Statistical averaging method for wind profiler doppler spectra

Claim 1

A method for Doppler spectral averaging, the method comprising: (a) collecting a statistical sample set for each Doppler velocity, antenna beam, and radar range using radar Doppler spectral values sampled over time; (b) detecting multiple signal sources in said radar Doppler spectral values by using differences in statistical and physical properties of said radar Doppler spectral values; (c) separating signals from said multiple signal sources over time, over Doppler velocity, and over radar range; (d) identifying contamination in said radar Doppler spectral values using said statistical and physical properties; (e) eliminating said contamination, leaving atmospheric signal and radar system noise when these are observable; (f) identifying said atmospheric signal based upon said statistical properties and physical properties; (g) estimating an average spectrum using a subset of said samples identified with said atmospheric signal and with said radar system noise; and (h) indicating a presence of useful signal from said radar Doppler spectral values in accordance with said average spectrum.

Claim 2

A method for Doppler spectral averaging, the method comprising: (a) collecting a statistical sample set for each Doppler velocity, antenna beam, and radar range using radar Doppler spectral values sampled over time; (b) detecting multiple signal sources in said radar Doppler spectral values by using differences in statistical and physical properties of said radar Doppler spectral values; (c) separating signals from said multiple signal sources over time, over Doppler velocity, and over radar range; (d) identifying contamination in said radar Doppler spectral values using said statistical and physical properties; (e) eliminating said contamination, leaving atmospheric signal and radar system noise when these are observable; (f) identifying said atmospheric signal based upon said statistical properties and physical properties; (g) estimating an average spectrum using a subset of said samples identified with said atmospheric signal and with said radar system noise; and (h) indicating a presence of useful signal from said radar Doppler spectral values in accordance with said average spectrum.

Claim 5

A method for radar Doppler spectral processing, the method comprising: (a) collecting a statistical sample set for each Doppler velocity, antenna beam, and radar range using radar Doppler spectral values sampled over time, wherein: (i) sampling is done over an observation time period for which atmospheric wind measurements are assumed to be stationary but for which the atmospheric return strength has random variations; and (ii) some properties of contamination behave with time over the observation time period differently from behavior of the atmospheric return strength; (b) detecting multiple signal sources using differences in statistical and physical properties of those signals sources, wherein: (i) a spectral power of an atmospheric signal has an exponential statistical distribution and is generally very weak for clear-air return; and (ii) a spectral power of contamination has different statistical distributions and/or relative strength from those of said atmospheric signal; (c) a separating of the signals from said different sources over time, over Doppler velocity, over antenna beam, and over radar range; (d) identifying said contamination using said statistical and physical properties; (e) eliminating said contamination, leaving said atmospheric signal and radar system noise, by setting an acceptance threshold for level of contamination and using said threshold to remove data samples identified as unacceptable from further signal processing; (f) identifying said atmospheric signal based upon said statistical properties and physical properties; (g) estimating an average spectrum using a subset of samples identified with said atmospheric signal and with said radar system noise; and (h) indicating a presence of a useful signal from the atmospheric return.

Claim 6

A method for radar Doppler spectral processing, the method comprising: (a) collecting a statistical sample set for each Doppler velocity, antenna beam, and radar range using radar Doppler spectral values sampled over time, wherein: (i) sampling is done over an observation time period for which atmospheric wind measurements are assumed to be stationary but for which the atmospheric return strength has random variations; and (ii) some properties of contamination behave with time over the observation time period differently from behavior of the atmospheric return strength; (b) detecting multiple signal sources using differences in statistical and physical properties of those signals sources, wherein: (i) a spectral power of an atmospheric signal has an exponential statistical distribution and is generally very weak for clear-air return; and (ii) a spectral power of contamination has different statistical distributions and/or relative strength from those of said atmospheric signal; (c) a separating of the signals from said different sources over time, over Doppler velocity, over antenna beam, and over radar range; (d) identifying said contamination using said statistical and physical properties; (e) eliminating said contamination, leaving said atmospheric signal and radar system noise, by setting an acceptance threshold for level of contamination and using said threshold to remove data samples identified as unacceptable from further signal processing; (f) identifying said atmospheric signal based upon said statistical properties and physical properties; (g) estimating an average spectrum using a subset of samples identified with said atmospheric signal and with said radar system noise; and (h) indicating a presence of a useful signal from the atmospheric return.