The purpose of the research is to develop recommendations for the use of mobile platforms for monitoring, early detection and assessment of fire danger based on the use of heterogeneous information about the state of the object of study and modern information and intelligent technologies that improve the quality of decisions about the fire situation.

Research methods. To assess the nature and type of fires, it is proposed to use sensors of electromagnetic fields generated by the flame, in combination with sensors of chemicals and burning temperature. Neural networks and hybrid fuzzy decision rules are used to synthesize mathematical models for assessing the fire situation and predicting and evaluating the condition of people in the fire zone.

Results. In the course of the conducted research, a range of the main tasks solved by mobile platforms interacting with automated operator workstations was formulated. The types of registered signals in the controlled zones are selected and recommendations for the selection of the element base of mobile platforms are formed. Recommendations are given for the synthesis of decisive rules for assessing the fire situation, forecasting and assessing the condition of people in the fire zone. An example of the classification of smoldering and open flames is given.

Conclusion. In the course of the conducted research, a list of tasks solved by mobile platforms during monitoring, early detection and assessment of fire danger was determined. It is shown that as a basic system of fire situation sensors, sensors of electromagnetic fields of the ultraviolet, optical and infrared range should be used in combination with sensors of temperature and chemical composition of substances in the area of ignition. It is advisable to select the composition of sensors and synthesize the decisive rules for classifying the type and nature of fires using modern mathematical methods, information and intelligent technologies.

The purpose of research is to substantiate the choice of developing a mobile solution for telemedicine technologies in ophthalmology. Creation of a mobile application for the recognition of photographs of images of the eyes of patients for diagnosing diseases of the retina of the organs of vision. The application is required for use with special equipment available in polyclinics fundus camera.

Methods. For a platform of mobile devices running the Android operating system, an application implementation containing a machine learning algorithm that diagnoses the presence of pathology of eye diseases.

Results. The article describes the results of the creation of a mobile application for telemedicine in ophthalmology, which collects, analyzes and processes digital information of eye diseases for diagnosis and diagnosis by an ophthalmologist. Description of the way devices interact and the technology of the client-server architecture of the model. The life cycle of the development of a pilot version of a software product is presented, future development prospects are indicated. This paper proposes the concept of a mobile tele-ophthalmology information system for deep learning-based image recognition. On the basis of the proposed concept, a mobile solution has been designed, implemented and tested, which improves the quality of medical service. The use of a mobile application for the primary examination by an ophthalmologist of patients and the diagnosis of visual impairment can increase productivity and efficiency of work, reduce the economic costs of providing services to the population. The doctor will get access to the necessary information as quickly as possible to make a decision on the patient's treatment, spend less time filling out outpatient cards and medical records, spending more time directly working with the patient.

Conclusion. As a result, the likelihood of medical errors will be minimized. As a result, the quality of medical services for the population will improve.

Purpose of research. Wireless sensor networks are actively developing and perspective direction in the information and communication field. Computational complexity optimization of data processing algorithms for wireless sensor network devices is still an important scientific and technical problem. This article is devoted to an algorithm that realized the method of intelligent quasi-indifferent data aggregation for decentralized devices - sensor nodes. The aim of the study is to further aggregation technologies improvement in wireless sensor networks by creating the new algorithm for quasi-stationary measurement data aggregation of sensor nodes. The developed algorithm ensures efficient quasistationary measurement data aggregation of sensor nodes by presenting this data in the form of the parabolic regression model coefficients vectors and combining them into groups based on the dynamic variations correlation of the recorded parameters, also takes into account undefined values and outliers in data segments and implements their elimination.

Methods. Methods of algorithms theories, probability theory, mathematical statistics, calculations in terms of complexity theory and technical calculations application software Matlab were used in the study during the theoretical research and algorithm development.

Results. The algorithm for quasi-stationary measurement data aggregation of sensor nodes that allows to reduce their volume when transmitting via wireless communication channels is developed. The practical significance of the developed algorithm lies in implementing the proposed theoretical and algorithmic structures to a level of programs.

Conclusion. The developed algorithm ensures efficient aggregation by presenting quasi-stationary measurement data of sensor nodes in the form of the parabolic regression model coefficients vectors, minimizes temporal and spatial correlations of data on the receiving side, eliminates undefined values and outliers in data segments exceeding three average absolute deviations.

Purpose of research. Development of a method for decoding electromyosignals in control systems for exoskeletons with virtual reality, allowing to adapt the rehabilitation program of a robotic device to the functional state of the patient.

Methods. To restore the motor functions of the lower extremities of post-stroke patients, it is proposed to use a biotechnical system with a robotic device, the control of which is based on the analysis and classification of electromyosignals. The robotic device is controlled by a fuzzy neural network. The formation of the vector of informative features for the neural network is carried out by means of a multilevel comparator, the number of levels of which is determined by the dimension of the vector of informative features determined by averaging the outputs of the comparators in a sliding window. The electromyosignal decoder includes a series-connected block of comparators, a block for calculating informative features, a multiplexer, a first neural network, a memory block and a second neural network, the outputs of which are intended to be connected to a servo motor controller, and a synchronizer connected by an output to the control inputs of the multiplexer, memory unit and servo motor controller.

Results. A classifier of electromyosignals has been developed, which is characterized by the use of multiple duplicate channels of EMG signals associated with a muscle or muscle groups that control the movement of the same joint of the extremities, as a result of which, at the output of the classifier of each channel, we obtain a number corresponding to the confidence in the command to rotate the servo motor of the exoskeleton, all the outputs of the channel classifiers are fed to a fuzzy neural network, the defuzzifier of which generates a control signal to the servo motor controller. In the course of the work, a software application was written that can control the exoskeleton using the analysis of electromyosignals.

Conclusion. The study showed that it is possible to change the indicators of clinical outcome in patients with subacute stroke experience after 12 sessions of BPS training. A biotechnical system with fuzzy control of a robotic device allows for an individual strategy for the rehabilitation of post-stroke patients (including targeted walking training).

The purpose of the research is to develop a method for recognizing the class (level) of quality of service in network routes under conditions of rapid changes in packet traffic and implicit (partial) monitoring of the level of quality of service in the links of the packet communication network.

Methods. In order to timely assess the level of service quality and reduce the load of network routes with test traffic, it is proposed to monitor the level of service quality only in part of the links of the packet communication network, the network route based on software and hardware sensors using ICMP echo requests, IP SLA trackers, TWAMP, etc. To recognize the quality of service in probable network routes at the stage of preparing initial data: important links from the network routes are determined based on the coefficients of paired correlation calculated using statistics between the values of the same quality indicators for different links and network routes, expert judgments about the weight of each service quality indicator; the packet traffic of the corporate network is divided into classes with different levels of service quality in the space of QoS quality indicators; fuzzy models are constructed for different classes (levels) of service quality for probable network routes and network links, based on expert judgments and the method of analyzing hierarchies, the functions of belonging gradations of features (models of the level of service quality in important network links, values of service quality indicators in network links) to the described (modeled) object. Within the framework of the developed method, models of service quality classes in important network links are consistently recognized based on the observed values of service quality indicators in the network links, and then models of the service quality class in the network route are recognized based on the recognized models of the service quality level in important network links.

Results. As part of the research experiment, the recognition of the level of service quality in the network route was performed in hypothetical situations under the conditions of accepted restrictions on changing the values of the observed indicators of service quality in the network links. At the same time, as a basis for constructing fuzzy models of different classes of service quality and for simulating the observed values of service quality indicators in the network links, the authors used statistical models based on theoretical calculation and known statistics for conditions close to the experiment. The expediency of using the developed method for recognizing the level of service quality in a network route under conditions of partial monitoring and changes in packet traffic under the conditions of accepted restrictions is proved.

Conclusion. The application of the proposed method for recognizing the level of service quality by packet traffic network routes in conditions of partial monitoring will increase the efficiency and reliability of monitoring the quality of service in network routes, reduce the load of transmission lines in autonomous packet communication networks, large corporate networks.

The purpose of the research is to study and search for options for the structure of the receiving path of the automatic dependent surveillance-broadcast message system.

Research methods are based on structural-parametric synthesis of measuring and computing facilities. Variants of the architecture of automatic dependent surveillance-broadcasting systems with a detailed composition of the receiving, converting and processing parts have been created. The study of the options for the scheme of the receiving path of the electronic module for receiving messages of automatic dependent observation-broadcasting showed that the best (in terms of gain) is the option with a sequential two-stage "gain-filtering" part, and the most reliable (in terms of the operating voltage level) is the option with an integrated processing circuit signal.

Results. Small astronautics is one of the most promising areas for the development of high-tech products. Small spacecraft as autonomous research mini-laboratories or robotic systems are used to solve problems of remote sensing of the Earth, organize space communication systems, conduct scientific experiments in space, etc. one area, the functions of which are expanded through the use of small spacecraft as additional transmitting and receiving means. Operating in an orbit of 400-500 km, a grouping of small spacecraft provides reception of messages of automatic dependent observation-broadcasting and their further processing in order to highlight the flight characteristics of aircraft (coordinates, speed, course, altitude, etc.). The obtained variants allow us to detail the architecture of the module and the structure of the receiving path, depending on the selected target indicators (gain coefficients, signal-to-noise ratio, application power).

Conclusion. The analysis of the module architecture options made it possible to detail the structural organization of the electronic module for receiving messages as part of the blocks for receiving, digitizing the signal and decrypting ADS-B messages. It is show, that the most critical block of signal reception should be evaluated in terms of gain, signal / noise, application power).

The purpose of the research is to use the potential didactic possibilities of discussion as a pedagogical technology for the development of soft skills of future IT specialists.

The research methods is based on the competence-based, contextual and integrative approach in education. We used traditional methods for theoretical and applied research - analysis, synthesis, pedagogical observation, systematization of facts. The material of this research is scientific works that explore practical issues of training future IT specialists in universities, as well as our own experience in organizing and conducting discussions on the results of practices provided for in the educational program of the "Software Engineering" direction.

Results. The relevance of the study is determined by the need to improve the educational process based on the widespread use of interactive technologies in higher professional education for training specialists of a qualitatively new level in the context of the development of the digital economy. These specialists must be competitive and in demand in the labor market, be able to solve problems in a rapidly changing situation. The experience of organizing and conducting discussions based on the results of practical training has proved that the discussion has great opportunities for professional training and development of soft skills of future IT specialists, since it creates situations and conditions for the discussion participant to form, develop and consolidate personal qualities, ensures the manifestation of individuality and active involvement of students in the search and proof of truth.

Conclusions. Discussion is possible only in the implementation of preliminary theoretical and practical training of students on the problem under consideration. As a pedagogical technology of interactive learning, it will ensure success only when analyzing problems for which there is no unambiguous solution, which require active joint discussion, have a personally and professionally significant character for the participant. A significant success factor is the willingness of the practice manager, experts and students to discuss, who are obliged to responsibly accept the rules of interaction before the discussion and observe them by each participant during the discussion, to have a culture of dispute management, adequate self-esteem, openness of thinking, breadth of creative imagination, tolerance.

The purpose of the research to improve the quality of fire situation assessment with an assessment of the health status of people in the fire zone based on the use of the methodology of synthesis of hybrid fuzzy decision rules and neural network technologies.

Methods. In the course of the conducted studies, it was shown that to improve the quality of decisions about the type and nature of fires, it is advisable to use three groups of sensors that measure the intensity of electromagnetic radiation in the ranges of ultraviolet, visible and infrared spectra, chemical composition (carbon monoxide, carbon dioxide, phenols, etc.) and temperature. An analysis of the nature, nature, types and characteristics of various types of fires has shown that a detailed classification of these classes of phenomena refers to poorly formalized problems with a fuzzy data structure. The assessment of the condition of people in the fire zone also belongs to the same class of tasks. Taking into account this circumstance, the fuzzy logic of decision-making was chosen as the basic mathematical apparatus, and in particular, the methodology for the synthesis of hybrid fuzzy decision rules in combination with neural network technologies.

Results. Using the methodology of synthesis of hybrid fuzzy decision rules as a basic mathematical theory, taking into account the specifics of the data structure, the paper proposes a method for synthesizing mathematical models for assessing the fire situation and the condition of people in the fire zone during the implementation of which, using the theory of measuring latent variables, a space of informative features is formed, fuzzy hybrid models of classification of types and nature of fires are synthesized, models of forecasting the development of the fire situation are built and models of forecasting and diagnosing the condition of people who are sent and are in the fire zone are synthesized. The effectiveness of the proposed method was tested on the task of assessing the severity of carbon monoxide poisoning of people in the fire zone.

Conclusion. The practical application of the proposed method allows us to create high-quality systems for assessing the fire situation and the condition of people in the zone of the studied class of emergency situations located on mobile and fixed platforms.

The purpose of the research is to propose a methodology for estimating the preparation time of the return stage of a space rocket for re-launch based on the Student's t-distribution using the Euler gamma function.

Methods. Statistical analysis of the dynamics of repeated launches of the first stage of the Falcon 9 launch vehicle (PH) showed that the first re-launch of the return stage was at 1046.2 was carried out on August 7, 2018, 88 days after her landing on the OCISLY offshore platform. In total, 3 re-launches of this stage took place in the period from 2018 to 2020. The average time interval required to prepare this stage for subsequent launch was 205 days. The 1058 stage, which was first launched in May 2020, was re-launched 6 times (the last time was in March 2021). The average time interval required to prepare this stage for subsequent launch was 50 days. Analysis of the available data shows that from the moment of the first re-launch of the stage in August 2018 to the present, the average preparation time of the returned stages for the next launch has decreased fourfold - from 205 days to 45-50 days.

Results. The application of the proposed methodology to the sampling of the time intervals of preparation for the relaunch of stages B1056, B1058, B1059, B1060 in the period from 2019 to 2021, allows us to obtain the following values of the time of preparation of the returned stage of a space rocket for re-launch: t = 55 ± 10 days, i.e. t (45; 65) days, which is confirmed by the analysis of the presented statistical data. The discrepancy between the calculation results and real data is within the statistical error.

Conclusion. The presented methodology makes it possible to estimate the average time required to prepare the return stages of the launch vehicle for subsequent launches, and, consequently, to predict the capabilities of participants in space activities who have reusable space systems at their disposal to increase their orbital groupings. Therefore, the proposed methodology can be used in the development of algorithms, models, methods for estimating the preparation time of the returned stages for re-launches of the PH, as well as for evaluating the effectiveness of the use of reusable (partially salvageable) PH of various classes for their intended purpose.

The purpose of the research is to improve the quality of classification of the severity of carbon monoxide poisoning through the use of hybrid fuzzy decision rules.

Methods. In the course of these studies, it was shown that modern approaches to classifying the severity of carbon monoxide poisoning are based on assessing the concentration of carbon monoxide in the air of the working area or carboxyhemoglobin in the blood, and there is a significant uncertainty zone between the studied classes of conditions, which reduces the quality of decisions made and makes it difficult to choose adequate treatment regimens In order to reduce the uncertainty zone, it is proposed to increase the number of severity classes and to adequately describe the uncertainty zones, use the methodology for the synthesis of hybrid fuzzy decision rules with a focus on models for assessing the health status of a person changing under the influence of exposure to harmful chemicals.

Results. Using the methodology of synthesis of hybrid fuzzy decision rules, we have obtained fuzzy mathematical models for the allocation of such classes of severity as normal, mild, medium, severe and critical degrees of poisoning with the confidence of decisions made no worse than 0.95, which allows us to accurately assess the health status of patients with the appointment of adequate prevention and treatment regimens.

Conclusion. In this paper, using the methodology of synthesis of hybrid fuzzy decision rules, fuzzy models of classification of the severity of carbon monoxide poisoning with the allocation of classes: normal, mild, medium, severe and critical poisoning are obtained. During the expert evaluation and mathematical modeling, it was shown that the diagnostic sensitivity, specificity and confidence in the decisions made exceeds the value of 0.95, which is a good practical result for this class of problems.