Vol 53, No 3 (2017)

Time-optimal algorithms of localization of pulsed-point sources randomly distributed over the search interval and detected due to generation of instantaneous pulses (delta functions) at random time instants are developed. Optimal search procedures ensure prescribed accuracy of localization and depend on the number of receivers in the detector.

This paper describes the method for determining the three-dimensional velocity vector of the flow using the laser Doppler measurements of unevenly sampled components. This method is based on the filtering and linear interpolation of the measured velocity components in the circuit channels. It is shown in the numerical simulation and physical experiment that the method error does not exceed 2% and the results of interpolation with respect to the closest points are several times greater.

This paper considers issues related to the identification of the parameters and form of the probability density function of generally non-Gaussian additive and multiplicative noise affecting the signal. The results of numerical simulation of methods for estimating the information parameters of random processes with a non-Gaussian probability density function for a finite sample.

This paper touches upon the synthesis of high-quality images in real time and the technique for specifying three-dimensional objects on the basis of perturbation functions. The recursive search method for the image elements of functionally defined objects with the use of graphics processing units is proposed. The advantages of such an approach over the frame-buffer visualization method are shown.

A monochromator capable of suppressing spurious laser radiation to a level required for low-frequency (<100 cm⁻¹) Raman spectroscopy is proposed. It has a high spectral resolution, contains a small number of optical elements, and can be easily included into the experimental optical scheme. The effect of using this monochromator in low-frequency Raman spectroscopy is illustrated by examples with test samples.

Potential accuracy of methods of laser Doppler anemometry is determined for the singleparticle scattering mode where the only disturbing factor is shot noise generated by the optical signal itself. The problem is solved by means of computer simulations with the maximum likelihood method. The initial parameters of simulations are chosen to be the number of real or virtual interference fringes in the measurement volume of the anemometer, the signal discretization frequency, and some typical values of the signal/shot noise ratio. The parameters to be estimated are the Doppler frequency as the basic parameter carrying information about the process velocity, the signal amplitude containing information about the size and concentration of scattering particles, and the instant when the particles arrive at the center of the measurement volume of the anemometer, which is needed for reconstruction of the examined flow velocity as a function of time. The estimates obtained in this study show that shot noise produces a minor effect (0.004–0.04%) on the frequency determination accuracy in the entire range of chosen values of the initial parameters. For the signal amplitude and the instant when the particles arrive at the center of the measurement volume of the anemometer, the errors induced by shot noise are in the interval of 0.2–3.5%; if the number of interference fringes is sufficiently large (more than 20), the errors do not exceed 0.2% regardless of the shot noise level.

This paper presents the results of measurement of the surface potential and the dwell time of Rb and Na atoms on the surface of S-52 molybdenum glass. It is found that at temperatures below the glass transition temperature, the temperature dependence of the dwell time of Rb atoms is well described by the Arrhenius formula. The surface potentials for Rb and Na are measured to be 0.67 and 1.37 eV, respectively. At temperatures above the glass transition temperature, the dwell time of these atoms increases abnormally. The reason for this is that during impact of an atom on the surface of molten glass, it can penetrate into the volume of the window and then return by diffusion and desorb from the surface. In this case, the dwell time of the atom on the glass is determined by the diffusion time and can be very significant, despite the relatively low potential barrier at the surface and high temperature.

A possibility of creating a capacitive accelerometer for measuring high-g accelerations (up to 10⁶g and higher) is discussed. It is demonstrated that insertion of a thin electret film with a high surface potential into the gap between the electrodes ensures significant expansion of the frequency and amplitude ranges of acceleration measurements, whereas the size of the proposed device is smaller than that of available MEMS accelerometers for measuring high-g accelerations. A mathematical model of an electret accelerometer for high-g accelerations is developed, and the main specific features of accelerometer operation are analyzed.