer windings for most common specific practical cases was proposed. Based on the simulated data, curves of the relationships of the winding losses as a function of the specified geometric parameters were built. The optimal type of approximating function was selected, the approximation was performed, and the approximating coefficients for the conductor diameter, frame diameter, and the number of turns in the winding for different operating frequencies were calculated.

The paper presents models of changes in the instantaneous frequency, instantaneous phase, and the signal itself at any instant of time for a complex broadband two-way infinite continuous signal of an FMCW (Frequency Modulated Continuous Wave) radar with linear frequency modulation (LFM), the law of variation of which has the shape of an equilateral triangle. The proposed models are simple to implement and can be helpful in further discrete-time modeling of the operation of FMCW radar and any processes therein. In this case, obtaining a complex broadband signal from time analysis is possible, rather than performing complex analytical calculations. The models consider various propagation options, including the case of a signal radiated by a moving transmitter, propagating towards a moving target, reflecting from it, and returning to the moving receiver. The ability to determine individual signal values at arbitrary moments over an infinite time interval allows us to take into account not only the Doppler effect, as a change of individual oscillation frequency associated with the movement of the radar and target, but also changes in the shape and duration of a complex signal, as well as the spectral shift and change of its width. The models make it possible to simulate changes in the speed of clocks in the transmitter and receiver due to changes in their ADC and DAC sampling rates.

A circularly polarized dual-band dual-sense coplanar waveguide (CPW)-fed square slot antenna is designed in this paper. The designed antenna consists of two additional patches. A good impedance matching can be achieved by varying the sizes of patches. In this antenna, there are two impedance bandwidth (IBW) ranges of 2.65–5.43 and 5.77–6.19 GHz, and there are two axial ratio bandwidth (ARBW) ranges of 2.88–3.00 and 4.98–5.35 GHz within the IBW. The dual impedance band designed antenna is meant to operate at resonant frequencies of 4.04 and 5.98 GHz. It generates a dual circularly polarized band that operates at 2.94 and 5.2 GHz. They are left-hand circular polarization (LHCP) and right-hand circular polarization (RHCP). Low-cost FR4_epoxy is the substrate material used with dielectric constant of εr = 4.4. The antenna is designed using the Ansoft HFSS software. The main parameters, including reflection coefficient, ARBW, gain, radiation pattern, and radiation efficiency, have been measured, simulated, and analyzed. The lower ARBW band of a proposed antenna can be suitable for some S-band wireless communication applications, and a higher ARBW band can be ideal for WLAN applications.

This paper proposes a method of definition of transistor dimensions for an 8-transistor (8T) cell of a resister file static memory RFSRAM (register file static random access memory) for development of low-power dual-port register files and dual-port SRAM (static random access memory) for power consumption decrease. This method can also be used for 6-transistor (6T) cells of single-port static RAM. The method is based on the analysis of so-called butterfly curves (BC) and the search for such values of transistor dimensions and their threshold voltage dispersion, providing for a given critical minimum supply voltage and the existence of one intersection and one contact of the BC. We compare the obtained characteristics of silicon memory samples and their critical voltage with the simulation results in the reading and writing modes dependent on the supply voltage. Experimental samples of memory cells manufactured at the TSMC factory using 180 nm have technology passed testing at 0.75–1.8 V.

As follows from the analysis of schematic diagrams of transceiver modules of antenna arrays, the phase shifter is a critical functional element. The results of developing phase shifters for the millimeter wave range of waveguide and integral types based on lumped high-speed semiconductor elements are presented. Diagrams and parameters of discrete phase shifters intended to create active phased antenna arrays (APAA) modules are considered. The results of known research studies and developments of integrated elements of phased arrays using the ferrite Faraday phase shifters in 8-mm and 3-mm wavelength ranges are presented that provide electrical scanning of the beam with deflection from the normal to the antenna opening up to 45–60° along two coordinates. The parameters of conical-shaped surface wave antenna radiators and the dependences of the gain and beam width on the ratio of the radiator size to the wavelength are also presented.

The paper considers the effect of temperature on an optical radiation sensor based on an oscillator with single-junction (SJT) and field-effect (FET) transistors, the output signal of which is the frequency proportional to the radiation intensity. Under the impact of temperature, the parameters of transistors change, causing frequency instability in the temperature. For thermal stabilization of the output signal, a second FET2 with the opposite sign of the current dependence on temperature and adjustable thermal sensitivity is connected in parallel to the first FET1. As a result, the temperature effects of all three transistors on the generation frequency are mutually compensated, and the output frequency is thermally stabilized.