Abstract <p>The safety of aircraft flights largely depends on the quality of measurements of weather formation parameters using the pulsed Doppler weather radars. This paper discusses the features of estimating the width of the Doppler spectrum of velocities of optically unobservable weather formations based on the paired-pulse method. A characteristic feature of such weather phenomena is the low power of radar reflectivity. It is shown that the estimation errors depend on the power ratio of the weather formations themselves and the receiver internal noise. A method for eliminating this dependence is proposed. It is based on representing the correlation matrix of input influences in terms of its eigenvectors, dividing the matrix into signal and noise components, and subsequently reducing the noise component’s influence. The effectiveness of the proposed method is illustrated using simulation modeling results. Analytical expressions have been derived for the distribution density of the absolute and relative estimation errors of the spectrum width of Doppler velocity fluctuations within the components of weather formations.</p>

错误:搜索内容不能为空,请输入英文关键词
错误:关键词超出字数限制,请精简
高级检索

Estimation of Weather Signal Parameters Based on Eigenvectors and Eigenvalues of Correlation Matrix of Input Influences

  • D. V. Atamanskiy,
  • V. V. Vasylyshyn,
  • R. L. Stovba,
  • L. V. Prokopenko,
  • I. V. Krasnoshapka

摘要

Abstract

The safety of aircraft flights largely depends on the quality of measurements of weather formation parameters using the pulsed Doppler weather radars. This paper discusses the features of estimating the width of the Doppler spectrum of velocities of optically unobservable weather formations based on the paired-pulse method. A characteristic feature of such weather phenomena is the low power of radar reflectivity. It is shown that the estimation errors depend on the power ratio of the weather formations themselves and the receiver internal noise. A method for eliminating this dependence is proposed. It is based on representing the correlation matrix of input influences in terms of its eigenvectors, dividing the matrix into signal and noise components, and subsequently reducing the noise component’s influence. The effectiveness of the proposed method is illustrated using simulation modeling results. Analytical expressions have been derived for the distribution density of the absolute and relative estimation errors of the spectrum width of Doppler velocity fluctuations within the components of weather formations.