Software & Systems

Articles of journal № 1 at 2021 year.

Order result by: Public date | Title | Authors

1. iLabit OmViSys: A photorealistic simulator based on the omnidirectional camera and structured light [№1 за 2021 год]
Author: Kholodilin, I.Yu.
Visitors: 2921
According to recent advances in neural network learning, which are supported by the demand for large training data, virtual learning has recently attracted a lot of attention from the computer vision commu-nity. Today, there are many available virtual simulation environments, but most of them are based on a standard camera and are limited to measure sensors that are on the mobile robot. To facilitate data collection in systems that were not previously integrated into existing virtual envi-ronments, this paper presents a photorealistic simulator "iLabit OmViSys", which includes an Omnidi-rectional camera, and a structured light source. An Omnidirectional camera and structured light have their own distinctive advantages compared to other computer vision systems. The Omnidirectional camera provides a wide viewing angle with a single shot. In addition, the laser light source is easy to detect and extract its information from this image for further processing. Developed using Unity, the iLabit OmViSys simulator also integrates mobile robots, elements of the internal environment, allows you to generate synthetic photorealistic datasets, and supports communi-cation with third-party programs based on the Transmission Control Protocol (TCP). iLabit OmViSys includes three primary screens that allow one to generate data for internal camera calibration, carried out experiments, and take measurements. A distinctive feature of the simulator is also its versatility, in terms of support for such operating systems as Windows, macOS, and Linux.

2. Automation of day-to-day tasks as a modernization of a marine rescue operation automation suite [№1 за 2021 год]
Authors: Karpov A.V., А.А. Sakharov
Visitors: 3525
The Armed Forces of the Russian Federation have phased in a marine rescue operation automation suite (MROAS) in August 2014. The suite is designed to automate the activities of specialists of the Navy search and rescue service during their daily activities, as well as when making decisions on search and rescue operations for emergency situations on Navy ships and vessels. The paper presents modern approaches to automating day-to-day tasks solved by specialists of the Navy search and rescue service: a hybrid method for developing special software, methods of forming functional requirements for a modernized MROAS, a generalized list of information that the suite ac-cumulates and keeps up to date, the basic principles of organizing information interaction between the suites distributed among the Navy command and control bodies at different levels.

3. The adaptation of the LSTM neural network model to solve the pattern recognition complex problem [№1 за 2021 год]
Author: V.S. Tormozov
Visitors: 3576
The paper examines the adaptation of the model of artificial neural networks of direct distribution with blocks of long short-term memory (LSTM) for the complex problem of pattern recognition. For artifi-cial neural networks (ANN), the context can be extracted from the input signal vector and from the weight values of the trained network. However, considering the context of a significant volume, the number of neural connections and the complexity of training procedures and network operation in-crease. Instead of receiving context from input values, the context can also be temporarily stored in a special memory buffer, from where it can later be extracted and used as a signal in the ANN's opera-tion. This type of memory is called LSTM. The advantage of networks of this type is that they use memory blocks associated with each neuron of the latent layer, which allows context-related data to be stored when forming recognition patterns. There is the method of linear switching of LSTM units depending on the value of the transmitted signal in the paper. A computational experiment was conducted aimed at investigating the effectiveness of the proposed method and the previously developed neural network of direct distribution of a similar structure. Machine learning was performed for each type of ANN on the same sequence of training ex-amples. The test results were compared for: an ANN of direct propagation, a recurring neural network (RNS) of a similar architecture: with the same number of neurons on each layer, and a network of neu-romodulating interaction with one feedback delay. The optimization criterion, in this case, is the error of the neural network on the training sample, consisting of examples not presented in the test. The effi-ciency of solving the classification problem is evaluated according to two criteria: learning error on the training sample and testing error on the testing sample.

4. Adaptive block-term tensor decomposition in visual question answering systems [№1 за 2021 год]
Authors: M.N. Favorskaya, V.V. Andreev
Visitors: 3161
The paper proposes a method for dimensionality reduction of the internal data representation in deep neural networks used to implement visual question answering systems. Methods of tensor decomposi-tion used to solve this problem in visual question answering systems are reviewed. The problem of these systems is to answer an arbitrary text question about the provided image or video sequence. A technical feature of these systems is the need to combine a visual signal (image or video sequence) with input data in text form. Differences in the features of the input data make it rea-sonable to use different architectures of deep neural networks: most often, a convolutional neural net-work for image processing and a recurrent neural network for text processing. When combining data, the number of model parameters explodes enough so that the problem of finding the most optimal methods for reducing the number of parameters is relevant, even when using modern equipment and considering the predicted growth of computational capabilities. Besides the technical limitations, it should also be noted that an increase in the number of parameters can reduce the model's ability to extract meaningful features from the training set, and increases the likelihood of fitting parameters to insignificant features in the data and "noise". The method of adaptive tensor decomposition proposed in the paper allows, based on training data, optimizing the number of parameters for the block tensor decomposition used for bilinear data fusion. The system was tested and the results were compared with some other visual question-answer systems, in which tensor decomposition methods are used for dimensionality reduction.

5. Algorithms for smart node patterns in a wireless sensor network [№1 за 2021 год]
Authors: Vinogradov G.P., Emtsev А.S., Fedotov I.S.
Visitors: 3760
One of the trends in the development of modern weapons is the "linking" of individual samples with a certain degree of autonomy into a complex using, as a rule, wireless sensor networks. The scope of such complexes is uncertain and poorly formalized environments. It is possible to achieve their desired efficiency mainly by improving the intellectual component of the management system of the complex as a whole and the individual node in particular. However, it should be noted that the vast majority of research in this area remains only at the theoretical level. There is a gap be-tween the primitive models of artificial entities’ behavior, for example, in swarm robotics, models of their interaction, and expectations from the practice. The situation is aggravated by the secrecy re-quirements, miniaturization, and low energy con- sumption. This paper presents an approach to the development of software for intelligent control systems for a separate network node that has a given degree of autonomy when performing problems. To offer rela-tively simple algorithms in the conditions of restrictions on power consumption and speed for giving the network node the properties of intelligent behavior, to provide the ability to study the situation and decide both independently, considering the data received from other network devices, and as part of a group. There are methods of the fuzzy sets theory, the theory of building fuzzy models and networks, and approaches and algorithms for building on-board intelligent control systems in this work. The work shows that the class identification of typical situations and successful action methods in actual conditions contribute to the development of the required algorithms. On this basis, it becomes possible to develop formal behavior models (patterns) for implementation in the node management system. The authors propose a two-level structure of an intelligent network management system. The upper level, implemented by the operator, corresponds to such properties as survival, safety, the fulfillment of mission obligations, accumulation, and adjustment of the knowledge base as effective behavior pat-terns. The object of control for it is the network, considered as a functional system. It calculates the current indicators of specific value based on the results and effectiveness at time t, calculates and im-plements the method of action (behavior) at time t according to a given behavior pattern, and monitors the results of implementing the behavior pattern.

6. Architecture and programming implementation of research testbed of a corporate Wireless Local Area Network [№1 за 2021 год]
Authors: L.I. Abrosimov , M.A. Orlova, H. Khayou
Visitors: 2449
The paper presents the architecture and implementation of a research testbed for obtaining and analyz-ing the probabilistic time characteristics of a corporate Wireless Local Area Network (WLAN). To de-velop this testbed, the authors obtained mathematical relations for calculating the guaranteed intensity of multimedia traffic. The research testbed architecture includes the two independent blocks. The block "Simulation testbed" contains the corporate WLAN description and the multimedia traffic flows in the discrete-event simulation system ns-3. The block "Analyzing simulation results" contains programs for analyz-ing files of transmitted traffic and simulation results and programs for calculating performance charac-teristics. To write the block "Analyzing simulation results ", the authors used the Python3 language, the analysis of the transmitted traffic files was performed using the pyshark library. The paper also contains the analytical equations of the WLAN model used in the block " Analyzing simulation results ". The above equations allow us to determine the maximum intensity of the delivered packets for a prescribed time of guaranteed packet delivery, for wireless communication channels us-ing a prescribed channel protocol. A software implementation of the research testbed affords the op-portunity to get the dependence of intensity guaranteed multimedia traffic for the specified parameters: structure WLAN, the settings of the wireless communication channel protocols, and channel access control. The developed testbed provides the capability of operation in two modes. In the development mode of a new WLAN, when the known parameters are the equipment passport data, logical characteristics of protocols, and expected traffic characteristics, a full set of functional modules and blocks is used, which allows both traffic matching with transmission and processing resources, and the specified per-formance of the WLAN. In operation mode, when monitoring allows you to get actual characteristics of traffic and protocols, the testbed allows the WLAN administrator to test the performance of the WLAN and the traffic intensity. This mode uses a limited module set, which requires much less time to evalu-ate the performance of the WLAN, provides the ability to adaptively change the settings of the WLAN, and provides the performance characteristics of the WLAN that meet the requirements of QoS.

7. The intelligent approach to automation of technological and production processes [№1 за 2021 год]
Authors: S.Yu. Ryabov, Yu.V. Ryabov
Visitors: 3194
The paper considers the approach to production automation, in particular, to the automated design of technological processes. Data processing in existing systems is reduced to a set of rules, and the exe-cuting program in its implementation is like a state machine. Obviously, this approach has its own ceil-ing. It is proposed to represent the production process as something whole, described by an intellectual model. The adopted model of automation of technological and production processes is based on graph theory and graph representation of data and knowledge. The graph is considered as some function of time and computation. It is proposed to use a supergraph as a set of abstract and defined given nodes and abstract and static relations. Thus, every script of the physical reality, every manufacturing situa-tion, considered at any scale, will be modeled as a subgraph of a supergraph. Akka, which is an imple-mentation of an actor computational model, can be an intelligent platform for the implementation of computations. It allows for an intelligent approach to solving the problem of automating production and technological processes. An example of constructing a part of a supergraph for machining a part element is considered by a typical transition-side, including the corresponding tool, processing modes, and a measuring tool. The result of such a system will be a graph with vertices and relations describing the knowledge of techno-logical operations or the state of the production process. The result can be transferred to another sys-tem for execution, saved in the database, or used to analyze the situation.

8. Using the entropy characteristics of network traffic to determine its abnormality [№1 за 2021 год]
Author: Efimov A.Yu.
Visitors: 3494
The number and scale of network computer attacks (intrusions) are constantly growing, which makes the problem of their prompt detection highly relevant. For this, network-level intrusion detection sys-tems are used, based on two approaches – abuse detection and anomaly detection, and the second ap-proach is more promising in the face of the constant appearance of new and modified types of intru-sions. The main objects of application of anomaly detection techniques are mass attacks (DoS- and DDoS attacks, scanning, spreading of worm viruses, etc.), which are difficult to detect by other (for ex-ample, signature-based) methods, since they are often based on regular network interactions. The entropy analysis method for detecting network traffic anomalies, compared to many other methods, is characterized by sufficient simplicity of implementation and speed of operation. The appli-cation of the method is based on the general assumption that abnormal traffic is more ordered or struc-tured than normal traffic in some parameters and more chaotic in others, which manifests itself as a de-crease or increase in the entropy of these parameters. This paper is devoted to determining the nature of the impact of attacks on the entropy of such traf-fic parameters as the source and destination IP addresses, as well as the destination port, considering several types of DoS- and DDoS attacks as objects. The author describes an approach to determining entropy (using Shannon entropy). The paper presents the results of the author's model, which reveal the ambiguity of the impact of attacks on entropy characteristics. The results show a clear dependence of such inpact (decrease or increase) depends on factors such as the source, target, attack power, and dis-tribution of normal traffic. Conclusions are made about the possibility of effective detection of anomalies corresponding to DoS and DDoS attacks by analyzing the entropy of network traffic parameters, but only if this analysis is carried out taking into account the distribution of normal traffic and the volumetric characteristics of normal and total traffic.

9. Mental models in designing the behavior of artificial objects [№1 за 2021 год]
Author: Vinogradov G.P.
Visitors: 3459
The authors relate the immediacy of the problem considered in this paper to the need to design artifi-cial systems to perform a certain mission like a person and when interacting with him. The analysis of approaches to the construction of artificial objects has shown that developers often give artificial objects their own patterns of behavior because of the specifics of the concepts and con-cepts they use. As a result, there are gaps between the formal models of patterns offered by developers and the expectations of users. The aim of the authors is to justify an approach to the development of software for intelligent sys-tems for managing the behavior of artificial entities based on the theory of patterns and to propose an approach that provides the development of digital products based on the patterns of behavior of the subject-leader for artificial entity management systems that automate the implementation of mission problems. The work uses the methods of the theories of reflexive games and information management of sys-tems with willpower and intelligence. There are ten constituent elements based on a formal model of the behavior pattern. The work shows that these elements form what can be called an intelligent digital machine designed to automate the execution of a mission in the interests of the host subject. These components are present when a person performs a mission, including all participants in the project to develop an artificial entity. The authors show that the ideas adjustment about the architecture of an ar-tificial entity through the exchange of information during the discussion allows us to determine the most effective model of the behavior pattern and all its components for implementation in an artificial entity. The authors proposed to conduct matching as a game by conducting experiments using the TOTHE method (a set of input data  impact  result), using spatial and visual logic to interpret the results. The possible logic of interpretation is justified. The authors briefly considered the models of the constituent elements of the behavior pattern for an artificial entity. An example of the approach im-plementation is considered. The authors show that in conditions of severe time scarcity; the choice based on behavior patterns makes it possible to implement effective behavior that does not require sig-nificant computational resources.

10. Method of testing the training models for the adequacy [№1 за 2021 год]
Authors: Ilin V.А., Kiryushow N.P.
Visitors: 3595
The paper substantiates the necessity of assessing the quality of simulated models of trainer systems and their adequacy to actual systems and describes a method for evaluating the adequacy of simulation models. The simulated model must provide the required accuracy and reliability of the process simulation. The simulation validity assumes that the model meets some specific requirements that allow us to test its quality. The model quality analysis involves testing for compliance with the modeling goals. In general, the evaluation of the model properties includes the model adequacy assessment, the simulation results’ ac-curacy (simulation error), the stability of the simulation results of the studied processes, and the study of the model sensitivity. The model adequacy assessment reflects its compliance degree with the actual system. The algo-rithm for the model adequacy checking comprises comparing the outputs (responses) of the model and the actual system with the same input values. Here, statistical methods are used to test hypotheses, for example, by the t-Student criterion. The path of determining the estimates of the mathematical expectation and variance of the deviation of the components of the response vector tests the accuracy of the simulation. The variance of the flow values tests the stability of the simulation results. The simulated model sensitivity refers to the degree to which the model's output parameters or re-sponses change depending on the input characteristics. Methods of assessing the adequacy of the models include the steps of selection criterion of the va-lidity of the simulation model to the subject of the study, the production of measurements of the re-sponse values of the real system and the simulation model, the computational stage with the assessment of the adequacy of the simulation model to real systems, the determination of the adequacy of the simulation model. The evaluation method for assessing the models' adequacy includes the selecting stages the criteri-on for the adequacy of the simulation model to the subject of research, making measurements of the re-sponse values of the real system and the simulation model, the computational stage with the assessment of the adequacy of the simulation model to actual systems, determining the adequacy of the simulation model.

| 1 | 2 | 3 | Next → ►