Vol 62, No 5 (2019)

A fast algorithm is proposed for choosing the blur factors of kernel functions of a non-parametric probability density estimate under conditions of large-scale statistical data. It is shown that the basis of the algorithm is the result of a study of the asymptotic properties of a new kernel probability density estimate. The properties of the developed algorithm are analyzed and the method of its application is formulated.

We obtain the dependence of the error variation of the overlap estimate of the Allan variance on the overlap coefficient in a given observation time interval with frequency fluctuation noise of the main types. Analytical expressions of normalized autocorrelation functions of frequency variations are used.

A comparative analysis of the results of estimation of measurement uncertainty is conducted. The results are obtained by a method described in the Handbook on the Expression of Uncertainty in Measurement and also by Monte-Carlo methods and empirical and Bayesian methods recommended in a tentative version of the Handbook. Examples of calculations of measurement uncertainty for several measurement techniques in the field of analytic chemistry are presented. Features of the application and the advantages and drawbacks of the methods are described.

A method is proposed to solve the problem of eliminating the ambiguity of reading information from a quasi-cyclic code scale. The solution is derived with the usage of a smaller number of reading elements by comparison with the application of the “dual brush” classical method. It is shown that the benefit in the number of required reading elements increases with the increase in the number of bits of the code scale. An example of a scale implementing this method is presented. The proposed quasi-cyclic scales are required in the development of digital angle converters and are used as an alternative to classical code scales, the mask of which is implemented in regular binary code.

A new algorithm has been developed for the determination and elimination of jumps in the Melbourne–Wübbena combination, formed from the code and phase measurements in global navigational satellite systems. The algorithm more reliably, in comparison with analogous known algorithms, detects jumps in the data that are associated with time losses in the radio-receiving devices for phase capture of RF signals transmitted from satellites. Determination of the positions and values of jumps in a Melbourne–Wübbena combination is performed against the noise component and outliers. The algorithm was tested for actual measurements in global navigational satellite systems at two carrier frequencies.

The most important sources of errors originating in monitoring angle-measurement structures (scales, circular scales, code disks) are obtained from within the reading systems of measurement devices. Prospects for using the differential method of measuring metrological parameters of angle-measurement structures are shown in the case of production real-time monitoring.

We propose the principle of operation of a special flowmeter of accumulative type and develop its design and electric circuit. This device is intended for the automatic monitoring of the process of wear of hydraulic seals of the ship pumps by means of periodic measuring of the leaks of liquid. It is shown that the device is adapted to the joint operation with the complex on-board system of technical diagnostics. The necessity of application of devices of this kind is explained by the fact that the rapid wear of seals is one of the main causes of idle time of the pumps.

A third-generation algorithm that implements the method of mirror noise images for extracting a useful signal from noise is developed. In the proposed algorithm initial processing of the initial signal is realized by means of a second-generation algorithm and the second half of the signal is then generated by a reversal of that segment of the signal where the noise is suppressed. The useful signal is practically invariant in the newly created signal while the noise signal is substantially distorted which in the course of subsequent filtration leads to an additional increase in the signal-to-noise ratio by comparison to narrow-band filtration of signals and the first- and second-generation algorithms. The algorithm is intended for information-measurement systems that function according to the principle, switched on – measured – switched off. It is shown that the new algorithm produces a substantial gain in signal-to-noise ratio by comparison with narrow-band filtration of signals and the first- and second-generation algorithms.

A method for the construction of microwave devices with longitudinal interaction is proposed. Devices of the present type create uniform cross-sectional distribution of temperature of rods made of polymer composite materials. Results from theoretical and experimental investigations of the cross-sectional distribution of the temperature of the material of the rod as well as the parameters of the microwave device are presented.

Algorithms for improving noise suppression in hydrophones are discussed. An increase is attained by using a priori information about the structure of the surface noise of the medium.

A silicon detector with a longitudinal silicon wafer has been developed. x-Ray spectra at the K-absorption edges of Au, Pb, and Bi are used to measure its energy resolution. The results are compared with measurements based on x-rays from a ²⁴¹Am gamma-ray source. The dependence of the energy resolution of the detector on the noise of a charge-sensitive amplifier and on statistical fluctuations in pair production of carriers in the silicon is calculated.

At the launch time, the engines of aircrafts emit electromagnetic signals uniquely described by a set of characteristics called signature. The simulators of signatures of the actual thermal objects are often used in testing the aircraft equipment. We analyze the main causes of errors appearing in the simulation of signatures on the bases of synthesizing of a three-dimensional function that describes the directivity pattern, spectral composition, and shape of the startup pulse. The computational relations for the error of the simulated parameter are presented. The error of the simulated parameter of the simulator is estimated according to the results of experimental investigations.