Oleksiy Roshchupkin

Mobile Ad Hoc wireless network for pre- and post-emergency situations in nuclear power plant

This paper describes the mobile Ad-Hoc (wireless) network (MANET) for emergency scenarios in nuclear power plant (NPP). Authors proposed the system with such properties as flexibility and a self-forming and self-healing network topology that dynamically adjusts to the moving configuration per each intermediate routing node. It is also proposed to integrate MANET and Bluetooth-like technologies to create an unmanned formation of autonomous quadcopters that provides both indoor and outdoor communications coverage inside and outside of the NPP.

Reducing the calibration points of multisensors

Goal of this paper is to develop the method identification of multisensors characteristic curve using neural networks and minimize the number of calibration points required to find an appropriate characteristic curve for the input-to-output relationship of multisensors. A proposed original method illustrates how to improve the prediction accuracy of calibration results for multisensors despite the high level of random error.

Development of Precision Information Measuring System for Ultraviolet Radiation

Results of studying the neural network method are presented to reduce the amount of calibration points for the multisensor (MS), in particular for the photodiode. This allows transmitti ...

Method of ensuring an interchangeability of the ultraviolet radiation sensors during a transition to its individual conversion function

This article show the results of the synthesis of the ultraviolet measurement system using photodiode. It was synthesized the generalized structure of the system and the scheme of its measurement channels. The measurement accuracy is increased by transition to individual conversion functions as a photodiode as well as measurement scheme. It is also proposed the scheme of data processing and evaluated the error of measurement of ultraviolet irradiation.

Multisensors signal processing using ANFIS

An improved algorithm of intelligent data processing by integrating the modified identification method of individual characteristic curve along with the adaptive neuro-fuzzy inference system (ANFIS) was proposed. The results of data prediction with training the ANFIS system and the different number of training epochs are described. The proposed approach provides the high accuracy of data prediction as well as low neural network training error.

Data communication crypto protocol for security systems sensor networks

This paper describes hardware implementations of crypto protocol elements for secured data communication in two wired networks of security system detectors. Those elements are basic to crypto protocol development for that type of network. In addition, hardware tools for the network controller and the network server that support this protocol were developed. These tools will allow us to test and debug the proposed security elements and the crypto protocol.

Methods for improving the accuracy of sensors with a significant influence of non-informative factors

During influence of non-informative factors on the sensors signals it is necessary to improve its accuracy. In this paper it described the reduction method of calibration points of multi sensors which allow significantly reduce the amount of calibration points which led the transition to individual transfer functions of such sensors and also is not too labor consuming. It is proposed the generalized block diagram of the system which provides the sensors signal processing. It is proposed the methods of providing sensors interchangeability. Also it is proposed the generalized scheme of the data transfer of measured physical quantities. The proposed solutions provide a base for improving the accuracy of sensors with a significant influence of non-informative factors.

Data acquisition system with low-accuracy sensors

The paper deals with improving the accuracy of the sensors under influence of non-informative factors. There are described the generalized block diagram of the data acquisition system and its measurement channels which provide the sensors signal processing. The generalized block diagram of the calculation process is proposed, and the structure of used neural networks is shown. Both scheme and algorithm of data processing are described. The measurement error of the proposed system was evaluated. It confirmed that proposed system adds the error of 0.32% which is very small in comparison with the etalon mean of 1%. The proposed system allows use of the sensors with a significant influence of non-informative factors.

Improved multisensors signal processing

The algorithm of the program was created to approximate the output signal of multisensor (MS) using the functional dependency of MS output voltage from MS operating frequency for a fixed thickness of tested films, the output voltage of the MS film thickness at a fixed frequency. The MS data obtained by sine wave eddy current measurement. It is investigated the influence of different type of function approximation accuracy of MS signals. It is shown that the smallest approximation error obtained using as polynomial function of the summation of first power polynomial and exponential function. This is indicates correctness and efficiency of noise reduction. This research is aimed to investigate and find effective, specialized and cheap tool for multisensor signals processing.