Radio Electronics, Computer Science, Control

IDENTIFICATION OF MOBILE DEVICES BY CORRELATION FEATURES OF THEIR SIGNAL SPECTRA

2024-11-21T09:10:01+02:00

Context. The mass spread of Wi-Fi networks is facilitated by the simplicity of their deployment, high speed, universality, and convenience of use. The development and dissemination of these networks continue despite a number of shortcomings. One of the shortcomings is their vulnerability to various types of attacks, including those based on the forgery (imitation) of identification data. At the same time, there are physical layer characteristics, knowledge of which expands the understanding of the network’s state, can contribute to increasing the reliability of network subscriber identification, and thus prevent a number of attacks. This research is aimed at the theoretical and practical substantiation of the possibility of their application.

Objective. The aim of the study is to assess the application of detailed analysis of signal spectra emitted by devices connected to wireless Wi-Fi networks for their identification. To achieve this goal, it is necessary to analyze the experimentally measured spectra of wireless devices connected to the Wi-Fi network and evaluate the possibility of using the spectrum for the identification of mobile devices.

Method. This work proposes a method for processing the results of measuring the spectra of Wi-Fi device emissions by evaluating the asymmetry coefficient of the Wi-Fi device spectrum’s cross-correlation function. Mathematical modeling was used to assess the effectiveness of the method.

Results. The research results show that the minimum value of the asymmetry coefficient when comparing the template with different positions of one’s own device, and large values of the asymmetry coefficient when comparing templates with foreign spectra. Therefore, this characteristic can also be used for the identification of Wi-Fi devices.

Conclusions. The research results suggest the possibility of applying the proposed method for the identification of mobile devices, which will qualitatively complement existing security models with another feature for detecting unauthorized access.

METHOD OF CONTROL THE MECHANICAL STATE OF THE OPTICAL FIBER OF THE DIELECTRIC SELF-SUPPORTING OPTICAL CABLE DURING OPERATION

2024-11-21T18:52:14+02:00

Context. One of the issues of theoretical and practical research studying phenomena that occur over time and lead to violations of the normative work of optical cables (OC) are ways to ensure and control their reliability during operation.

Today, electronic communication (telecommunications) has already gained significant integration and widespread use due to the urgent need to exchange large volumes of information between users or network devices at high speeds and over long distances, as well as the provision of a wide range of electronic communication services.

The electronic communication service has a high level of demand and consists in receiving and/or transmitting information through electronic communication networks, which is transmitted using electronic communication networks and services.

In an electronic communication network, the transmitting/receiving of optical signals is provided by the fiber optic transmission system (FOTS). It is capable of converting electrical signals from a variety of digital devices into optical signals and transmitting them over fiber-optic communication lines (FOCL), which is the main transmission medium in an electronic communication network.

The problem of ensuring the reliability of the FOCL, which includes a wide range of issues related to the development and production of all its elements, design, construction and technical operation of the communication line, continues to gain more and more importance.

In general, the transmitting/receiving of information between end users equipment, communication nodes, network devices (servers, databases, etc.) takes place through an electronic communication network.

Normative and technical documentations for fiber-optic communication lines regulates the control of the mechanical state of the optical cable during operation, but do not provide the full control of the mechanical state of the optical fiber to ensure the quality and reliability of the line during the specified service life.

As known, to ensure the reliability of the optical cable, as a rule, the permissible elongation of the optical fiber (OF) is ɛpOF < (0.2...0.25) %, adopted during the designing of the cable. However, during operation, the appearance of multiple excess of elongations exceeding these values is possible in the fibers.

Thus, the development and substantiation of methods for evaluating the mechanical characteristics of a dielectric self-supporting optical cable (DSOC) and the method of full control of the mechanical state of the optical fiber is necessary. The last can lead to premature failure of the optical fiber.

Objective. Development and substantiation of the method of control the mechanical state of an optical fiber of the suspended DSOC, as well as assessment of the conditions of deformation of optical fibers in its core with the appearance of longitudinal tensile/compressive loads during operation.

Method. Two ways of evaluating the mechanical characteristics of DSOC and the method of control the mechanical state of its fibers have been developed and proposed. For this, the following characteristics of the cable and fiber are adopted in the work: relative elongation of the cable and fiber (ɛcх, ɛОFх), span length (Lsх) of the line, cable sag in the span (fx) and tensile load (TL) of the cable (Ftlх), which causes longitudinal deformation ɛ. At the same time, the method proposes to control the mechanical state of the optical fiber during the operation of the DSOC by determining its effective relative elongation according to the mechanical, physical and climatic conditions of the line location.

In the paper, it is proposed to measure and calculate the following mechanical characteristics, due to the developed reference data for the selection of the cable type and the climatic zone of the line location, measuring equipment and mathematical tools:

– equivalent mechanical tension in DSOC;

– calculated and actual cable sag in the span;

– actual effective relative elongation of the cable;

– actual tensile load acting on the cable.

The ways and method presented in the work allow a complete evaluation of the mechanical characteristics of the cable and control of the mechanical state of the optical fiber during operation of the DSOC. It creates an opportunity to monitor its changes to prevent the appearance of excessive loads during operation and failure of the fiber-optic communication line.

It is possible to recommend this method for use by relevant departments for technical operation of telecommunication lines and networks based on hanging optical cables.

Results. The work presents the results of the development and justification of the method of control the mechanical state of optical fibers of dielectric self-supporting optical cables during operation. For example, using the developed method, it is shown that in the cable OKL-3-D2A14 produced by PJSC “Odeskabel” in the conditions of the Odesa climate zone (Black Sea region), optical fibers with a span length of 100 m are subject to elongation by 0.16 %, and DSOC is subject to an actual tensile force of 2.722 kN. This result of the control of the mechanical state of the OF established that such span of the line ensures its mechanical integrity within the limits of the permissible deformation of 0.25 % adopted in the design of the cable, but exceeds its permissible tensile load of 2.6 kN.

Conclusions. The scientific novelty of the work results is that, for the first time ways of fully evaluating the mechanical characteristics of the DSOC during operation and the method of fully control the mechanical state of its optical fiber have been developed. It allows to monitoring changes in the mechanical state of the optical fiber of the cable.

OPTIMIZATION OF RADIO SYSTEM STRUCTURES FOR DIRECTION FINDING OF SIGNAL SOURCES WITH COMPLETELY KNOWN PARAMETERS

2024-11-21T19:40:18+02:00

Context. Direction finders are critical components of radar and radio navigation systems, particularly when installed onboard UAVs. The increasing use of unmanned systems has heightened the need for precise direction finding and wide-angle unambiguous measurements. The primary challenge is to strike a balance between achieving high accuracy and maintaining a broad range of unambiguous measurement angles.

Objective. To simultaneously enhance direction finding accuracy and expand the range of unambiguous measurement angles through the statistical synthesis of functionally deterministic signal processing methods in multichannel direction finders.

Method. The approach is grounded in the statistical theory of optimization for radio remote sensing and radar systems. For the specific type of signals considered in this study, represented by functional-deterministic models, the likelihood function is constructed, and its maxima are determined for various configurations of multi-antenna direction finders. The statistical synthesis results are validated through simulation and in-situ experiments.

Results. Theoretical analysis and simulation modeling confirm that in dual-antenna direction finders, there is a trade-off between high resolution and the range of unambiguous direction finding angles. An improved signal processing method is developed for a four-antenna direction finder, utilizing a pair of high-gain and a pair of low-gain antennas. To achieve the maximum possible bearing accuracy within the range of unambiguous direction finder measurements, a new signal processing method is synthesized for a sixelement radio receiver, combining the processing of signals in two amplitude direction finders and one phase direction finder.

Conclusions. The proposed approach achieves an optimal balance between resolution and angle range, making it particularly suitable for onboard systems of unmanned aerial vehicles. Simulation results confirm the effectiveness of the proposed method, highlighting its potential for implementation in modern radio systems.

ALGORITHMIC DIFFERENCES OF COMPLETE AND PARTIAL ALGEBRAIC SYNTHESIS OF A FINITE STATE MACHINE WITH DATAPATH OF TRANSITIONS

2024-11-26T13:40:40+02:00

Context. The problem of algorithmization the search for formal solutions of the problem of algebraic synthesis of a finite state ma-chine with datapath of transitions is considered. The concept of complete and partial solutions of this problem is proposed. The object of research is the automated synthesis of the finite state machine in the part of the function of transitions without taking into account the function of outputs. The basis of the algebraic implementation of the transition function is the author's approach to the transformation of state codes using a set of arithmetic and logical operations. The search for formal solutions to the problem of algebraic synthesis is a complex process that requires the use of appropriate methods and algorithms aimed at special coding of states and mapping of operations of transitions to individual state machine transitions. The use of partial solutions of the problem of algebraic synthesis can contribute to reducing the executing time of such algorithms and reducing the overall design time of digital control devices based on a finite state machine with an datapath of transitions.

Objective. Development and research of algorithms for finding complete and partial solutions to the problem of algebraic synthesis of a finite state machine with datapath of transitions.

Method. The research is based on the structure of a finite state machine with datapath of transitions. The synthesis of the circuit of the state machine involves the preliminary solution of the problem of algebraic synthesis. The result is the so-called formal solution of this problem, which contains two components. The first component is defined state codes, the second component is arithmetic and logic operations mapped to separate state machine transitions. Finite state machine can be synthesized in that case if transformation of given state codes in the process of making transitions is possible using a given set of operations. Verification of this possibility is carried out using the known matrix approach. It involves the formation and element-by-element mapping of two matrices – the matrix of transitions and the combined matrix of operations. As a result, a coverage matrix is formed, which reflects the possibility of implementing (covering) of state machine transitions by specified arithmetic and logic operations. Changing the way of encoding states or the set of operations can give different solutions to the problem of algebraic synthesis with a greater or lesser number of covered state machine transitions.

Results. Using the example of an abstract graph-scheme of the control algorithm, it is demonstrated that the solution of the problem of algebraic synthesis of a finite state machine with datapath of transitions can be considered a situation when one or more state machine transitions cannot be implemented using a given set of operations. It is proposed to call such situation as a partial solution of the problem of algebraic synthesis. The implementation of all transitions by specified operations gives a complete solution of this problem, but the number of complete solutions is always much smaller than the number of partial solutions. Therefore, in the general case, the search for complete solutions takes much more time and, moreover, is not always possible.

Conclusions. The design of a logical circuit of a finite state machine with datapath of transitions is possible in the case of a complete or partial solution of the problem of algebraic synthesis. In the case of a partial solution, those transitions that cannot be implemented by any of the available operations are implemented in a canonical way using a system of Boolean equations. The search for complete solutions generally takes more time than the search for partial solutions. This makes actual the development of algorithms and methods of synthesis of this class of finite state machine, based on the search for partial solutions of the problem of algebraic synthesis.

METHODOLOGY FOR OPTIMIZING THE FUNCTIONING OF THE OPTOELECTRONIC SURVEILLANCE SYSTEM

2024-11-26T14:08:00+02:00

Context. Radar, thermal imaging, and video surveillance means are actively used in the protection of the state border. Together with the appropriate communication equipment, they allow to create optoelectronic surveillance systems, which are the basis for the intellectualization of border protection. The effectiveness of such systems is significantly affected by the peculiarities of their functional and structural design. A rational structural design, even in difficult physical and geographical conditions, allows for a high level of surveillance efficiency. However, the functional component also has a significant impact on improving the system efficiency. In many cases, the functioning of the elements of the optoelectronic surveillance system occurs under conditions of power supply restrictions. Such limitations determine the need for a rational choice of modes of use of certain types of surveillance equipment at certain time intervals in order to ensure effective surveillance, taking into account the time of day and weather conditions. The imperfection of the scientific and methodological apparatus for optimizing the functioning of optoelectronic surveillance systems determines the relevance of this study.

Objective. The aim of the work is to develop a methodology for optimizing the optoelectronic surveillance system functioning by rationally selecting the modes of operation of different types of surveillance equipment in certain time intervals, taking into account the time of day and weather conditions in which they are used.

Methods. The paper sets and investigates the two-criteria problem of choosing the modes of operation of different types of observation equipment of an optoelectronic surveillance system at separate time intervals, taking into account the time of day and weather conditions in which they are used, which ensures maximizing the efficiency of the optoelectronic surveillance system while minimizing the power consumed by active types of surveillance equipment in the presence of boundary restrictions on the efficiency and power consumed by the system. The proposed indicator for assessing the effectiveness of the system allows us to assess the level of impossibility of uncontrolled crossing of the perimeter of the protected area by an intruder. The peculiarity of this methodology is the possibility of ensuring a significant reduction in the level of energy consumption by the system components due to a slight decrease in the efficiency of monitoring.

Results. The paper proposes an alternative approach to assessing the effectiveness of the optoelectronic surveillance system, the idea of which is that instead of assessing the effectiveness of surveillance over the entire sector of the controlled area of the border, the effectiveness of control is assessed only along the perimeter of this area. This approach significantly reduces the computational complexity of the problem of finding the value of efficiency which further simplifies the solution of problems of structural optimization of surveillance systems. A software and algorithmic implementation of the methodology for optimizing the functioning of an optoelectronic surveillance system is proposed. Using the developed software, a rational choice of modes of operation of certain types of surveillance equipment at certain time intervals was carried out taking into account the time of day and weather conditions.

Conclusions. The use of the proposed methodology makes it possible to optimize the modes of operation of the optoelectronic surveillance system, taking into account the limiting factors in terms of efficiency and power consumption when using the same types of surveillance equipment on all towers of the system. A possible direction for improving the methodology is its adaptation to cases where different types of surveillance equipment are used on different towers of the system.

ADAPTIVE FILTERING AND MACHINE LEARNING METHODS IN NOISE SUPPRESSION SYSTEMS, IMPLEMENTED ON THE SoC

2024-11-26T15:00:11+02:00

Context. Modern video conferencing systems work in different noise environments, so preservation of speech clarity and provision of quick adaptation to changes in this environment are relevant tasks. During the development of embedded systems, finding a balance between resource consumption, performance, and signal quality obtained after noise suppression is necessary. Systems on a chip allow us to use the power of both processor cores available on the hardware platform and FPGAs to perform complex calculations, which contributes to increasing the speed or reducing the load on the central SoC cores.

Objective. To conduct a comparative analysis of the noise suppression quality in audio signals by an adaptive filtering algorithm and a filtering algorithm using machine learning based on the RNNoise neural network in noise suppression devices on the technological platform SoC.

Method. Evaluation using objective metrics and spectrogram analysis using the Librosa library in Python. Neural network training and model design are performed on the basis of Python and Torch tools. The Vitis IDE package was used for the neural network implementation on the platform SoC.

Results. The analysis of two noise suppression methods using the adaptive Wiener filter and the RNNoise neural network was performed. In the considered scenarios, it was determined that the neural network shows better noise suppression results according to the analysis of spectrograms and objective metrics.

Conclusions. A comparative analysis of the effectiveness of noise suppression algorithms based on adaptive filters and a neural network was performed for scenarios with different noise environments. The results of objective SIGMOS metrics were obtained to evaluate the quality of the received audio signal. In addition, the possibility of running the RNNoise neural network on the technological platform SoC ZYNQ 7000 was verified.

CONSTRUCTING SENSOR SIGNAL PROCESSING CHANNEL FOR AUTONOMOUS ROBOTIC PLATFORMS

2024-12-01T11:06:32+02:00

Context. The development of autonomous mobile robotic platforms has advanced rapidly, especially in cyber-physical systems where integrating physical components and computational resources is vital. A key challenge in such platforms is the efficient realtime processing of sensor signals under limited computational resources, enabling robots to operate independently of human intervention. Traditional signal processing methods demand significant power, which may limit mobile platforms constrained by energy and resources. This research focuses on restructuring sensor signal processing channels using digital bandpass filters while overcoming technical challenges posed by limited resources.

Objective. The goal is to create an efficient method for processing sensor signals in autonomous mobile platforms with constrained resources. This involves using low-order bandpass filters, capable of adjusting their characteristics and improving quality through sequential connection of identical filters. Reducing the computational load allows for enhanced overall performance of cyber-physical systems, improving efficiency under changing conditions and enabling autonomous task completion. New computational formulas are also proposed to simplify the design and better utilize onboard resources.

Method. The improved method for constructing sensor signal processing channels uses identical low-order frequency-dependent components, sequentially connected to solve challenges faced by higher-order components. This approach simplifies coefficient calculations for cutoff frequencies and improves filter performance by increasing the order and quality. The method achieves a quasilinear phase-frequency characteristic, ensuring minimal distortion in the processed signals, while significantly reducing computational requirements.

Results. The proposed method effectively reduces computational costs while maintaining high performance in sensor signal processing. The new formulas allow for calculating filter coefficients with fewer resources, suitable for autonomous systems. Modelling and experimental verification confirm that this method lowers the computational load and enhances filter performance, enabling more efficient sensor data processing, extended battery life, and improved system reliability.

Conclusions. This research presents an efficient approach to sensor signal processing for resource-constrained autonomous robotic platforms. Sequentially connecting identical frequency-dependent components reduces computational costs while maintaining high signal processing quality. These findings are recommended for real-time applications requiring efficient resource utilization, contributing to improved autonomy and adaptability in mobile robotic systems.

VECTOR-LOGIC FAULT SIMULATION

2024-12-01T12:44:54+02:00

Context. The technological trends of Design&Test computing for the IT industry and academic science are determined by the following directions: in-memory computing, immersive computing, AI computing, focused on energy saving and reduction of computing time when providing services. A mechanism for simulating faults as addresses on smart data structures is proposed, which eliminates the algorithm for simulating input test sets to obtain a test map for logic functionality. The proposed mechanism is focused on the service of SoC IP-cores under the control of the IEEE 1500 standard, which can be perceived positively by engineers in the EDA market.

Objective. The purpose of the research is time- and energy-saving mechanisms for simulating malfunctions, such as addresses, by using read-write transactions of in-memory computing to build a test map of any functionality on smart data structures.

Method. Smart data structures are represented by a logical vector and its derivatives in the form of truth tables and matrices. The test map is a matrix whose coordinates are determined by the combinations of all logical faults that are tested on the binary sets of the comprehensive test. The construction of the test map is focused on the architecture of in-memory computing based on read-write transactions, which makes the simulation mechanism economical in terms of simulation time and energy consumption due to the absence of a central processor. A logical vector as a single component of input data does not require synthesis into a technologically permitted structure of elements. Synthesis of smart data structures based on four matrix operations creates a fault test map like addresses for any logic.

Results. Deductive matrix vectors are effectively used to model faults as addresses in digital structures of any configuration, including convergent branches and feedback loops. The resulting test map is used to find the minimum fault-checking test of the input variables. The proposed fault simulation mechanism technologically easily fits into the architecture of in-memory computing and uses only read-write transactions. The vector logic engine can also be used to test graph structures that are described by a truth table or a logical vector. The truth table addresses used for fault simulation are effectively used for processorless processing of large data in the in-memory computing architecture.

Conclusions. Scientific novelty – a vector-logic in-memory computing mechanism for building a test map is proposed, characterized by the construction of intelligent data structures that reset the fault modeling algorithm. The proposed mechanism has no analogues in the design & test industry in terms of simplicity and predictability of data structure sizes and the absence of a test set modeling algorithm. The practical significance is determined by the application of the mechanism for testing logical functionalities of any complexity to solve verification tasks. Prospects of the research – increasing the object of diagnosis to the scheme, i.e. building a test map of the scheme logical structure.

ENSEMBLE OF SIMPLE SPIKING NEURAL NETWORKS AS A CONCEPT DRIFT DETECTOR

2024-11-26T00:08:03+02:00

Context. This paper provides a new approach in concept drift detection using an ensemble of simple spiking neural networks. Such approach utilizes an event-based nature and built-in ability to learn spatio-temporal patterns of spiking neurons, while ensemble provides additional robustness and scalability. This can help solve an active problem of limited time and processing resources in tasks of online machine learning, especially in very strict environments like IoT which also benefit in other ways from the use of spiking computations.

Objective. The aim of the work is the creation of an ensemble of simple spiking neural networks to act as a concept drift detector in the tasks of online data stream mining.

Method. The proposed approach is primary based on the accumulative nature of spiking neural networks, especially Leaky Integrate-and-Fire neurons can be viewed as gated memory units, where membrane time constant Г_m is a balance constant between remembering and forgetting information. A training algorithm is implemented that utilizes a shallow two-layer SNN, which takes features and labels of the data as an input layer and the second layer consists of a single neuron. This neuron’s activation implies that an abrupt drift has occurred. In addition to that, such model is used as a base model within the ensemble to improve robustness, accuracy and scalability.

Results. An ensemble of shallow two-layer SNNs was implemented and trained to detect abrupt concept drift in the SEA data stream. The ensemble managed to improve accuracy significantly compared to a base model and achieved competitive results to modern state-of-the-art models.

Conclusions. Results showcased the viability of the proposed solution, which not only provides a cheap and competitive solution for resource-restricted environments, but also open doors for further research of SNN’s ability to learn spatio-temporal patters in the data streams and other fields.

HARDWARE IMPLEMENTATION OF AN ANALOG SPIKING NEURON WITH DIGITAL CONTROL OF INPUT SIGNALS WEIGHING

2024-11-26T09:46:53+02:00

Context. Significant challenges facing hardware developers of artificial intelligence systems force them to look for new nonstandard architectural solutions. One of the promising solutions is the transition from von Neumann’s classic architecture to neuromorphic architecture, which at the hardware level tries to imitate the work of the neural network of the human brain. A neuromorphic processor built as hardware implementation of a spiking neural network consists of a large number of elementary electronic circuits that structurally and functionally correspond to neurons. Thus, the design of hardware implementation of a spiking neuron as the basic building element of a neuromorphic processor is of great scientific interest.

Objective. The goal of the work is to design an analog spiking neuron hardware implementation with digital control of input signals by binary synaptic weighting coefficients.

Method. Designing is performed at the logical/schematic and topological levels of the design flow using modern tools of electronic design automation. All proposed schematic and layout solutions are verified and simulated using computer aided design tools to prove their functionality.

Results. The schematic and layout solutions have been developed and investigated for the hardware implementation of the spiking analog neuron with digital control of input signals by binary synaptic weighting coefficients to be the basic building element of a spiking neural network of the neuromorphic processor.

Conclusions. The proposed hybrid design of the spiking neuron hardware implementation benefits by combining the simplicity of analog signal processing methods in the neuron with digital control of the state of the neuron using binary weighting coefficients. The simulation results confirm the functionality of the obtained schematic/layout solutions and demonstrate the possibility of implementing logical functions inherent in the perceptron. The prospects for further research may include the design of hardware implementation for a spiking neural network core based on the developed schematic and layout solutions for the spiking neuron.

IMPACT OF PREPROCESSING AND COMPARISON OF NEURAL NETWORK ENSEMBLE METHODS FOR SEGMENTATION OF THE THORACIC SPINE IN X-RAY IMAGES

2024-11-26T10:32:48+02:00

Context. Automatic segmentation of medical images plays an important role in the process of automating the detection of various diseases in the spine and the use of radiography is the most accessible means of predicting diseases. Over the years many studies have been conducted on the topic of image segmentation. One of the many methods for improving image segmentation is the use of neural network ensembles.

Objective. The aims of this study were to investigate the impact of preprocessing and compare the main methods of neural network ensembles and their effect on the segmentation of the thoracic region, in this study the area was considered which consists of the vertebrae: Th8, Th9, Th10, Th11.

Method. To begin with, the influence of preprocessing of X-ray images was considered, which included the following methods: histogram equalization for contrast enhancement, contrast-limited adaptive histogram equalization, logarithmic transform method, median filter, Gaussian filter, and bilateral filter. To study the influence of neural network ensemble on segmentation quality, several methods were used. Averaging method – a simple half-averaging method. Weighted averaging method – an improved version of the averaging method which uses weights for each network, the higher the network weight, the greater its influence on averaging. Method of cumulative averaging – a modified averaging method in which each ensemble receives an averaged image, after which all the results of the ensembles are averaged. Bagging – method of averaging networks trained on different data, n networks are used, the training sample is divided into n parts, and each neural network is trained on its own subset of data, as a result, the averaging method is used for predictions. Averaging method for a large number of networks – in this method, 100 neural networks were trained, after which the averaging method was used. Method of averaging mask shapes – this method uses a distance transform to average multiple masks into one shape average.

Results. It was investigated that the use of different methods of image preprocessing does not guarantee an improvement in the quality of segmentation of the spine region on X-ray images, but even on the contrary worsens the quality of segmentation. Different methods of combining predictions of neural network ensembles were considered, which made it possible to find out the pros and cons of specific methods for the task of segmentation of X-ray images.

Conclusions. The experiments conducted allowed us to conclude that the use of any preprocessing methods should not be used for segmentation of X-ray images. Also, due to a large number of architectures and methods for combining predictions, the behavior of ensemble methods was studied, which will allow us to further determine the necessary approach for segmentation of X-ray images. Further study of the weighted averaging method and the mask shape averaging method will make it possible to improve the obtained result and achieve even greater success in segmentation.

ENSEMBLE METHOD BASED ON AVERAGING SHAPES OF OBJECTS USING THE PYRAMID METHOD

2024-11-26T11:38:35+02:00

Context. Image segmentation plays a key role in computer vision. The quality of segmentation is affected by many factors: noise, artifacts, complex shapes of objects. Classical methods cannot always guarantee good success, depending on the quality of the image and the existing noise, they cannot always achieve the desired result. The proposed method uses an ensemble of neural networks, which makes it possible to increase the accuracy and stability of segmentation.

Objective. The goal of the work is to develop a new method of combining predictions of neural network ensembles, which can improve segmentation accuracy by combining images of different image sizes.

Method. A method is proposed that averages the shapes of objects depicted on prediction masks. A pyramid of images is used to improve segmentation quality, each level of the pyramid corresponds to an increased size of the original image. This approach allows obtaining image characteristics at different levels. For a test image, a prediction is obtained from each neural network in the ensemble, after which a pyramid is built for the image. All pyramid levels are combined into the final image using SAAMC. All obtained final images for each neural network are also combined at the end using SAAMC. The use of an ensemble of neural networks combined with the pyramid method allows for reducing the impact of noise and artifacts on the segmentation results.

Results. The use of this method was compared with the usual use of individual neural networks and the ensemble averaging method. The obtained results show that the proposed method outperforms its competitors. Application of the proposed method improved the accuracy and quality of segmentation.

Conclusions. The conducted research confirmed the sense of using an ensemble of neural networks and creating a new method of combining predictions. The use of an ensemble of neural networks makes it possible to compensate for the errors and shortcomings of individual neural networks. Using the proposed method can significantly reduce the impact of noise and artifacts on segmentation. Further study and modification of this method will make it possible to further improve the quality of segmentation.

AIRCRAFT DETECTION WITH DEEP NEURAL NETWORKS AND CONTOUR-BASED METHODS

2024-11-26T12:14:45+02:00

Context. Aircraft detection is an essential task in the military, as fast and accurate aircraft identification allows for timely response to potential threats, effective airspace control, and national security. The use of deep neural networks improves the accuracy of aircraft recognition, which is essential for modern defense and airspace monitoring needs.

Objective. The work aims to improve the accuracy of aircraft recognition in high-resolution optical satellite imagery by using deep neural networks and a method of sequential boundary traversal to detect object contours.

Method. A method for improving the accuracy of aircraft detection on high-resolution satellite images is proposed. The first stage involves collecting data from the HRPlanesv2 dataset containing high-precision satellite images with aircraft annotations. The second stage consists of preprocessing the images using a sequential boundary detection method to detect object contours. In the third stage, training data is created by integrating the obtained contours with the original HRPlanesv2 images. In the fourth stage, the YOLOv8m object detection model is trained separately on the original HRPlanesv2 dataset and the dataset with the applied preprocessing, which allows the evaluation of the impact of additional processed features on the model performance.

Results. Software that implements the proposed method was developed. Testing was conducted on the primary data before preprocessing and the data after its application. The results confirmed the superiority of the proposed method over classical approaches, providing higher aircraft recognition accuracy. The mAP50 index reached 0.994, and the mAP50-95 index reached 0.864, 1% and 4.8% higher than the standard approach.

Conclusions. The experiments confirm the effectiveness of the proposed method of aircraft detection using deep neural networks and the process of sequential boundary traversal to detect object contours. The results indicate this approach’s high accuracy and efficiency, which allows us to recommend it for use in research related to aircraft recognition in high-resolution images. Further research could focus on improving image preprocessing methods and developing object recognition technologies in machine learning.

POST PROCESSING OF PREDICTIONS TO IMPROVE THE QUALITY OF RECOGNITION OF WATER SURFACE OBJECTS

2024-11-26T12:52:47+02:00

Context. The significance of this work stems from the growing need for UAV technologies integrated with artificial intelligence, aimed at detecting and identifying objects on the surface of water bodies. Modern needs in water body monitoring, especially in the context of environmental monitoring, protection and resource management, require accurate and reliable solutions. This work demonstrates methods for improving the performance of neural networks and offers approaches for processing NN predictions, even if they are trained on irrelevant data, which increases the versatility and efficiency of the technology.

Objective. The goal of the work is to solve the problem of false recognition of objects on the surface of water bodies, which is due to a decrease in the accuracy threshold for the neural network. This provides more accurate and reliable detection, reducing the number of false positive predictions and increasing the efficiency of the system in general.

Method. It is proposed to add a stage of post-processing of NN predictions, which inherits concepts of min-max suppression used by YOLO models. This algorithm suppresses the re-detection of the object by the network and relies on the cross-sectional area of the detected rectangles. It uses a threshold value of 0.8 for the two points of the rectangle, which can effectively reduce the number of re-predictions and improve the accuracy.

Results. As a result of the implementation of the proposed algorithm and the script created on its basis, a result was achieved in which groups from several predictions are combined and filtered. The received data is stored in the database as found and detected objects. The proposed post-processing algorithm effectively removes redundant predictions while maintaining forecast accuracy. This ensures the reliability of the system and increases its performance in real conditions.

Conclusions. Detected images of objects on the surface of water bodies are stored in the database in the form of records with unique file name identifiers. After tests with pre-taken images algorithm proved it`s persistence against data duplication scenarios. This increases the efficiency and reliability of the monitoring system, ensuring accurate and timely detection of objects on the surface of water bodies.

MODELING OF THE SPREAD OF TUBERCULOSIS BY REGIONS IN UKRAINE

2024-11-21T20:28:14+02:00

Context. Modelling the spread of tuberculosis in Ukraine is particularly relevant due to the increasing number of cases, especially in 2023.

Objective. The aim of this study is to solve modeling tasks by applying modern machine learning methods and data analysis to build predictive models of tuberculosis spread at the regional level.

Method. To model the spread of tuberculosis at the regional level in Ukraine, it is proposed to use several approaches, such as the SIR model, cellular automata, and Random Forest. Each of these methods has its unique advantages and can provide a more detailed understanding of the dynamics of disease spread. The SIR model (Susceptible-Infectious-Recovered) is a classical epidemiological model that describes the spread of infectious diseases in a population. The model assumes three groups of the population: S (Susceptible) – susceptible to infection; I (Infectious) – infected and capable of transmitting the infection; R (Recovered) – those who have recovered and gained immunity. Cellular automata are a discrete model that uses a grid of cells to simulate spatiotemporal processes. Each cell can be in different states (e.g., healthy, infected, immune) and change its state depending on the states of neighboring cells. Random Forest is a machine learning method that uses an ensemble of decision trees for classification or regression. This method can be applied to predict the spread of tuberculosis based on a large number of input parameters. Using these methods will allow for a deep analysis and comprehensive results regarding the spread of tuberculosis at the regional level in Ukraine. This, in turn, will facilitate the development of effective strategies to combat the disease and improve public health..

Results. The results of applying the Random Forest and SIR methods were described and analyzed in detail. For Random Forest, the metrics MSE and R2 were evaluated, showing high prediction accuracy. In the case of the SIR algorithm, visual assessment of the results revealed insufficient accuracy due to model limitations. Comparing the chosen methods with other studies, a conclusion was made about the need to consider more complex algorithms to obtain more accurate results.

Conclusions. Based on the research results, it can be concluded that the Random Forest method is sufficiently effective for predicting vulnerable social groups and that the SIR algorithm is less effective for modeling the spread of tuberculosis. For further research development, it is recommended to consider more complex algorithms and account for additional factors influencing the spread of the disease. Moreover, to better understand further actions to combat the disease, it is advisable to simulate the spread of tuberculosis among the population of Ukraine.

ABOUT OF THE ANNEALING METHOD USING FOR THE TRAVELING SALESMAN PROBLEM SOLUTION WITH THE FUZZY TIME PERCEPTION

2024-11-23T00:42:12+02:00

Context. The article considers a technique for the use of fuzzy numbers and the annealing method for solving the traveling salesman problem, which is formulated as the problem of finding a route to visit a given number of cities without repetitions with a minimum duration of movement. The task of formalizing the algorithm for solving the traveling salesman problem by the annealing method using fuzzy numbers for subjective time perception is posed. The use of fuzzy numbers to increase the accuracy to represent real-world circumstances is proposed.

Objective. The goal of the work is to develop an algorithm for solving the traveling salesman problem based on the implementation of the annealing method with fuzzy numbers representing the subjective time perception for traveling between the cities with the minimum perceived duration of movement along the route.

Method. This paper proposes a method for solving the traveling salesman problem by the annealing method with fuzzy numbers for subjective time perception. A scheme for formalizing the procedure for solving the traveling salesman problem with the minimal perceived duration of movement along the route is described. A variant of the original traveling salesman problem is proposed, which consists in using fuzzy numbers to represent the uncertainty and subjective time perception in traveling between cities as opposed to regular crisp numbers to show regular distance and/or time of traveling. The results of the proposed algorithm for calculating solutions to the traveling salesman problem with minimization of the perceived duration of movement are presented, the obtained solutions are compared with the solutions found by other heuristic methods.

Results. The method for solving the traveling salesman problem using the annealing method with fuzzy numbers for subjective time perception is developed. A variant of the original traveling salesman problem is proposed, which consists in using fuzzy numbers to represent the uncertainty and subjective time perception in traveling between cities as opposed to regular crisp numbers to show regular distance and/or time of traveling. The application of fuzzy numbers makes it possible to perform calculation over possibly uncertain or subjective data, making results more accurate in the case of realistic deviations from the expected mean values in distance coverage. The results of the proposed algorithm for calculating solutions to the traveling salesman problem with minimization of the perceived duration of movement are presented, the obtained solutions are compared with the solutions found by other heuristic methods.

Conclusions. The paper considers a method for formalizing the algorithm for solving the traveling salesman problem using fuzzy numbers for subjective time perception. The use of fuzzy numbers to increase the accuracy to represent real-world circumstances is proposed. The scheme for formalizing the procedure for solving the traveling salesman problem with the minimal perceived duration of movement along the route is described. A variant of the original traveling salesman problem is proposed, which consists in using fuzzy numbers to represent the uncertainty and subjective time perception in traveling between cities as opposed to regular crisp numbers to show regular distance and/or time of traveling.

INNOVATIVE IMPROVED APPROXIMATE SOLUTION METHOD FOR THE INTEGER KNAPSACK PROBLEM, ERROR COMPRESSION AND COMPUTATIONAL EXPERIMENTS

2024-11-23T10:24:56+02:00

Context. Mathematical models of many optimization problems encountered in economics and engineering are taken in the form of an integer knapsack problem. Since this problem belongs to the class of “NP-complete”, that is, “hard to solve” problems, the number of operations required by known methods to find its optimal solution is exponential. This does not allow solving large-scale problems in real time. Therefore, various and fast working approximate solution methods of this problem have been developed. However, it is known that the approximate solution provided by those methods can differ significantly from the optimal solution in most cases. Therefore, after taking any approximate solution as a starting point, there is a demand to develop methods for its further improvement. Development of such methods has both theoretical and great practical importance.

Objective. The main purpose solving of this issue is as follows. The main purpose in performing this work is to first find an initial approximate solution of the problem using any known method, and then work out an algorithm for successively further improvement of this solution. For this purpose, the set of numbers with which the coordinates of the optimal solution and the found approximate solution can differ should be determined. After that, new solutions should be constructed by assigning possible values to the unknowns corresponding to the numbers in that set, and the best among these solutions should be selected. However, the algorithm for constructing such a solution should be simple, require a small number of operations, not cause difficulties from the point of view of programming, be new and be applicable to practical issues.

Method. The essence of the proposed method consists of the following. First, the initial approximate solution of the considered problem and the value of the objective function corresponding to this solution are found by a known rule. After that, the optimal solution of the problem is easily found by a known method, without taking into account the condition that the unknowns are integers. Obviously, this solution can take at most one coordinate fractional value. It is assumed that the coordinates of the optimal solution of the integer knapsack problem and the initial approximate solution may differ around a certain fractional coordinate of the optimal solution of the continuous problem. Then, the minimum number of non-zero coordinates and zero coordinates in the optimal solution is found. Corresponding theorems have been proved for this. It is assumed that the different coordinates of the optimal solution and the initial approximate solution located between those minimal numbers. Therefore, the best solution can be selected by successively changing the coordinates between those minimum numbers one by one.

Results. Extensive calculation experiments were conducted with the application of the proposed method.To have a high quality of this method was confirmed once again through experiments.

Conclusions. The proposed method is new, simple in nature, easy to consider from the programming point of view, and has important practical importance. Thus, we call this solution the innovative improved approximate solution.

SOLUTION OF A MULTICRITERIA ASSIGNMENT PROBLEM USING A CATEGORICAL EFFICIENCY CRITERION

2024-11-25T23:44:10+02:00

Context. The paper considers a problem of assigning a set of employees to a finite set of operations in a multicriteria statement, under condition of a hierarchical structure of a partial efficiency criterion of performing a set of operations, being presented in such a way that each employee possesses a finite set of competencies and each operation has a finite set of characteristics. Numerical and categorical data types are provided for the use as exogenous parameters of the problem. The relevance of the assignment problem being considered is determined by an extremely wide range of practical applications, both in the classical statements and new modifications, the high demand for which is constantly generated by the dynamically developing economic environment. At the same time, a critically smaller number of scientific publications propose means of modeling and solving multi-criteria assignment problems, despite the importance of this type of problems in decision-making, both in theoretical and practical aspects. In general, in conditions of lack of information, the exogenous parameters of the problem cannot be specified in numerical form, therefore there is a need to use categorical data with further numerical coding.

Objective. The goal of the work is to build a multicriteria mathematical model and, on this basis, carry out a numerical study of the optimization assignment problem, taking into account a hierarchical structure of a partial efficiency criterion of the selection of «operation – employee» pairs.

Method. The study proposes a novel method of solving the assignment problem that implemented as a multi-stage process, which includes the stage of transformation of exogenous parameters of the model, given by categorical variables, based on the implementation of the Pareto principle and logistic mapping, the stage of constructing linear scalarization of the efficiency and the cost criteria.

AN IMPROVED MATHEMATICAL MODEL OF THE METHOD OF FULLY PREPARING THE DETERMINATION OF FIRING UNITS FOR HITTING THE INFORMATION AND CALCULATION COMPONENT OF THE AUTOMATED FIRE CONTROL SYSTEM OF COMBAT VEHICLES OF REACTIVE ARTILLERY

2024-12-01T13:34:57+02:00

Context. As part of the automation of the fire control system of rocket artillery combat vehicles, in relation to the preparation of data for firing and fire control, the information and computing process of this system has been improved, namely, the mathematical model of the method of fully preparing the determination of installations for firing projectiles used in rocket salvo fire systems has been improved. In the system of differential equations of the mathematical model of the information and computing process of the component of the automated fire control system of combat vehicles of jet artillery, weighting functions for air temperature, wind influence for the active and passive sections of the projectile flight trajectory and the section of the opening of combat elements have been introduced, which allows determining the weighting coefficients for them for each projectile type.

Objective. To improve the information and calculation component of the automated fire control system of combat vehicles of reactive artillery, by improving the mathematical model of the method of full preparation of the determination of installations for firing on damage. Having proposed a system of differential equations that will take into account the weighting functions of air temperature, wind influence for active and passive sections of the projectile flight path and the section of the opening of combat elements, and will also give the opportunity to determine weighting coefficients for each type of projectile based on them, which in turn will lead to an increase the accuracy of determining firing settings.

Method. The proposed analytical method allows: to calculate the weighting coefficients for each type of rocket, characterizing the process of the approach of the rocket flight to the tabular trajectory and to set the initial conditions necessary for solving the differential equations of the mathematical model of the information-computing process of the component of the automated fire control system of combat vehicles of rocket artillery; to increase the accuracy of determining firing positions when performing firing tasks, which makes it possible to quickly respond to a change in the combat situation by means of changes in the softwaremathematical process of the automated fire control system; effectively and efficiently ensure the development or clarification of textual and graphic administrative and combat documents based on the results obtained using differential equations of the mathematical model of the information-computational process of the component of the automated fire control system.

Results. The improved information and calculation component of the automated fire control system of combat vehicles of jet artillery was tested during the conduct of hostilities. The system of differential equations of the mathematical model of the information-computing process of the component of the automated fire control system of combat vehicles of reactive artillery ensures a timely response to a change in the situation in the information-computational process of the component of the automated fire control system of combat vehicles of reactive artillery during firing and fire control. Provides an opportunity to efficiently and quickly ensure the development or clarification of textual and graphic administrative and combat documents based on the information received during the execution of fire missions.

Conclusions. The calculations based on the proposed system of differential equations confirm the improvement of the information-calculation component of the automated fire control system of jet artillery combat vehicles and allow timely response to changes in tasks in the information-calculation process during firing and fire control, as well as effectively and quickly ensure the formation of formalized messages and documents based on the information received during the execution of a fire mission by units of reactive artillery. Prospects for further research are the creation of agreed mathematical methods, models, algorithms and programs for the implementation of the goals and tasks of firing and fire control when compiling Firing Tables for prospective or received combat vehicles of reactive artillery from partners.

GENERAL PRINCIPLES OF FORMALIZATION OF TECHNOLOGICAL PROCESS CONTROL OF MINING PRODUCTION IN A DYNAMIC DISTRIBUTED SYSTEM

2024-12-01T14:24:35+02:00

Context. The problem of synthesis, modeling, and analysis of automated control of complex technological processes of mining production as a dynamic structure with distributed parameters.

Objective. On the example of the technological line of ore beneficiation, the general principles of formalization of control of mining production processes as a dynamic system with distributed parameters are considered.

Method. The modeling of interactions between individual components of the control system is carried out using the methods of coordinated distributed control. In accordance with this approach, the technological line is decomposed into a set of separate subsystems (technological units, enrichment cycles). Under these circumstances, the solution to the global optimization problem is also decomposed into a corresponding set of individual subproblems of optimizing the control of subsystems. To solve the global problem, this formulation uses a two-level structure with coordinating variables that are fed to the input of local control systems for technological units and cycles. At the lower level of control, sets of subtasks have independent solutions, coordinated by the coordinating variables formed at the upper level.

Results. The paper proposes a method for forming control of a distributed system of technological units of an ore dressing line based on the decomposition of the dynamics of the distributed system into time and space components. In the spatial domain, the control synthesis problem is solved as a sequence of approximation problems of a set of spatial components of the dynamics of the controlled system. In the time domain, the solution of the control synthesis problem is based on the methods of synthesizing control systems with concentrated parameters.

Conclusions. The use of the proposed approach to the formation of technological process management at mining enterprises of the Kryvyi Rih iron ore basin will improve the quality of iron ore concentrate supplied to metallurgical processing, increase the productivity of technological units and reduce energy consumption.