H.O. Boyo (Japan/Nigeria), M. Fujiwara (Japan), V.G. Moshnyaga, and A. Boyo (Nigeria)
Signal processing, noise localization, time-frequency analysis,DSP application
The difficulties in extracting useful information from the noise-corrupted stratospheric laser echoes have elevated application of digital signal processing techniques to the atmospheric research. This paper focuses on the problem of noise elimination from the transient returns of stratospheric laser echoes, proposes a new solution approach to it. Unlike related methods which address signal modeling in time domain only, we formulate the problem in both time and frequency domain, present a new modeling environment for the joint time-frequency analysis and use a non-linear median filter for noise elimination. Experiments with various patterns of spectral clouds demonstrate that such an approach significantly increases readability and accuracy of results in terms of the back-scattering function and depolarization.
Important Links:
Go Back
