| Authors |
Kozyrev Gennadiy Ivanovich, doctor of technical sciences, professor, sub-department of telemetric systems, integrated processing and information protection, Military Space Academy named after A. F. Mozhaysky (13 Zhdanovskaya street, St. Petersburg, Russia), Е-mail: gen-kozyrev@yandex.ru
Kravcov Aleksandr Nikolaevich, candidate of technical sciences, associate professor, head of sub-department of metrological support of arms, military and special equipment, Military Space Academy named after A. F. Mozhaysky (13 Zhdanovskaya street, St. Petersburg, Russia), Е-mail: kan1970@bk.ru
Usikov Valentin Dmitrievich, adjunct, sub-department of metrological support of arms, military and special equipment, Military Space Academy named after A. F. Mozhaysky (13 Zhdanovskaya street, St. Petersburg, Russia), Е-mail: usikov_1989@list.ru
|
| Abstract |
Background. The relevance of the topic of work is due to the need for a discrete presentation of various models of continuous communications from a single energy and precision positions. The aim of the work is to obtain simple analytical expressions for estimating the polling (sampling) frequency depending on a given reconstruction error, the maximum frequency of the spectrum Fm of the initial message, taking into account the subjectivity of the determination of Fm when restoring the original message by Lagrange interpolation polynomials.
Results. The three most common models of measuring messages are considered, when the energy spectrum and the correlation function of the initial process are presented as: the response of an ideal low-pass filter (LPF) to white noise, the reaction of a low-frequency RC filter to white noise, the reaction of a Gaussian low-pass filter to white noise. The polling frequency was calculated using the approximate formulas obtained in the work when restoring the initial process for the above models using Lagrange interpolation polynomials. zero and first orders for a given error recovery. The determination subjectivi y was reflected by introducing a proportionality coefficient Kp corresponding to 100p – the percentage of the power of the initial process, enclosed in the frequency range [0, Fm]. The values of the parameter p were set equal to 0.9; 0.95 and 0.99, as the most commonly used in practice.
Conclusions. The introduction of the proportionality coefficient corresponding to 100p — the percentage of the power of the initial process, enclosed in the frequency interval [0, Fm], allows one to take into account the subjectivity of the determination of Fm and approach the analysis of the discrete representation of various models of continuous messages from the same energy and accuracy positions. In this case, there is no need to use other parameters of the energy spectrum, such as the effective value of the spectrum, the width of the spectrum at the level of 0.707, 50 %, 25 % of the power, etc.
|
| Key words |
interrogation frequency, energy spectrum, correlation function, maximum spectrum frequency, reconstruction error
|
