Spectral Estimation of a Narrow-Band Gaussian Process from the Distribution of the Distance Between Adjacent Zeros

1980 ◽  
Vol BME-27 (2) ◽  
pp. 108-110 ◽  
Author(s):  
Bjorn A. J. Angelsen
1986 ◽  
Vol 30 (02) ◽  
pp. 123-126
Author(s):  
A. E. Mansour

Introduction and background - The probability distribution of the peak process of a stationary random process with zero mean was first determined by Rice [1]. 2 Following his basic derivation, Longuet-Higgins [2] and Cartwright and Longuet-Higgins [3] evaluated various wave statistics, first for a narrow-band Gaussian process, then extended the results for a Gaussian process of any spectral width.


2018 ◽  
Vol 11 (3) ◽  
pp. 215-219
Author(s):  
C. F. Hu ◽  
N. J. Li

AbstractThe measurement accuracy of low-frequency narrow-band antenna is heavily influenced by its environment, which is also difficult to remove the clutter with a time gating. This paper proposes a method to improve the measurement accuracy of low-frequency narrow-band antenna using signal processing technique. The method is to predict the unknown value out of received original signal with an auto-regressive model (AR model) based on modern spectral estimation theory, and the parameters in AR model are calculated by maximum entropy spectral estimation algorithm. Thus, a wideband signal compared with the original band is obtained, and then the time-domain resolution is enhanced. The time gating is more exactly to separate the antenna radiation signal from multipath signals. The simulation and experimental results show that about 50% extended data for each ends of original band can be obtained after spectral extrapolation, and the time-domain resolution after extrapolation is twice than the original narrow-band signal, and the influence of measurement environment can be eliminated effectively. The method can be used to improve accuracy in actual antenna measurement.


Sign in / Sign up

Export Citation Format

Share Document