Organizations

Major Research Results

  • Main Research Results in 2016

    • Developed the methods, models, algorithms and software tools for visualization of nonformalized data that differently from the currently existing ones implement hierarchical component-oriented architectures, advanced libraries of visualization models, universal interfaces for source data import and visualization results export to the external systems, that provide for highly efficient processes  of monitoring and information security management. Dr. Sci. (Tech) Prof. Igor V. Kotenko, Dr. Sci. (Tech) Prof. Igor B. Saenko

    • Developed the models of main elements of the users’  infomobility support system namely the : (1) model of infomobility support system based on the “blackboard” architecture and knowledge ontological representation; (2) ontological model of the task at hand context where the methods are implemented as computational services; (3) ontological model of the user context consisting of information given by the user (his/her profile) and environment information collected automatically; (4) ontological models of the infomobility support system services. Dr. Sci. (Tech) Prof. Alexander V. Smirnov, Dr. Alexey M. Kashevnik

    • Proposed an architecture for multimodal information and navigation cloud system aimed at corporate cyber- physical intelligent space (environment) incorporating a plurality of interrelated services intended for  users’ authentication, corporate television, visitors’ navigation, videoconferencing,  sensory information processing and, at that, functioning in concordance with scenarios of interaction with the users and an enterprise protection ensured by integration of heterogeneous infrastructure components (information screens, turnstiles, video surveillance cameras and other proximity sensors, network and computing resources).  Ms. Irina V. Vatamanyuk, Mr. Dmitry K. Levonevsky, Dr. Sci. (Tech) Prof.  Andrey L. Ronzhin

    • Developed the set of technologically ordered scalable, efficient and stable algorithms intended for Big Data processing to solve data mining and machine learning task for data values assessment and prediction as well as for decision making. The Big Data processing technology comprises (1) task-driven adaptive granulation algorithms, (2) application-driven ontology-based semi-automatic semantic data meta-model and data structure design algorithm, (3) causal association discovery in Big Data and (4) optimization of causal decision making model algorithm. The algorithms are designed to process heterogeneous data of big sizes and dimensionalities including natural language texts. The Java-based software library implementing all the algorithms is developed. The developed complex of algorithms and software are validated experimentally using real-life Big Data sets.  Dr. Sci. (Tech) Prof. Vladimir I. Gorodetsky, Dr. Olga N. Tushkanova

    • Developed mathematical methods, models, algorithms and software prototype for aggregation, normalization and analysis of large volumes of heterogeneous structured, semi-structured and unstructured data for monitoring and secure managing the “Internet of Things” networks. The above  apparatus is based on the technology of parallel data-flow processing differing by the use of genetic algorithms for load balancing, providing high scalability, validity, performance, and stability of data processing and focused on the use in a wide range of information infrastructures based on the "Internet of Things" concept. Dr. Sci. (Tech) Prof. Igor V. Kotenko, Dr. Sci. (Tech) Prof. Igor B. Saenko

    • Done a comparative analysis of the status and practices of the national legislation of the Collective Security Treaty Organization (CSTO) member States in the sphere of protection of state secrets. Exposed the main provisions of legal regulation of ensuring international information security in the CIS and the CSTO.  Developed a draft Model Law of the CSTO "On State Secrets" that is a kind  of "bridge" between international and domestic laws contributing to the introduction into the national regulatory system of principles and norms of international law, thus,  transporting them in a normative-concentrated  form into the national legislative acts.   Dr. Mikhail A. Vus,   Cor. Member of RAS Dr. Sci. (Tech) Prof. Rafael M. Yusupov

    • Developed a methodology and an intelligent technology intended for design and implementation of a certain universal software platform and a matching distributed multifunctional decision-making support system at control of the state-level objects.  The technology is based on the combined logic and dynamic models, methods and algorithms of proactive monitoring and management that allow for real-time processing of super big amounts of arriving measurement information that contains erroneous, inaccurate and contradictory data and for making timely and valid control actions. Dr. Sci. (Tech) Prof. Boris V. Sokolov, Dr. Sci. (Tech) Prof. Vyacheslav A. Zelentsov.

    • Developed and studied the hybrid acoustic models, in which Deep Neural Networks (DNN) are used for estimating the posterior probabilities of a Hidden Markov Model state, have been developed and studied. An implication of DNN with a various number of hidden layers, hidden units, and different activation functions is researched. The developed models were introduced in a very large vocabulary automatic Russian speech recognition system (the vocabulary size is more than 150K word-forms). The use of the hybrid models allowed decreasing a relative word error rate by 20% with respect to the baseline system with Gaussian mixture models.  Dr. Sci. (Tech) Associate Professor Alexey A. Karpov, Dr. Irina S. Kipyatkova

    • Developed an ontological model for self-organisation of a group of mobile robots based on the publish/subscribe mechanism.  The ontological model consists of robot ontologies and task ontologies published in the smart space (SS). The robot ontology describes its competencies and constraints to be satisfied. When a task is published in the smart space, its ontology is matched with robot ontologies. Based on the matching results the robot or several robots are chosen for task solving. They start to solve the task and publish the result in the smart space. If another robot or robots meet the requirements of the updated task, they perform corresponding actions. This sequence is repeated until the task is solved completely . Dr. Sci. (Tech) Prof. Alexander V. Smirnov, Dr. Alexey M. Kashevnik

    • Scientific basis has been developed for employing the method of structural resonance (SR) for solving the fundamental problem of discovering the patterns of chemical structures and structure-activity relationships (SAR). The unique capabilities of the SR method for finding strong and complex logical associations in multidimensional data as well as its ability to analyze Big Data allow to use known descriptor systems of chemical compounds in a new way, operate the new descriptor systems and create high-precision SAR models.   It has been shown that, compared to rivaling solutions, the SR method shows higher precision and output as well as interpretability of the results that allows to carry out targeted design of pharmaceutical preparations. Dr. Sci. (Tech) Vyacheslav A. Duke

    • Proposed the software defined method for real time remote aircraft monitoring that reduces the limit of simultaneously monitored aircrafts.  The novelty consists in the dynamic organization of software defined communication channel with the aircraft on the basis of public communication networks. The novelty is in the improved efficiency of the radio channel use and providing legacy support, technology transition, which allows the monitoring and the localization of the aircraft  in real time. At that, technical limitations on the number of aircrafts subjected to simultaneous monitoring in most cases are withdrawn.  Dr. Sci. (Tech) Sergey V. Kuleshov,  Dr. Alexandra A. Zaytseva,  Dr. Alexey J. Aksenov

    • Proposed approaches to developing the software-reconfigurable telecommunication systems using the concept of active data. The novelty consists in a software reconfiguration of formats in the application of the concept of active data and the use of the principle of nested containers in a configurable format. The importance of the approach is the ensuring of an access to telecommunications services at points not covered by the public cellular networks, and the use of special radio equipment or satellite system is impractical or at the dynamic deployment of specialized data networks from the standard telecommunication devices.   Dr. Sci. (Tech) Sergey  V. Kuleshov,  Dr. Sci.  (Tech) Prof. Victor V. Alexandrov, Dr. Alexandra A. Zaytseva.

    • Developed theoretical basics and practical approaches to building a software defined systems in support of import substitution of hardware components by replacement of specialized branded import  components of their software implementations. The novelty consists in the replacement of specialized branded import components by their software implementations. The importance is to develop a software defined data channels (technology development SDR), allowing to save on infrastructure specialized channels (physical transport component) by the software implementation of the required properties are implemented on the basis of the available channels as well as  in the flexible management of the properties of such channels. Dr. Sci.  (Tech) Prof. Victor V. Alexandrov, Dr. Sci. (Tech) Sergey  V. Kuleshov,  Cor. Member of RAS Dr. Sci. (Tech) Prof. Rafael M. Yusupov

    • Designed a new system-on module characterized by the use of an external real-time microchip; an independent power supply system; low cost and a small size (82 * 50 * 5 mm); low power consumption (2-3 W); the processor frequency of up to 1000 MHz; RAM memory DDR3 of up to 1024 Mb; non-volatile random-access memory of up to 8GB. It supports external interfaces of peripheral devices (USB, UART, ADC, GPIO, Ethernet and others.) and is intended for the processing of sensory information and control of activation devices in embedded systems and mobile robotic complexes [19, 31]. Dr. Vladimir P. Dashevsky,  Dr. Victor Yu. Budkov, Mr. V. G.  Rzhimsky

    • Developed an approach to check feasibility of real-time multi-task applications in various combinations of scheduling disciplines and protocols of access to shared information resources, when applications run on a multi-core computer platform and the application structure is specified with a simple formalized profile of segments of three types. The approach is based on the concept of software application density which characterizes the potential efficiency of computer resource usage by the profiled application, determined by assessing its feasibility depending on the processor performance with the developed software simulation tools. Dr. Sci. (Phys. & Math.) Prof. Sergey N. Baranov, Dr. Sci.  (Tech) Prof. Victor V. Nikiforov

    • Developed an approach to the construction of employees’ social graph based on social and structural data from social network sites. Methods for constructing user's vulnerabilities profile based on such information from accounts on social network site were analyzed.  Also proposed a probabilistic graphical models (PGM) that incorporate copula as other class of probabilistic graphical models for the problem of estimation of person's risky behavior rate on the data about several last episodes.  The  developed  approaches and models facilitate the estimation of  the informational system's working processes sustainability in case of social engineering attack actions on the users or group of users of the informational systems .  Dr. Sci. (Phys. & Math.) Associate Professor Alexander L. Tulupyev, Dr. Alena V. Suvorova

  • Main Research Results in 2015

    • Developed the concept and methodology of exploring the potential of complex socio-technical systems (CTS) under risk conditions; introduced principles, models and methods that support an analytical evaluation of the complex CTS potential indices  in risk conditions and allow to reveal the functional dependencies of these indices values from the characteristics of being made decisions. The results were successfully applied to the solution of actual scientific problems to improve the defense industry and at solving  other important government tasks. Dr. Sci. (Tech) Prof. Igor V. Lysenko,  Dr. Alexander S. Geida .

    • Developed the model of the malefactor committing social and engineering attacking impacts on the information systems’  users;  its critical properties are particularized. Dr. Sci. (Phys. & Math.) Associate Professor Alexander L. Tulupyev, Dr. Alena V. Suvorova

    • Developed the "Commentary to the Model Law of CIS IPA "On State Secrets" ¬in the interests of ensuring national security of the Russian Federation. The document was adopted by the CIS IPA Resolution No. 43-16 of 26.11.2015 for the use by  CIS Member States.  Dr. Mikhail A. Vus,   Cor. Member of RAS Dr. Sci. (Tech) Prof. Rafael M. Yusupov

    • Computer system for paralinguistic analysis of natural speech for automatic recognition of human’s emotional states using speech classification and speech paralinguistic phenomena. The system won the First  place in the International competition within the 16-th International conference Interspeech-2015 (Dresden). Dr. Sci. (Tech) Associate Professor Alexey A. Karpov, Dr. Irina S. Kipyatkova, Ms. Vasilisa O.  Verkhodanova V.

    • Developed methodology and smart technology, and software complex (SC) for an operational flood forecasting based on a complex anticipatory modeling and integration of heterogeneous data. SC provides for hourly geolocated contours and the depths’ map, a list of objects in the zone of risk of predictable flood, automatic notification of interested persons and entities, visualization of simulation results using the developed geo portal. Dr. Sci. (Tech) Prof. Boris V. Sokolov, Dr. Sci. (Tech) Prof. Vyacheslav A. Zelentsov.

    • Developed  the theoretical basics and firmware of the technological platform for the design and operation of multimodal interfaces based on spatial-temporal structuring of data about users, application modules of multichannel processing of audio and video signals, characterized by a focus made on design and support of information management services and heterogeneous robot systems. Dr. Sci. (Tech) Prof.  Andrey L. Ronzhin, Dr. Victor Yu. Budkov, Dr. Anton I. Saveliev.

    • Developed the efficient algorithmic support for building and optimizing the Big Data target models based on concepts’ semantic analysis that is a new approach to further develop the formal concept analysis, and also uses well-known concepts’ database to build and optimize the causal Big Data models for decision-making. Dr. Sci. (Tech) Prof. Vladimir I. Gorodetsky.

    • Developed a methodology for the self-organization of mobile robots’ group and computing resources for their joint work on a task in an intellectual space, that allows for a joint access to information and robots’ knowledge / resources of such a space using technology of ontology management. Dr. Sci. (Tech) Prof. Alexander V. Smirnov, Dr. Alexey M. Kashevnik

    • Exposed the previously unknown regularity of the indices’ distribution for a situational anxiety in people with malignancies. Dr. Sci. (Tech) Sergey B. Roudnitsky.

    • Developed methods for efficient representation and storage of spatial data obtained through 3D scanning of physical objects within the extension of the methodology for cognitive technologies concerning the additive technologies at physical level of info-communication processes. Dr. Sci.  (Tech) Prof. Victor V. Alexandrov, Dr. Sci. (Tech) Sergey V. Kuleshov,  Dr. Alexandra A. Zaytseva,  Dr. Alexey J. Aksenov

    • Developed a method of high-speed denied encryption, different in meeting a criterion of the computational indistinguishiability over the cipher text from the probabilistic encryption and prospective to be used as a mechanism of ensuring the information security against attacks that enforce  the messages’ sender and receiver to reveal the encryption key. The method can be applied to new methods of authenticating users, protecting critical information and guiding the attackers on the false path.  Dr. Sci.  (Tech) Prof. Nikolay A. Moldovyan.

    • Developed new theoretic  propositions, models, methods, and devices that implement a cognitive analog neural network operative processing of heterogeneous information with the use of memristors. The novelty lies in the development of associative and spatial addressing to the neural networks’ memory. The received solutions allow for a breakthrough in the development of compact cognitive analog neural network machines, new cognitive weapons, military and civil hardware. These machines will allow to operatively solve such application tasks with a difficulty that cannot be resolved through the modern supercomputers. In 2011-2015 five results in regard to the mentioned methods and devices have been patented. Dr. Sci.  (Tech)  Prof.  Vasily Yu. Osipov.

    • Developed the theory of computer sonovision of underwater objects in an ocean waveguide. The main idea consists in the forming  in the virtual space of the underwater environment of the distributions of the intensities of the acoustic field of underwater objects based on the agreed with the environment processing of  the signals received by an acoustic antenna. Novelty is determined by the development of new methods and algorithms for calculation of acoustic fields and the inverse problems solving in underwater waveguides. Dr. Victor I.  Ermolaev,  Dr. Dmitry V. Chirov.

    • The novelty is the Software platform for the development and operation of multimodal interfaces based on spatial-temporal structuring of data about users, application of modules for multichannel processing of audio and video signals in design and support of heterogeneous robot systems, some of them are successfully tested and operated in organizations of the RF Minpromtorg. Dr. Sci. (Tech) Prof.  Andrey L. Ronzhin.

    • Developed a computer system for paralinguistic analysis of natural speech for automatic recognition of human’s emotional states using speech classification and speech paralinguistic phenomena. Dr. Sci. (Tech) Associate Professor Alexey A. Karpov

    • Developed an intelligent information technology and distributed information system for the integrated processing and analysis of heterogeneous land-aerospace data intended for solving the monitoring problems, multivariate forecasting and proactive control of natural-technological objects and territories in a dynamically changing environment. Developed technology and information system approved in 2014-2015 at solving the problems of management of territories in urban and rural municipal districts of St. Petersburg and Leningrad region. Dr. Sci. (Tech) Prof. Boris V. Sokolov, Dr. Sci. (Tech) Prof. Vyacheslav A. Zelentsov., Dr. Semen A. Potriasaev.

  • Main Research Results in 2014

    • Scientifically grounded and developed: the project "Information security strategy" and three model legislative acts (for the states – participants of the CIS). The documents were adopted at 41st plenary session of the IPA CIS,28.11.2014 ; "Recommendations for rapprochement and harmonization of the national legislation of member states of the Collective Security Treaty Organization in the sphere of ensuring information and communication security". The document was adopted at the 7thplenary session of the CSTO PA 27.11.2014 (R.M. Yusupov, M.A.Vus)

    • We have created theoretical framework as well as software and hardware for technology platform for development and operation of multimodal interfaces based on spatial-temporal structuring data on users, application of modules of multichannel audio and video signals processing, differing by a focus on the design and support of personalized multimodal information management services and heterogeneous robotic systems. (A.L. Ronzhin)

    • We have developed a method of computer sonography of underwater objects in an ocean waveguide. Its essence lies in the estimation of the distributions of the intensities of the acoustic fields of underwater objects based on the harmonized with the environment processing of signals received by acoustic antenna. The novelty consists in the development of new algorithms for calculation of acoustic field and solution of the inverse problem in an underwater waveguide. The method is intended to illuminate the underwater environment in sonar underwater surveillance systems. (V.V. Popovich)

    • We have developed amodel and algorithmic support for the technology of model building of big data target variables. The technology is fully implemented in software, tested experimentally and confirmed its computational efficiency and performance. (V. I.Gorodetskiy)

    • We have developed an optional algorithm for analysis and prediction of the behavior of cloud applications that useprofile activity and time series forecasting models.  Anomaly detection is based on data received from the analyzer. The prospects of the method lie in the possibility of detecting new types of attacks. (V. I. Vorobyev)

    • We have developed a new method for constructing crypto schemes, allowing overcoming the difficulties of solving factorization and discrete logarithm problems simultaneously. The method provides for the extension of the range of cryptographic algorithms and protocols with high security level. (A.A. Moldovyan)

    • We have developed models of socially important behavior on the basis of Bayesian Belief Networks, which offer new approaches to processing specific data type (survey results) and provide support for decision making under circumstanceswhere it is impossible to organize long-term observation for data reception. (A.L. Tulupyev)

    • We have created algorithms and methods for preprocessing of electrophysiological data, consistent with mathematical methods for subsequent data mining. Tools for structural-linguistic processing and analysis of electrophysiological signals have been upgraded and new methodshave been developed that use constructive and statistical associations of parameters and fragments of electrophysiological signals, which allows improving assessment of functional patients’state and conducting a deeper analysis of the psychological effects. (S.B.Rudnitskiy)

    • We have developed an algorithm for compression of 3D data, which is characterized by the use of space-filling curve to reorder elements in point clouds and does not require a reconstructed surface of the object. The concept of replication of 3D model and its propertieshave been defined. (V. V. Alexandrov)
      We have developed models, a methodology and software environment for modelling of protection processes of information and telecommunication networks against infrastructure attacks.These models differ from the existing ones by the use of hierarchical componentbased architecture and developed libraries of attacks and protection mechanisms.These results are important for studying performance of various types of malware, search of optimal configurations of counteraction mechanisms andassessment of the effectiveness of their functioning. (I.V.Kotenko)