The purpose of the research is to substantiate and select the architecture of a neural network for the possibility of implementing the cognitive functions of network software for controlling a grouping of interacting small spacecraft.Methods are based on the concepts of AI theory for managing the grouping of small spacecraft - the use of adaptive methods and tools that allow making decisions, similar to the mechanisms of human thinking. With regard to space communication systems with a heterogeneous structure, AI methods and technologies are aimed at the processes of predicting the state in communication channels between network nodes and automatic reconfiguration of the network of devices based on the learning processes of a neural network (NN).Results. In the learning and forecasting mode, it is necessary to use time series of parameters and coordinates of specific pairs of small spacecraft with non-zero line of sight. Especially for time series analysis, recurrent neural networks (RNN) are used, in particular, LSTM. The idea of RNN operation is to use as input data for the current forecast not only the state vectors of the SVs and their coordinates, but also the previous value of the communication quality, actual or predictive. The paper shows that the onboard computing power of a separate MSC does not allow performing forecasting and training on board. Therefore, a dedicated ground segment of forecasting and monitoring is required, which will collect a posteriori information, periodically train the cognitive model, use it to predict the quality of communication, and transmit the results to the network nodes to build data transmission routes.Conclusion. The analysis of modern solutions and the choice of neural network architecture for the implementation of the cognitive functions of the network software for controlling the grouping of interacting small spacecraft showed that the neural networks of the Transformer architecture, which are based on the mechanism of internal attention, most fully meet the requirements of the project. The Transformer architecture allows using the entirety of a priori data, has a high learning and forecasting speed.

The purpose of the research is to develop a simulation model and tune the hyperparameters of a neural network to predict possible states of a network of small spacecraft.The research methods are based on the concepts of the theory of artificial intelligence to control a group of small spacecraft (SSV) – the use of adaptive methods and tools to make decisions, similar to the mechanisms of human thinking. In relation to space communication systems with a heterogeneous structure, artificial intelligence methods and technologies are aimed at predicting the state of communication channels between network nodes and automatically reconfiguring a network of devices based on neural network learning processes. One of the most important functions of network software for the application of cognitive algorithms is to predict the quality of communication between pairs of SSVs.Results. A method has been developed for using the Transformer architecture neural network to predict possible states of the SSV network, which provides aggregation and time synchronization of data on the state of the SSV network, their use for training the neural network, as well as using the neural network to predict the quality of communication. The data format for the training sample has been created, based on the representation of the state of the SSV network, which ensures the generation of the initial state of the network, modeling the proactive mode of its operation, collecting SSV network state markers to generate training data sets in the form of chronological sequences grouped into frames, and allowing to reduce the amount of data transferred between SSVs when creating a training set. A simulation model of the SSV network has been developed, which provides generation of the initial state of the network, modeling of the proactive mode of its operation, and collection of information about the state of the SSV network to generate sets of synthetic training data.Conclusion. The article develops a simulation model of the SSV network for generating synthetic training data and predicting possible states of the SSV network, as well as a method for using a neural network to predict possible states of the SSV network.

One of the important achievements of the theory of error-correcting coding is the discoveryof graph codes and their important subset - low-density parity check codes (LDPC codes). Usingthe parity check matrix of the code on the graph, one can obtain a Markov random field. LDPCcode can be embedded in an Ising model (a type of Markov random field) by using a torus topologywith negative curvature. In this case, codewords correspond to saddle points (extrema) in themodel, and trappin sets correspond to local minima. The use of LDPC codes with an increasedcode distance allows for maximum separation of saddle points, and thus increases the noise resistanceof the neural network and the representation power. At the same time, the block andsparse structure, characteristic of a torus of negative curvature, simplifies multiplexing and reducesthe number of trainable parameters of the neural network. The aim of the research is toreduce the computational complexity and increase the accuracy of neural networks through theuse of a priori structural (quasi-cyclic) sparse graphs for a wide class of machine learning problemson Markov random fields. The paper presents a new approach that allows the synthesis ofneural network architectures based on graph codes. The proposed approach provides an effectiverepresentation of Markov random fields through the use of QC-LDPC matrices and tensors.The proposed approach allows us to reduce the number of trainable parameters and logarithmicallyreduce the complexity of tensor multiplexing. The proposed approach provided an accuracyof 94.95% (1.72% to first place) of the binary classification problem “Pathfinder” of the “LongRange Arena” competition, with more than 5 times fewer parameters (multiplications). Applicationof the proposed approach to factorization problems on dense graphs, network problems, surfacemeshes, covariance matrices made it possible to increase the accuracy of reconstruction usingthe Frobenius metric in individual problems by more than 8 orders of magnitude in combinationwith simplifying the structure of the multiplexer.

The work consists in describing the types of modulation and their parameters for the selection and construction of modern digital communication networks with multiple subscriber access to network resources and high-speed interconnection. The goal is to systematize communication systems and provide multiple access of network subscribers to resources and high-speed data exchange. Solution method. The article discusses the technologies and standards of modern communication systems in terms of the structure of the construction of receiving and transmitting devices from the conversion of an analog signal into a digital form to the flow of symbolic solutions from the output of the demodulator. The classification of types of wired/wireless communication, communication standards, types of modulation on an analog carrier is carried out. The features of the hardware implementation of receiving and transmitting devices, transmission speeds and operating conditions of communication channels were taken into account. The largest manufacturers of the element base and specialized digital signal processing chips are presented, on which the interfaces of the physical layer of most receiving and transmitting devices of communication systems are implemented. As a result of the analysis, a set of modulation types that are currently used in most communication channels was determined, as well as the main parameters of their operation.

The article describes approaches and methods for managing a group of moving objects,characterized by the ability to autonomously make decisions about their status within the group.Another problem of managing such a grouping is weak predictive solutions for the connectivity ofpairs of elements and their dependence on a single control center. Nanosatellites operating underconditions of uncertainty in the internal and external environment are considered as such objects.The goal is to ensure the coherence of the group’s apparatus through a decentralized change instructure. It is shown that methods and algorithms for dynamic reconfiguration of a group of movingobjects predominantly use a centralized approach and a single ground control center, which isimpractical for small space exploration. A class of management methods using knowledge processingmethods and technology (artificial intelligence technology) is considered, allowing for theidentification and use of additional information about the configuration of the group. Configurationis understood as a dual system that describes the composition and connections between neighboring elements with some quantitative assessment. The article checks the connectivity configurationof elements to ensure continuous data transfer between a pair of arbitrary groupingelements. The proposed reconfiguration method is hierarchical: at the upper level, reconfigurationis based on the principles of self-organization; at the lower level, the grouping is understood as anadaptive system that changes its state based on a trained neural network based on historical data -time series of parameters of devices and their locations. The method is a two-level cycle of pollingeach element for grouping its neighbors and drawing up a network map. This network map showsthe available connections, taking into account the current steam numbers of each device. The second(nested) polling cycle uses control information about the future state of the device and theconnectivity of the group as a whole. Making changes to the network map instances by each deviceand updating the network map instances allows, upon completion of the polling cycles, to obtainthe configuration of working devices. The results of the comparative analysis showed that managementmethods based on the principles of self-organization and adaptive change in structure arethe most suitable for dynamic reconfiguration of the group. This result is possible due to the supportof forecasting steps.

The purpose of the study is determined by a significant expansion of the classes of threats tomodern automated systems, the dynamic development of tactics and techniques for attacking theirinformation resources. The available methods and hardware and software tools effectively resistsingle-stage attacks that have a fixed scheme of destructive impact and time-limited activity. Moderntypes of destructive influences are understood as multi-stage complex attacks, for which it isimportant to create an adequate and effective apparatus for describing, modeling and repellingnew types of attacks. Research methods are based on the development of a structural-algebraicapproach, primarily on the apparatus of formal grammars and metagrammars. It has been establishedthat the well-known formal models for describing and modeling multi-stage complex attacksare cumbersome, and their modification is difficult. Most attack descriptors are not equipped witha representative set of methods for structural and algebraic analysis of such complexly structuredobjects. To describe, model and repel such attacks, a class of reconfiguration metagrammars hasbeen developed. These metagrammars contain a set of regular and reconfiguration rules formatching between grammar elements within the grammar. These rules allow you to select specificbranches of the search graph depending on the achieved parsing states. This property significantlyreduces the search space and thus increases the specific efficiency of the search. The developedapparatus of reconfiguration metagrammars creates the necessary theoretical basis for their effectiveuse in modeling and reflecting existing and prospective ICAs that have a structural-linguisticdescription. The resulting qualimetric five-dimensional diagram, built on a set of practically significantindicators (homogeneity, connectivity, compactness, adaptability, directionality) showedthe advantage of reconfiguration metagrammars over general metagrammars. Methods of parsingin reconfiguration metagrammars differ in structural rules of reconfiguration (structural adaptation)and selection criteria for their adaptation. These procedural features make it possible toexpand the possibilities of attack modeling and improve the efficiency of procedures for repellingmulti-stage complex attacks.

The purpose of the study is to develop a method for encoding of transmitted ADS-B messagesbetween aircraft. The open format 1090ES of transmitted data is critical in terms of carryingout various types of attacks that can lead to a violation of the safety of aircraft operations.The work is aimed at using means of encoding and decoding messages with a private key. Researchmethods are based on the application and development of streaming data encryption usingone-dimensional cellular automata. They operate as a generator of pseudo-random sequences thattransform the elementary states of a cell of a one-dimensional cellular automaton based on typicalhardware-oriented operations. The processes of encoding and decoding data fields are based onan analytical expression using typical logical operations (or, xor). This property allows parallelprocessing of message data fields. The result is the created method for ensuring the protection oftransmitted data, additionally encoding on transmission and decoding on message reception.A distinctive feature of the method is the preservation of the protocol forma. The method uses aone-dimensional cellular automaton that encodes and decodes the target fields (coordinates, heading,etc.) using a pseudo-random number generator. The developed method belongs to the class ofhardware-oriented methods. Critical for encoding and decoding properties of periodicity of datafields and key length are eliminated by choosing an initial irrational value and organizing the“streaming” work of the encoder. If the encoding automaton is running in streaming mode, thecurrent value depends on the history of some depth, determining the length of the "automatic key"from the ADS-B message will be algorithmically impossible due to data loss. The linear complexityof the method allows you to perform transformations at the data rate. Conclusion: the developmentof hardware-oriented methods of data encoding makes it possible to increase the efficiency ofusing the ADS-B system by counteracting various types of destructive actions.

The organization of information exchange with a small spacecraft is performed using signal-code structures with high power and spectral efficiency. These criteria are fully satisfied by two-position and four-position phase-shift signals that effectively use a dedicated frequency band and do not require high signal-to-noise ratios for reliable reception. A significant problem that arises when transmitting radio signals via satellite communication channels is the presence of an ionospheric propagation region which due to its inhomogeneous electronic configuration, distorts the transmitted signal. In this paper, an analytical study of the effect of changes of the dielectric permittivity index of the ionosphere on the phase distortion of signals in the "small spacecraft – Earth station" path is carried out. A mathematical model for determining the integrated electron concentration in the ionosphere for different heights and inclinations of signal propagation paths in satellite communication channels is provided. The paper presents a mathematical model for estimating changes in the phase incursion of a phase-shift signal passing through dense layers of the ionosphere depending on various values of the signal frequency, the integrated electron concentration in the signal transmission path, the type of electromagnetic wave propagated, the magnitude of the components of the magnetic induction vector, as well as on the coordinates of the observation point. A tool for modeling radio waves based on a bicharacteristic system of equations is presented in an analytical form, which allows taking into account both the isotropy and anisotropy of the ionosphere.

Purpose of research. Reducing the time and increasing the efficiency of search and rescue of crews of land vehicles, air, surface and underwater craft. Combining physical, informational, and geometric parameters allows reliable detection of single ferromagnetic bodies by analyzing additional information in working algorithms.Methods. The paper contains a practically significant variant of building a search engine that is implemented on two or more unmanned flying vehicles moving in the search area in a coordinated formation at the basic distance. The solution system is based on using magnetic field components as magnetometric information without using primary gradient information. This feature allows us to get compact and functionally complete algorithms. Calculations using such algorithms ensure the stability of an unmanned flying vehicle in various combinations of vector, roll and pitch.Results. Algorithms for the functioning of the magnetometric system for searching ferromagnetic bodies based on the magnetostatic equations have been constructed in the article,. The intelligent system is based on the scheme of a measuring and computing circuit with staggered vector three-axis blocks of magnetometers in space. The system is able to solve the whole range of problems of search, detection, direction finding, localization, positioning and identification of metal objects that have their own magnetic field.Conclusion. The main result of the study is that the working algorithms for detecting ferromagnetic bodies take into account the characteristics of the air environment, the instrumental error of measurement units of physical quantities and the geometric proportions of measuring equipment location on board the unmanned and manned search and rescue helicopters. The obtained research results have been used in the development of a hardware and software complex with helicopter-type unmanned flying vehicles designed to support search and rescue operations in the Arctic.

Oбъект исследования - алгоритмы параллельной сортировки с применением базовой операции «сравнение-обмен». Цель исследования - сокращение шагов работы алгоритма сортировки массива данных за счет уменьшения количества промежуточных перестановок элементов массива. Цель достигается разработкой оригинальной схемы коммутаций элементов массива. Данная схема составляет основу модели диагональных коммутаций пар элементов массива. Массив имеет 2d-представление, что позволяет объединить в пары элементы из его различных половин. За счет 2d-представления массива образуемые пары элементов позволяют уменьшить количество перестановок. Новизна модели диагональных коммутаций состоит в том, что операции «сравнение-обмен» параллельно выполняются на неконфликтующих парах элементов, взятых из различных половин массива. Это свойство модели позволяет «прыжками» перемещать элемент в необходимую позицию массива. Модель диагональных коммутаций объединена с известной моделью четно-нечетной сортировки. В результате объединения был получен алгоритм параллельной сортировки с гибридной схемой коммутации. Эта схема реализует на четных шагах предложенную модель, а на нечетных шагах - модель четно-нечетной сортировки. Моделирование алгоритмов четно-нечетной сортировки и гибридной сортировок показало преимущество разработанной модели. Расширение четно-нечетной сортировки моделью диагональных коммутаций позволяет сократить среднее число шагов сортировки. Кроме четно-нечетной сортировки, модель диагональных коммутаций применима для алгоритмов параллельной сортировки, использующих базовую операцию «сравнение-обмен» - сортировки Батчера, Шелла, слиянием The object of the research is parallel sorting algorithms using the basic operation "compare-swap". The aim of the research is to reduce the steps of the algorithm for sorting the data array by reducing the number of intermediate permutations of the array elements. The goal is achieved by developing an original switching circuit of the array elements. This circuit forms the basis of the model for diagonal commutation of pairs of array elements. The array is 2D, which allows one to pair elements from its different halves. Due to the 2D representation of the array, the formed pairs of elements allow reducing the number of permutations. The novelty of the diagonal commutation model is that the “compare-exchange” operations are performed in parallel on non-conflicting pairs of elements taken from different halves of the array. This property of the model allows one to "jump" the element to the desired position in the array. The diagonal commutation model is combined with the well-known odd-even sorting model. The combination resulted in a parallel sorting algorithm with a hybrid switching scheme. This scheme implements the proposed model at even steps, and the even-odd sorting model at odd steps. Modeling algorithms for odd-even sorting and hybrid sorting showed the advantage of the developed model. Extension of even-odd sorting by the diagonal commutation model allows to reduce the average number of sorting steps by 6-10%. In addition to odd-even sorting, the diagonal commutation model is applicable for parallel sorting algorithms using the basic comparison-exchange operation - Butcher, Shell, merge sorting

Article contains the description of parallel search in associative memory. Initial operands (words) receive a dual impression. The source text is presented in the form of a line or in the form of a matrix. The sample is presented by a two-dimensional form. It consists in splitting a sample into even and odd character sets. Addition of the second coordinate for a sample allows to execute the alternating development of symbols and provides the anticipatory analysis of symbols of a sample of rather next symbols. This feature of the form of a sample allows to process each character set parallely. Processing of character sets is carried out on bit cutoffs in the associative memory at the same time. For the hardware support of steps of retrieval operation the original structure of the associative memory is offered. The structure contains new communications between storage cells, two retrieval registers for storage of character sets and original a circuitry element for foliation of characters on sets.

In article approach to search in a sample for production systems is described. Search in a sample is applied to processing of symbolical information. For this purpose operands are presented in the form of a matrix in which processing of vectors is conducted by a conveyor way. Elements of two vectors are logically multiplied among themselves with shift by the beginning of bit of the previous vector of rather current vector. Such operation has received the name "shift" conjunction. She allows to find a sample as a part of the search text.

Hierarchy of high-rate multiplexing systems with time compression of channels, principles of construction and basic characteristics of first-level hierarchy systems according to the recommendations by ITU-T G. 732-G. 734 have been discussed.