Search

The study of information and psychological security raises a number of issues related to ensuring human security and preserving human health in connection with various types of information threats: targeted information influence on the population through the media, the Internet, which can lead to negative socio-political consequences; incomplete realisation of citizens' rights to receive and exchange reliable information; provoking social, ethnic, religious tensions through the activities of certain media; manipulation of Predicting the results of information and psychological influence is an important and urgent problem not only on the scale of an individual, but also for the national security of the state. The purpose of this paper is to increase the efficiency of forecasting information and psychological influence by modifying the model of information and psychological influence based on the analysis of influence factors. The paper reviews existing models for predicting the results of influence, from which a model was selected for modification to improve the forecasting results. The result of the work is a methodology that takes into account the channels of information and psychological influence, the emotional component of memes and the methods of counteraction that can be used by the opposing side, which allowed to create a theoretical basis for further practical research. The results obtained can be used to predict the information and psychological impact on society and/or to choose methods of counteracting this impact to reduce its consequences. [ABSTRACT FROM AUTHOR]

The work developed a modification of the algorithm for detecting cloning of frames in video sequences. It is impossible to imagine the life of a modern person without the constant presence of digital technologies in it. They have become an integral part of our daily life. Everyone has a smartphone, almost everyone owns a personal computer or laptop. Photo and video recording technologies, as well as the distribution and publication of images and videos, such as social networks, have become an integral part of digitalization. Digitization has also spread to very important areas for human life, such as forensics or mass media. In order to have confidence in video materials, which is extremely important in criminal investigations, court cases and information networks, it is necessary to develop the field of exposing modifications and falsifications in video and photo materials, develop new methods that allow you to verify the integrity of the video or detect falsification, and modify and improve already developed ones. The paper analyzes the methods for finding falsifications in video sequences and modern formats and algorithms for saving and compressing video sequences. As a result of the analysis, the current state of development in the field of detection of fake videos was revealed and the most popular video formats were determined. Experiments were conducted with videos of different formats and resolutions, data were collected and analyzed, after which the necessary threshold values of the Pearson correlation coefficient were established for each format studied. Threshold values for searching for cloned frames were proposed in the work. The threshold value is chosen depending on the format and resolution of the video. Parameters such as video format and resolution were taken to modify the algorithm for detecting cloning of frames in a video sequence by analyzing the threshold values of the Pearson correlation coefficient. A software implementation of the modification of the algorithm for detecting cloning of frames in video sequences was created. Errors of the first and second kind are cited in the work. [ABSTRACT FROM AUTHOR]

The paper aims to develop a new method for the design of hybrid multivariable automatic control systems. We determine a hybrid control system as a system that includes two different types of parallel operating controllers. The paper's aim is achieved by solving the following tasks. The first task is to develop a method for the design of hybrid control systems. Investigated systems include PI controllers and a linear-quadratic controller. The second task is to investigate the resulting control system. The investigation is based on direct quality indicators, robustness, and fragility analyses. The most significant result of the paper is the development of a method for the tuning of a linear-quadratic controller, which should work with the decentralized control system in parallel. BLT (Biggest Log Modulus) method is used for decentralized control system tuning. BLT is a well-known method of PI controller detuning. The idea of BLT is to ensure the robustness of multivariable control systems tuned by Ziegler Nichols's rules. The proposed method for the design of the control system is not complicated. The possibility of correcting the tunings of decentralized controllers remains. Design of the new hybrid automatic control system for the distillation plant made using the developed method. The analysis of the dynamic properties of the designed control system showed that the use of a parallel linear-quadratic controller provides a small percentage of overshoot in the control by the reference and high accuracy of stabilizing the values of controlled variables at the largest disturbances. At the same time, the robustness does not decrease and the control system remains fragile, i.e., it is possible to change the settings of the controllers without significant loss of robustness. [ABSTRACT FROM AUTHOR]

The paper describes the development of an application for data encryption with format preservation. Basic definitions such as personal information, identifying information are provided. Over the past few years, a growing number of personal data breaches have resulted in the loss of data integrity for hundreds of millions of user records. Such attacks are aimed at both large companies and small businesses. The cyber-attacks aimed at stealing personal data from enterprises and it is ensured due to the negligence of companies regarding the conscious encryption and masking of personal data of employees and customers. The paper reveals the concept of format-preserving data encryption, as the process of encrypting data in such a way that the output data remains in the same format as the input data. Two modes of operation are considered in format-preserving encryption - FF1 and FF3. When analyzing the existing FPE modes with format preservation, the FF3-1 mode was selected due to the ability to solve the problem of encryption on small data domains. FF3-1 mode was implemented in a software application. The work also reveals the concept of pseudonymization and methods of achieving it. Pseudonymization can be achieved using various methods such as data masking, encryption or tokenization. Data masking process change the value of data while leaving its input format. The goal is to create an output that cannot be deciphered or reverse engineered. The method of data masking was implemented. Thus, the main task was to implement methods that would help companies encrypt and mask data with minimal contributions to restructuring and preserving functionality on the fly. In order to achieve the goal of clear pseudonymization, so that the fact of replacing real data is imperceptible, it was decided to modernize the approach to masking and encryption for fields such as first name, last name, and email address. The methods of pseudonymization and data fields masking: first name, last name, and email address, and the use of the FPE method in FF3-1 mode for fields such as phone number and credit card were implemented. An improved method of pseudonymization and masking of data fields is implemented in the software application. [ABSTRACT FROM AUTHOR]

The paper considers the improvement of the method of detection and localization of cloning areas in digital images. A significant role in the modern world is played by digital signals, in particular, digital images used in science, medicine, court proceedings, the press, etc. Today, everyone has a camera, if the event is significant, you will have thousands of photos at your disposal. However, changing the photo is very easy. Even amateurs can do it. Often, such falsifications are almost impossible to detect with the naked eye. One of the most commonly used operations in the forgery of digital images is the cloning operation, in which a part of a digital image is replaced by a part of the same digital image. The paper considers the method of detection and localization of cloning regions based on the Pearson correlation coefficient. The basic method uses 8x8 blocks as a compromise between image processing time and the accuracy of cloning region detection. Increasing the block size will speed up the base method. Another way to improve the method of detecting and locating cloned blocks is to use markers. Markers are predefined pixels in a block that are used to define the perspective of the block to identify the cloning area. Markers for square blocks and blocks of complex shape are proposed in the work. Blocks of a complex shape in the work are considered to be blocks consisting of a suitable set of pixels, which are used to detect and localize areas of cloning. The main steps of the improved method of detection and localization of cloning regions in digital images are presented. Evaluation of the effectiveness of the improved method of detection and localization of cloning areas was performed by measuring the time of detection and localization of cloning areas before and after improvement. [ABSTRACT FROM AUTHOR]

The system of distributed data storage includes, in addition to the devices themselves, on which information will be stored, as well as communication channels through which it is forwarded. When transmitting information to considerable distances, there are a number of new problems that need to be solved. An analysis of the current state and development of data transmission and storage systems shows that it is necessary to use various technical and software methods of information protection from distortion, loss or theft. The use of only one of them is not effective, so they usually use their combination. The most widely used data protection methods are the methods based on the use of correction codes, particularly the residue number system (RNS) codes. Due to the presentation of information in the form of remainders from the division of data into each element of the system of mutually simple modules, a reduction in the number of digits with which operations are carried out is achieved. The simplification of information collection and transmission system due to the presentation of information in the form of residuals is investigated in this paper. The initial range of calculations expansion implemented by the introduction of additional backup modules is suggested for data protection, that allows to increase the probability of error detection to 100%. The advantages and disadvantages of detecting and correcting errors with the use of corrective codes in RNS using the methods of calculating the syndrome and calculating the number projection is analyzed. The number of information modules in the designed systems can vary depending on the task, therefore, in the conducted studies, it was determined that with an increase in the number of information modules and combinations of distorted symbols, the number of undetected errors in percentage compliance remains practically unchanged. It is suggested to use RNS without additional error correction for designing information collection systems, as it is possible to choose a set of modules in such a way that the maximum number of errors that will not affect the system functionality. [ABSTRACT FROM AUTHOR]

The paper aims to develop a new method of designing one-dimensional automatic control systems with P, PI, and PID control laws. The method scope is static and astatic plants, and it aims to minimize the influence of measurement noise on the dynamic properties of the control system. Achievement of this aim should be without complicating the structure of the control channel. The scope of the method is the problems of stabilization of relatively fast processes, which most of the time are controlled by regulators in the mode when the control error is close to the measurement error. Examples of such problems are: temperature stabilization in a heat exchanger or flow (pressure) control. The goal is achieved by solving the following tasks: 1) to develop rules for PID-type controllers for static and astatic control plants with a dead tune, which achieve a guaranteed value of the ratio of the output dispersion to the input (according to the white noise model); 2) to investigate the rules on the model of the technological plant, which takes into account the peculiarities of the automation in the control channel. The most significant result of the work is a method that represents a set of simple rules similar to the rules of other known methods of tuning PID-type controllers. A plant's reference dynamic mode is supposed in the form of a combination of a first-order aperiodic or integrating link with a dead time link. The method is also suitable for hard-controlled plants with a significant or dominant dead time. An example of the application of the developed method for controlling the temperature in the heat exchanger of a beer distillation column using a PI controller is demonstrated. It is shown that the method allows a reduction of the dispersion by a quarter compared to the controller obtained using the pidtune program. Similar results are achieved for other PID-type controllers. [ABSTRACT FROM AUTHOR]

The cryptographic S-box is the crucial component of modern ciphers which determines their efficiency, cryptographic security, and performance. Today, the development of quantum cryptanalysis, as well as the appearance of possible attacks on cryptographic algorithms by describing them using many-valued logic functions made urgent the task of developing algorithms for the synthesis of S-boxes, which would be characterized by high quality not only when represented by component Boolean functions, but also with any other representation by component functions of many-valued logic. At the same time, most of the existing methods of synthesis of S-boxes presented in the literature are focused only on the research of their cryptographic quality when represented by component Boolean functions. In this paper, on the basis of S-boxes of length N=16, which corresponds to the strict avalanche criterion of component Boolean and 4- functions, we propose a method for synthesis of a set of high cardinality equal to J=117588 of S-boxes of practically valuable length N=256, which simultaneously corresponds the strict avalanche criterion of component Boolean functions, the strict avalanche criterion of component 4-functions, as well as the criterion of correlation immunity of component Boolean functions, i.e., they have ideal matrices of correlation coefficients between output and input vectors. The high cryptographic quality of the developed S-boxes when they are represented by component Boolean and 4-functions makes it possible to recommend them for practical use both in the tasks of increasing the effectiveness of existing cryptographic algorithms and in the development of promising ciphers, while the cardinality of the class of synthesized S-boxes allows them to be used as a long-term key. [ABSTRACT FROM AUTHOR]

In today's world it is impossible to do without the use of various communication systems. They are now the integral part of our lives. Almost every modern person uses one or another system of information exchange with other people. Mobile platforms are most often used for information exchange. The vast majority are platforms such as Android, iOS and Windows Phone. Information transmitted from one user to another can often be confidential or restricted. So, the appropriate methods of information protection should be used for this purpose. Today, information security researchers focus on improving the performance of cryptographic algorithms and the primitives on which they are based. Nevertheless, the issues of building of integrated security systems for instant messaging systems remain insufficiently considered. This paper is devoted to the development of a comprehensive application protection system for instant messaging. Any IM application consists of three main parts. The first component is the program code executed on a mobile platform. The second part is the communication channel. The last part is the program code executed on the server side. This paper discusses each of the main parts for building a comprehensive application protection system for instant messaging. The choice of certain components for building a security system is substantiated, the schemes of operation of the program code executed on the Android mobile platform are given, the choice of forms of transmission of user information through the communication channel is substantiated. Recommendations for choosing a communication channel and ensuring its security are given. The schemes of the server side of the application are also given. The choice of certain hashes functions, functions of input information checks, approaches to the implementation of the protection system, storage system and processing of user information on the server side is substantiated. The developed application also shows some shortcomings of the mobile platform Android in terms of information security. It is shown how attackers can take advantage of these shortcomings, and suggestions to prevent their use are provided. The results of the work can be used to protect information in practical systems. [ABSTRACT FROM AUTHOR]

We research a scheduling algorithm for a feasible schedule construction on one machine by the total earliness of jobs minimization criterion. We propose an improvement to this algorithm. During the research, we have developed a software system to verify reliability of the algorithms. We compared the schedules built by the algorithm with those constructed by an exhaustive enumeration. We calculated the percentage of a non-exact schedule construction by the investigated algorithm for different characteristics of the schedule. The object of research is the process of schedules' optimization and their investigation. The purpose of the paper is to increase the efficiency of feasible schedules construction process for one machine in the presence of penalties for jobs completion prior to their due dates. We need to use appropriate economic and mathematical methods, in particular those from scheduling theory, to organize the production and planning on an enterprise efficiently. Scheduling problems have a great application value today. Rapid development of communication and electronics technologies increasingly requires the construction of schedules. The schedules are related to functioning of industrial enterprises and the service sector, education, transport and many other areas. Tasks of scheduling theory include study of the computational complexity of the problems and development of exact, approximation and heuristic algorithms for their solution. Most papers are devoted to combinatorial approaches' development. However, as practice shows, possibilities of combinatorial algorithms are substantially limited by the dimension of the problems they solve. This is why we believe in importance of the research and improvement of the scheduling algorithm to minimize the total earliness of jobs on one machine. We use combinatorial optimization methods, scheduling theory, heuristic algorithms and statistical analysis. The scientific novelty of the results consists in modification to the existing scheduling algorithm that minimizes the total earliness on one machine. New algorithm has better efficiency than the existing ones. [ABSTRACT FROM AUTHOR]

In the 21st century, cyber security is becoming an important issue for the national security of Ukraine. This is due to the active formation of the sixth technological order and the risks that society faces today. Currently, there is an improvement in the technical level of the implementation of cyber threats, the latest tools, algorithms and mechanisms of cyber-attacks are being introduced and successfully used, and new ideas are being developed to manipulate public behavior and opinion. The impact of the COVID-19 pandemic and the quarantine restrictions associated with it must be taken into account. The impact on the systems of the world order of these restrictions will be long-term, which will lead to an increased role of information technology and electronic communications in everyday communication and at work. This causes an increase in the vulnerability of information and personal data processing processes. It is of great importance to strengthen the capabilities and effectiveness of the national cybersecurity system to counter cyber threats in the modern security environment, to ensure a high level of protection of personal data, organization data and information resources by introducing appropriate additional measures in the state and organizations. The paper considers the problem of the spread of cyber threats in almost all areas of human life, describes the importance of appropriate changes in algorithms, strategies and tactics to counter them, emphasizes the importance of the ability to identify vulnerabilities and cyber-attacks as quickly as possible for accurate and effective response, taking appropriate measures in order to increase the level of cyber security to better protect organizations. Due to rapid changes, the digital space requires a greater ability of the cybersecurity system of both organizations and the state as a whole, its effectiveness and balance, which can change depending on changes in the security environment, ensuring their safe functioning. The purpose of further research is to study the strategies used by cybersecurity leaders, to collect best practices in countering cyberattacks to understand how various organizations have adapted to new forms of work, maintaining / reducing their vulnerability to cyber risks, as well as the experience used to maintain employee confidence, team development. Work, protecting the mental health of team members, reducing insider risk and the human factor. The article contains a collection of data on recent research conducted as part of the study of the dynamics of telecommuting, remote work and cybersecurity risks both due to quarantine restrictions due to COVID-19 and not due to this reason. [ABSTRACT FROM AUTHOR]

Budget planning has always been and remains an important part of any person's life. The purpose of this work is increasing users savings through the use of developed web service for dynamic generating a forecast, which is based on user's basic model and automatic selection of combinations of forecasting methods in each specific case by analyzing the quality and quantity of the user's story. Correct distribution of income, planning expenses and increasing savings remain quite relevant. Correct budgeting and expenditure require more and more effort to achieve high quality, consistent forecast estimates. In addition, questions in these areas are associated with a high degree of uncertainty. When analyzing such processes, the main problem is the ability to generate high-quality forecasts. Due to the flaws of the existing solutions, it was proposed to develop its own web service for the dynamic formation of a budget forecast depending on the user's story. The paper proposes its own models, the importance of each of the elements of the model is analyzed, shows how the basic model is transformed into a user's model, the use of models in the method of dynamic generating a forecast is shown. Thanks to the use of the developed models, the method allows for each specific case, taking into account the individual designated elements for constructing a forecast. The presented method analyzes the quality and quantity of the user's story and automatically selects a combination of forecasting methods in each specific case to solve the economic and financial issues of users. The increase in household saves after using the application based on the proposed models was analyzed. The analysis showed an average increase in income of 1.3 times among 74 families compared to the amount of savings saved without using the developed application based on the proposed models and method. [ABSTRACT FROM AUTHOR]

Successful identification of a person in a real-time video stream is a critical function in a wide range of relevant subject areas. On the other hand, such a task is complex and can be effectively solved only with the help of modern artificial intelligence approaches. In this work, the previously developed MTCNN approach is used as a key algorithm for face image recognition and vectorization, which uses a cascade of several neural networks to identify key points of the face of a recognizable personality. In addition to the main algorithm, an integrated system was implemented that provides the functionality of person identification available to the end user. Additionally, face identification modules were implemented based on vectors representing a person in a video frame, a subsystem for storing and replenishing the database of recognizable persons, and a convenient user interface. Modifications of the MTCNN architecture proposed in this paper made it possible to achieve real-time performance and acceptable identification quality at the level of 0.92 according to the F1-metric. [ABSTRACT FROM AUTHOR]

Today, video surveillance is the most sought after system for security and monitoring purposes. The development of video surveillance systems opens up new opportunities not only for fixing offenses but also for preventing them. With regard to object tracking on video, it can be noted that this need arises in many areas of life, such as security systems, the creation of systems for the automated analysis of sports competitions, the quality control of processes, the creation of human-machine interface. It is essential that CCTV users not only be able to use the system, store and process data, but also obtain the necessary information about a particular object that is displayed on video using a tool such as object tracking. This paper describes a developed algorithm for searching and tracking objects on video. The algorithm developed consists of the following steps: finding or detecting objects and tracking objects on video. You should also consider the metrics that will be used to evaluate the similarity of the two blocks in the frame. Most video quality metrics are based on still image quality metrics, which is why video images are compared frame by frame. In this regard, the metrics are only applicable to compare videos that have the same number of pixels. The paper analyzes the existing methods of object detection, namely human faces. The Viola-Jones method was chosen to detect objects on video. This method consists of several steps, one of which is to use the Haar traits to find the desired object. Both standard and additional features of Haar were considered in the paper. The Viola-Jones method was programmatically implemented and an experiment was conducted. You should also consider the metrics that will be used to evaluate the similarity of the two blocks in the frame. [ABSTRACT FROM AUTHOR]

The paper proposes a modification of the steganographic method of embedding a digital watermark into Wavelet transform domain of a digital image. The proposed method embeds a watermark in the high-frequency detail coefficients of a three-level Wavelet transform using the Dobeshi-43 wavelet, which ensures the high quality of the resulting stego. Stego is a result of embedding of digital watermark into container and is saved in a losses format that is caused by wide dissemination of such images due to their small size. Discrete cosine transform is applied to a digital watermark, thereby increasing the efficiency of detecting the presence of a watermark in a stego. In work the analysis of efficiency of detecting of the digital watermark depending on the threshold values defining the choice of the detailing coefficients used for embedding is carried out. A result of this analysis is establishment the fact that increase in thresholds without application of Discrete cosine transformation to the watermark leads to considerable deterioration in accuracy of detecting. Conversely, the proposed method with the threshold values Т1=80 and Т2=90 of the choice of detailing coefficients in conjunction with the Discrete Cosine Transformation of the watermark matrix increased the percentage of correctly identified stegos. The results of the experiments showed the resistance of embedding a digital watermark to compression attacks (even with image quality QF=5), imposing noise, sharpening, shifting, for which the detection accuracy averages 97%. Also, the method is resistant to scaling provided QF>50. This paper compares the resistance of the original method of embedding a digital watermark and different variations of the proposed modification of the steganographic method to complex attacks (compression in combination with other types of digital image processing). [ABSTRACT FROM AUTHOR]

This paper discusses the developed algorithm for creating a panoramic video of several video sequences. The algorithm was developed for video surveillance systems. In recent years, CCTV systems have become the basis of a robust security system. CCTV systems allow you to obtain information about the current status of a protected object by collecting, processing, archiving, storing, displaying, and analyzing this information. The developed algorithm processes a video sequence of several cameras and creates a single video sequence, which contains all the information from the processed video sequences, based on the search algorithms for a singular point in the images. Special points are those points by which you can classify images, recognize it. They determine the peculiarity of the image, its uniqueness. The paper considers the most well-known and widely used algorithms for finding a singular point. The SIFT (Scale Invariant Feature Transform) and SURF (Speeded Up Robust Features). The SIFT method looks for special points using the Gaussian pyramid and the Gaussian difference. Descriptors are constructed by computing a histogram of oriented gradients in the vicinity of a singular point. To find special points, the SURF method uses the Hessian matrix. The basic steps of the developed algorithm are considered: search of special points, selection of one or more special points by criterion, stitching of video frames. Here are some examples of how to create a panoramic video algorithm for the first video frames. SIFT and SURF methods have been investigated in situations where the image is scaled, rotated, dimmed and blurred and a comparison of SURF and SIFT methods is made. The paper presents conclusions about the use of SURF and SIFT methods in the developed algorithm for creating panoramic video. [ABSTRACT FROM AUTHOR]

The actual standard for industrial technological processes and many other branches of automation is the PID-type controller. The problem of synthesis of such controllers has been studied in the literature in sufficient detail. The advantage of such controllers, which influenced on its wide popularity, is that it is a controller in which the automation system tuner can empirically change the numerical coefficients (from 1 to 4 depending on the controller type), based on the direct criteria of transient quality: the maximum deviation, overshoot, setting time, as well as on signs of instability. The problem of optimal synthesis of the PID-type controller based on typical models can be generally considered as solved. With the development of computer techniques and engineering mathematical software, the problem of synthesis of the optimal PID-type controller for any control object is solved by combining frequency criteria and optimization algorithms. Other approaches are also used, which are mainly designed for multivariable control objects. This paper solves the problem of synthesis of a discrete SISO (Single Input and Single Output) controller for a stable linear model of arbitrary order with dead time using engineering mathematical software. The main idea of the method is a modification of the classical Dahlin controller. The proposed controller has the following advantages: simple software implementation, no necessity to solve the optimization problem, high robustness, possibility to take into account all dynamics, and simple tuning procedure. A study of the controller's operation for typical forms of models was carried out and attention was paid to the problem of neutralization of stepwise disturbances when controlling integrating plants. [ABSTRACT FROM AUTHOR]

The issues of constantly changing geopolitical situation in the world and the desire of leading powers to obtain or strengthen the new and existing spheres of influence, as well as the urge of other countries to maintain or regain their former positions, leads to the use of new methods of hybrid warfare. One of the methods employed for information environment (i.e. network system) risk assessment is the probability of an attack on it. Information Security Assessment is determined by the values of the interaction between the attacking and defensive sides. This raises the challenge of the need to assess and predict the actions of both the defensive and the attacking sides. The process of interaction between attackers and defenders can be viewed as a game play, understood as the process of strategic interaction between the opposing or cooperating players under certain constraints and their plans of action. Deriving from this interpretation of the game, resolving the issues of hybrid war on the basis of the theory of games, predicting the actions of the attacking side, and determining the decisions by the defensive side is proposed. As in any theory of applied mathematics, game theory provides some formal structure. It is assumed that all possible actions of each player can be accurately established and for each combination of such actions the expected outcome of the competition can be quantified. The goal of each player is to maximize their expectation of winning. The player can assume that each of his opponents is busy with a similar optimization. The paper considers various possible situations of interaction between the attacking and defending sides. Functional expressions are given to determine the optimal strategies of the opposing sides and to find the guaranteed maximum level of the cost of an attack. Methods for solving these expressions are proposed. In particular, these include the following: methods for reducing matrix and continuous games to linear programming problems, methods for reducing continuous games to matrix games, algorithms for the approximate solution of continuous games by the random search method, and methods for modeling games. [ABSTRACT FROM AUTHOR]

Steganography and cryptography are at their peak these days. One of the types of steganography is hash steganography. This is a fundamentally new direction in the transmission of secret information. Hash steganography methods are based on the fact that instead of embedding secret information into a container, we transmit a specific sequence of digital signals and recover messages from their hash codes. Digital images are commonly used for these principles to work. However, digital audio files, thanks to the many services and messengers, are becoming an increasingly popular method of transferring information. Therefore, the development of hash steganography methods based on the use of audio signals is urgent. The aim of this work is to improve the efficiency of transmission of a secret message by modifying the hash steganography method. In this paper, the Shazam's algorithm is used to obtain hash codes of a digital audio signals that are resistant to attacks and file transformations. this method does not use embedding additional information in the audio file as a container. In this method, it is proposed to use a hash code (hereinafter the hash) obtained from audio files to build a stegan message. But when transferring and converting audio files, the integrity of their hash code may be compromised due to possible loss of quality and noise. This can interfere with matching files and hash codes and restoring the secret message. One of the popular audio file identification services is the Shazam software application. Shazam allows you to correctly identify audio tracks, even in the presence of noise, extraneous sounds and more. Consider the main steps of this algorithm for their use in obtaining attack-resistant hash codes of audio signals. This algorithm allows representing audio as a set of special points, which are then converted into hash codes using the MD5 method. Secret messages are recovered from these hash codes after receiving a specific sequence of audio signals. The developed algorithm has no analogues for comparison, but it is resistant to attacks. [ABSTRACT FROM AUTHOR]

In this paper we study the effect of perturbing influences aimed at the spatial domain of digital images on the change of color triplets in the matrix of unique colors of a digital image. In the course of computational experiments established that the modification of only blue color matrix of the image, the quantity of Blue-triads (which indicate the blue color component of the image) in the matrix of unique colors of the digital image significantly exceeds the quantity of Red- and Green-triads that will allow for the examination of images clearly establish the fact of modifying the blue color component. The identified feature is typical for most containers in a lossy format which is associated with the digital image's formation technologies and compression algorithms therefore despite the low sensitivity of the human eye to blue color it is not recommended to use the blue color component for spatial transformations of the image, in particular, for steganographic transformations. During the experiments the most suitable for spatial transformations green and red color components revealed, since their modifications lead to the appearance of a smaller quantity of Green- and Red-triads (which indicate the green and red color components of the image, respectively) and to obtain a comparable quantity of all types of triads of triplets in a matrix of unique colors which will complicate the correct examination of a digital image from the point of view of unambiguously identifying a modified color component. The obtained results are substantiated from the standpoint of color perception by the human eye, which underlies the formation of colors by digital cameras, and the image compression features of the JPEG standard. Criteria for determining the color component of a digital image providing less disturbance in the quantity of sequential triads in a matrix of unique colors and the basic steps for determining for a particular image the most and least suitable for steganographic transformation color components in terms of changes in color triplets in a matrix of unique colors are given. [ABSTRACT FROM AUTHOR]

The process of optimization work schedule is important for meeting the needs for staff and requirements of work time standards and also for accounting staff availability and preferences. The paper considers shifts scheduling problems for staff with a flexible work schedule. It was formulated three problems of determining the number of staff, who start to work in a given interval of the day. The problems deal with a given duration of the shift and lunchtime, which can be planned in the given bounds and the given staff number. The objective of the problems is to minimize the total deviation of staff count from needs for them in each time interval during the day. There have been proposed two methods for determining the staff number in each interval, which envisages the solve of the two optimization problems. The problem, in which it is needed to define the number of staff without the count of lunchtime is solving in the first stage of the first method and with the count of lunchtime in given intervals in the first stage of the second method. After the solution of the first stage problem, in both methods, it is needed to assign lunchtime in allowed bounds in the solution got in the previous problem to minimize the total deviation of staff numbers from needs for them. It was got nonlinear problems, which were transformed into the integer linear and binary integer programming problems. There have been developed heuristic algorithms for solving the problem with unfixed lunchtime. There were two series of experiments with the purpose of research their efficiency. The input dataset for experiments was generated randomly in given bounds of interval time. The number of intervals and bounds of flexible lunchtime are variated parameters. The results of experiments prove the efficiency of the first method and the needs of additional research on the second one. [ABSTRACT FROM AUTHOR]

In this paper we investigate methodic of testing Turbo-Product-codes decoder on software and hardware level. The proposed methodic allows achieving results that characterize decoder's correction ability and through put. The objective of this in vestigation is development of unified methodic that makes it possible to hold testing and evaluation of decoder's properties for any sub-codes of Turbo-Product-code sand for any hard ware platform. The research has allowed emphasizing problems during decoding stage for the specified algorithm usageandidentifying condition for stable work of the decoder. From the practical point of view, it makes possible to define the code that is appropriate for the certain circumstances according to the necessary values of throughput and correction ability. Thus, the decoder can be dynamically configured in obedience to the outer conditions and reach maximum efficient ratio of the two mentioned parameters. The testing methodic supposes division on several stages of testing. In this work several software means have been used during the test stages and among them there is software modeling environment ModelSim, programming language Python. The developed hardware is intended for testing wide range of codes that are supported by the decoder. Hardware tools have also been occupied extensively. Hardware tools include developed laboratory stand using two development boards with FPGA integrated circuit and interface adapter for data transfer between computer and stand. The first one is used for direct implementation of decoder in hardware, while the second one provides necessary throughput capability for transmission speed testing. Combination of developed software and laboratory stand has been utilized during testing in accordance with methodic presented in this work. In compliance with the developed methodic, it is possible to develop universal approach for testing of hardware solutions for error-correcting codes. [ABSTRACT FROM AUTHOR]

Modern information protection issues are changing with the development of information technology. Every year, the level of cybercrime is getting higher everywhere. Often cybercriminals are exposed to digital files of various formats. One of the main tasks of modern information security is the detection of violations of the integrity of digital signals, in particular, digital imaging. Detection of digital image processing is currently a topical task due to the increased incidence of integrity of important documents. Digital images are almost always processed by the filters of graphic editors. One of the popular filters is the sharpening. In the open edition, a single algorithm for detecting artificial sharpening was found. The existing algorithm is based on the analysis of close pairs of colors of a digital image in relation to the total number of pairs of colors of the matrix of a digital image. Threshold value is detected, considering what certain ratio of these indicators indicates presence or absence of sharpening. However, his main disadvantage is a large number of errors of the first and second kind. In this paper, the influence of the partition of the image matrix on square blocks of different sizes on the effectiveness of detecting sharpening is investigated. As a sharpening filter, Adobe Photoshop's Intelligent Sharpness tool is chosen because of its wide-ranging popularity. On the basis of the conducted research modification of the algorithm was performed. This modification made it possible to reduce the number of errors of the 1st kind by 12%, and the 2 genera by 4%, which results in 40% of errors of the second kind and 14% of errors of the 1st genus. Also, possible further variants of development and improvement of the process of detection of the indicated processing, namely, the study of the influence of the modification of the formula of close pairs of colors on the efficiency of the algorithm, are proposed in the work. [ABSTRACT FROM AUTHOR]

The well-known combinatorial problem of finding a coverage by the theorems method on the properties of the cover table is considered. The author's work "Methods and algorithms of coverage (part 2)" [1] provides a consistent solution to this problem, which has a cyclic character. In the new author's work the method of parallelization of the solution of this problem based on the property of independence of branches of computational process (subprocesses) is offered; external cycles of subprocesses possess such property search nuclear/antinuclear rows and columns of absorbing/absorption lines. A series-parallel information graph of such a solution is constructed and its description is given. Of particular importance is the definition of such parallelization of the computational process, in which the computational load on the processors is uniform, that is, balanced. In practice, in many cases of cycles, there is a gradual decrease in the volume of calculations in the body of the cycle from maximum to single, resulting in a load on the processors becomes significantly uneven. The considered problem of finding a covering is just such a case. The paper proposes a geometric representation of the computational load. The case of nonuniform load described above is represented by the triangle of the computational load. The method of converting the load triangle into an equal-sized rectangle is shown, which provides effective load balancing for processors in a parallel system. Assessment of underloaded processors is proposed. [ABSTRACT FROM AUTHOR]

The paper investigates the main approaches for detecting fraudulent operations with credit cards and proposes a combined implementation of behavioral assessment of credit card holders and credit card transactions in order to detect possible fraud attempts. An incoming sample of 4,000 active credit cards was generated, for which 55 thousand transactions were carried out, the main types of credit cards were distinguished by their behavior. According to the results of a study on the behavior of credit card clients, a behavioral model was built. It describes how they will act at the next moments, what transactions carry out, of what amount and with what limits, and thereby non-typical queries could be identified that are evidence of fraudulent attempts. Along with the standard characteristics describing the client and credit card, aggregated characteristics were formed for client's behavior describing (balance, income, expenses, limits) in dynamics and due to this managed to obtain more qualitative models. In analyzing the risk of fraudulent transactions, transaction transactions for previous periods (typical and fraudulent) were used, models based on logistic regression, regression equation, regression with integrated slip medium, and also a multi-layer perceptron and radial-basic function were used. Such models carried out the classification of the transaction for fraud, and then predicted the amount of the transaction. The best method of classification of transactions was a multilayer perceptron, and the highest accuracy values of the forecast showed autoregression integrated moving average (ARIMA) model with an error value of 4.46% for forecasting transaction amount. The proposed combined approach proved to be quite effective in analyzing and monitoring of credit cards and detecting fraudulent transactions, and it could be implemented on other types of financial risks (for billing and payment systems, money transfer systems and other trading operations). [ABSTRACT FROM AUTHOR]

The theoretical principles of the formation of the criterion of structural complexity of binary, halftone and color optical images in the Hamming space are highlighted, analytical expressions of the Hamming distance estimates are given on the basis of different crosscorrelation functions, and the examples of applying the method for recognition of optical images in the Hamming space while solving applied problems of structuring multifunctional data in distributed computer systems are presented in this paper. The research results can be used while developing high capacity processors for data encryption. These devices are based on a functional scheme of the device for calculating the modular correlation function. This leads to increased efficiency of the device for estimating the Hamming distance between two signals. [ABSTRACT FROM AUTHOR]

The paper considers the process of development from information security to cybernetic security of infrastructure objects, in particular, the processes of cyber security audit. It has been established that the aspects of the dynamics of information and cyber security processes have not been adequately studied. and do not disclose the full formal and systematic approach to solving cyber security problems. In this regard, developed a logicallinguistic model of a risk-oriented approach to planning cyber security activities and the simplest mathematical model of two-stage cyclical management of audit processes. The use of the mathematical model will contribute to a more accurate understanding of the dynamic properties and the development of the necessary system and non-systemic measures to ensure cybersecurity. [ABSTRACT FROM AUTHOR]

In the paper we study the effect of modification of DCT coefficients on the spatial domain of digital images using analysis of sequential triads of color triplets in a matrix of unique colors of the image, for further steganalysis of stego formed by modifying the DCT domain. On the basis of the computational experiments the improvement of the previously developed steganalytic method for detecting the presence of additional information attachments in digital images was carried out. A software interface for steganalysis of digital images was developed in MATLAB environment. [ABSTRACT FROM AUTHOR]

In this paper we consider the basic fundamental factors of the content of the mathematical model, which we seek to construct, namely, predicting the reliability of the process of functioning of complex systems and the risk of their failure. Moreover, in the most basic form of these indicators. This is the probability of a fail-safe process of functioning of the system (reliability and survivability) in certain spatial and technological conditions over a period of time and under the same conditions - the probability of failure (risk) of systems. [ABSTRACT FROM AUTHOR]

In this paper a new approach of digital watermarking for QR-code protection is developed. This approach can be used for authentication and integrity checking of QR-code after its reading with mobile device from paper. On the basis of proposed approach a software application for mobile device on Windows Phone is implemented. [ABSTRACT FROM AUTHOR]

The paper presents a comparative efficiency analysis of algorithms for detecting the presence of attachments of confidential information in digital video, that implement the author's previously developed steganalytic methods, with modern analogs. Proposed algorithms are based on the analysis of the spatial domain of digital contents, that allowed to avoid additional accumulation of computational error and, as a result, to increase the efficiency of steganalysis of digital video in conditions of low hidden capacity. [ABSTRACT FROM AUTHOR]

Prevalence of breast cancer ranks first in the structure of women cancer. Expert diagnosis of pathological states based on qualitative performance analysis of histological images, so development processing allows making the correct diagnosis process faster and more accurate for the expert. This paper proposes a fuzzy system diagnosing of precancerous and cancerous breast conditions based on attributes obtained from the doctor-diagnostician. The system can be used in cancer medicine that will allow diagnosing in real time and eliminating subjective diagnosis by a doctor. [ABSTRACT FROM AUTHOR]

This paper examines the possibilities for simulation of complex radio systems. Modelling and simulation is developed in Simulink environment and validated through experimental data. In this paper we offer example to show how simple simulation model of radar transmitter. [ABSTRACT FROM AUTHOR]

Methods of complex evaluation of the state and operation quality for complex hierarchical network systems are described in the paper. These methods are using a large number of characteristics and parameters of evaluations for obtaining the adequate conclusions. Such procedures require performance of huge amounts of computations. Therefore, approaches to the optimization of the proposed methods of evaluation based on the efficient organization of parallel computations are developed and investigated. Algorithmic constructions for the implementation of the methods of complex evaluation on the modern and perspective computer systems are offered. Features of this implementation on multicore computers, cluster systems and high-performance computing environments are determined. Some theoretical estimates of the acceleration of parallel computing are presented. Results of this work can be used for complex evaluation of real hierarchical network systems from different subject areas (transport systems, communication and power supply systems etc.) based on the use of modern software and hardware. [ABSTRACT FROM AUTHOR]

The paper presents a modification resistant to attacks against embedded message steganographic method developed earlier by the author, which provides a spatial region stehanoperetvorennya container image, but is limited to applications with the ability to breach the reliability of the image due to embed additional information. The result of the modification is to eliminate any restriction on the scope of the relevant algorithm by ensuring the reliability of perception formed stehanopovidomlennya regardless of the specific image container. Resistance algorithm that implements a modified method to attacks against embedded message in this case is reduced slightly. As discussed zburni steps: overlay stehanopovidomlennya different noises with different parameters, filtering, compression lossy stehanopovidomlennya with different coefficients of quality (saving in format Jpeg, Jpeg2000). The results of computational experiments. [ABSTRACT FROM AUTHOR]

The paper presents the steganalytic algorithm, based on the accounting of differences of character in the change of the number of blocks with the equal color brightness values of color matrices of sequence of images / frames of a video sequence as a result of primary and repeated insertion of additional information. The proposed steganalytic algorithm analyzes digital content in the spatial domain and is aimed to detect the presence of the embedding of additional information by modification of the least significant bits with a small hidden capacity (not more than 0.5 bit / pixel). Results of computational experiment are shown. [ABSTRACT FROM AUTHOR]

Social Internet services belong to the class of nonlinear dynamical systems, so the phenomenon of social communication processes are characterized by unpredictability of agents' interaction due to external influences and transition of the system into a state of uncontrolled chaos. To achieve a steady state of the system as a result of self-managed agents, it is suggested to use the concept of the synergistic management of agents' interaction in social networking services. The paper synthesized synergistic management of agents' interaction in social networking service that will enhance the resistance of virtual communities to destructive information impacts through regularization of the demand for content. As an example, the article presents a model of the management of agents' interaction in order to provide the synergistically managed transition to a given state of the information security of a virtual community. [ABSTRACT FROM AUTHOR]

The paper presents the modification of simulation for Gaussian random variables using a normalizing Johnson SU transform. This approach, unlike many others, uses a single random flat value to generate a value of a Gaussian random variable, and improves the simulation results at the boundaries of empiric distribution of the Gaussian random variable simulated. [ABSTRACT FROM AUTHOR]

The paper presents an approach to decision-making support in the development of comprehensive special-purpose national information security programs in the presence of dangers and risks. The approach relies on the introduction of danger-and-risk models into the hierarchy of program/task targets and on the target-based assessment of the support. The risk model has a random risk factor and a special target. The danger is simulated by a special program which is introduced into the target hierarchy. [ABSTRACT FROM AUTHOR]

In this paper, by analyzing the most popular platforms for develop online - shops, was selected the most common. The analyze of the most common platforms, conducted by considering the following factors, such as the popularity of the platform, the size of the community of users and developers, ease of administration, technical support. By analyzing the shortcomings were found existing tools to protect information. Develop more complete plugin for information protection for the online shops based on the platform WordPress (WP). [ABSTRACT FROM AUTHOR]

The paper discusses the problem of classification and codification of professional skills in the context of the conflicts found in the practice of the development of higher education standards for technical and economical branches using a competency-based approach. We found the conflicts caused by the following: (a) no methodology exists to form a uniform list of professional skills from the skills of solving typical professional tasks and those ensuring the formation of competency; (b) neither the feedback model nor the model of the competency supplier to the labor market exists. Scaling, data analysis and correlation cluster analysis methods were used. An education activity model was proposed. We proposed the principles to develop a consolidated classification of professional skills which would be suitable in technical, economical and natural science branches of higher education. A consolidated classification is rather simple, and takes account of the current science and technology trends to convergence and integrated approach, thus making it possible to improve education standards. [ABSTRACT FROM AUTHOR]

In this paper, blood glucose dynamics during digestion is discussed. A mathematical model of glucose uptake distribution was developed based on the simplest parabolic law. Additionally, a mathematical model was developed based on the exponential function to characterize the accelerated process of glucose digestion. It was found that the model based on the exponential function is more accurate than that based on the simplest parabolic law. [ABSTRACT FROM AUTHOR]

In this paper, the authors propose the techniques incorporated into a strategy for justification of basic general system requirements for designing data security systems based on group argument accounting methods, analogy approach, expert assessment methods, and statistical data processing methods. [ABSTRACT FROM AUTHOR]

In this paper, a method to identify a multisensor output signal by means of modular neural networks was discussed. A multisensor output signal was defined by mathematical modeling. Mathematical models were considered for the instances when physical quantities transformation characteristics have positive derivatives, negative derivatives, and both positive and negative derivatives for the increase and decrease in the amplitude of multisensor output signal. The improvement in accuracy of identification was confirmed by the results of simulation modeling. [ABSTRACT FROM AUTHOR]

This paper focuses on problems of evaluation of the networks protection level taking into account the emergence coefficient. Criterion that allowed to set optimally correspondence between network emergency and protection level was first developed. According to the proposed approach, it is possible to identify the necessary protection level of information streams in computer networks that required for specific class of architectures. [ABSTRACT FROM AUTHOR]

The paper presents the model of the organization of the linguistic database as part of linguistic support of the automatic processing of natural language text information. The peculiarity of the proposed model is to separate the knowledge of the language system of knowledge about the world (domain). The developed model of a linguistic database in comparison with known characterized by significantly lower volumes of dictionaries, data format allows you to simulate patterns as inflected languages and analytical. It is able to provide the processing of natural language text for a variety of applications. [ABSTRACT FROM AUTHOR]

This paper focuses on mathematical models of the relationship between sets and subsets based on analysis of the logic categories for further analyze threatening impacts on information security systems. The description of the pullback, nuclear relations and amalgams in the sets of information security systems are given. [ABSTRACT FROM AUTHOR]