RFI mitigation in weather radar data

1. A method to mitigate radio frequency interference (RFI) in weather radar data, comprising:

using a radar antenna to generate weather radar data;

computing p norms of radials of the weather radar data to construct an p norm profile of the weather radar data as a function of azimuth angle, the weather radar data comprising Level 2 or higher weather radar data in polar format, wherein each p norm comprises an 0 norm, a Manhattan Distance, or a Euclidian Norm;

determining that a given radial in the weather radar data is an RFI radial based on the p norm profile of the weather radar data; and

displaying an image from the weather radar data in which at least one of:

the RFI radial is identified in the image as including RFI; or

the RFI radial is omitted from the image.

2. The method of claim 1 , further comprising deleting data points of the RFI radial from the weather radar data before displaying the image to cause the RFI radial to be omitted from the image when the image is displayed.

3. The method of claim 2 , further comprising imputing at least some replacement data points for the deleted data points from data points of two or more non-RFI radials in the weather radar data that are near a location of the RFI radial.

4. The method of claim 3 , wherein imputing at least some replacement data points for the deleted data points from data points of two or more non-RFI radials in the weather radar data that are near the location of the RFI radial comprises interpolating from each side of the RFI radial.

5. The method of claim 1 , further comprising, prior to computing the p norms, applying a threshold to the weather radar data to suppress noise and precipitation data.

6. The method of claim 5 , wherein applying the threshold to the weather radar data comprises temporarily disregarding each data point that has a magnitude below the threshold for computation of the p norms.

7. The method of claim 1 , further comprising determining that the given radial includes the RFI radial based on a consensus of first and second data products included in the weather radar data.

8. The method of claim 7 , wherein:

computing the p norms of the radials of the weather radar data to construct the p norm profile comprises computing first p norms of radials of the first data product to construct a first p norm profile;

the method further comprises:

computing second p norms of radials of the second data product to construct a second p norm profile; and

averaging the first p norm profile and the second p norm profile to form an average p norm profile; and

determining that the given radial is the RFI radial based on the consensus of the first and second data products comprises determining that the given radial is the RFI radial based on the average p norm profile.

9. The method of claim 1 , wherein determining that the given radial in the weather radar data is the RFI radial based on the p norm profile of the weather radar data includes determining that a convolution of a derivative of the p norm profile with a RFI kernel exceeds a threshold at an azimuth angle of the given radial.

10. The method of claim 9 , further comprising, prior to determining that the convolution exceeds the threshold:

computing the derivative of the p norm profile with respect to azimuth angle;

obtaining the RFI kernel; and

computing the convolution;

wherein obtaining the RFI kernel comprises one of:

computing a template RFI kernel to use as the RFI kernel, the template RFI kernel including an approximation of typical radials in an azimuthal neighborhood of RFI;

deriving a wavelet RFI kernel to use as the RFI kernel, the wavelet RFI kernel derived from data that represents an actual RFI radial and neighboring radials; or

deriving an average wavelet RFI kernel to use as the RFI kernel, the average wavelet RFI kernel derived from data that represents multiple actual RFI radials and respective neighboring radials.

11. The method of claim 1 , further comprising, prior to displaying the image of the weather radar data, despeckling and masking the weather radar data to remove noise, wherein displaying the image of the weather radar data comprises displaying the image of the despeckled and masked weather radar data.

12. A non-transitory computer-readable medium having computer-readable instructions stored thereon that are executable by a processor to perform or control performance of operations comprising:

using a radar antenna to generate weather radar data;

computing p norms of radials of the weather radar data to construct an p norm profile of the weather radar data as a function of azimuth angle, the weather radar data comprising Level 2 or higher weather radar data in polar format, wherein each p norm comprises an 0 norm, a Manhattan Distance, or a Euclidian Norm;

determining that a given radial in the weather radar data is a radio frequency interference (RFI) radial based on the p norm profile of the weather radar data; and

displaying an image from the weather radar data in which at least one of:

the RFI radial is identified in the image as including RFI; or

the RFI radial is omitted from the image.

13. The non-transitory computer-readable medium of claim 12 , the operations further comprising deleting data points of the RFI radial from the weather radar data before displaying the image to cause the RFI radial to be omitted from the image when the image is displayed.

14. The non-transitory computer-readable medium of claim 13 , the operations further comprising imputing at least some replacement data points for the deleted data points from data points of two or more non-RFI radials in the weather radar data that are near a location of the RFI radial.

15. The non-transitory computer-readable medium of claim 12 , the operations further comprising, prior to computing the p norms, applying a threshold to the weather radar data to suppress noise and precipitation data.

16. The non-transitory computer-readable medium of claim 12 , the operations further comprising determining that the given radial is the RFI radial based on a consensus of first and second data products included in the weather radar data.

17. The non-transitory computer-readable medium of claim 16 , wherein:

computing the p norms of the radials of the weather radar data to construct the p norm profile comprises computing first p norms of radials of the first data product to construct a first p norm profile;

the operations further comprise:

computing second p norms of radials of the second data product to construct a second p norm profile; and

averaging the first p norm profile and the second p norm profile to form an average p norm profile; and

determining that the given radial is the RFI radial based on the consensus of the first and second data products comprises determining that the given radial is the RFI radial based on the average p norm profile.

18. The non-transitory computer-readable medium of claim 12 , wherein determining that the given radial in the weather radar data includes the RFI radial based on the p norm profile of the weather radar data includes determining that a convolution of a derivative of the p norm profile with a RFI kernel exceeds a threshold at an azimuth angle of the given radial.

19. The non-transitory computer-readable medium of claim 18 , the operations further comprising, prior to determining that the convolution exceeds the threshold:

computing the derivative of the p norm profile with respect to azimuth angle;

obtaining the RFI kernel; and

computing the convolution;

wherein obtaining the RFI kernel comprises one of:

computing a template RFI kernel to use as the RFI kernel, the template RFI kernel including an approximation of typical radials in an azimuthal neighborhood of RFI;

deriving a wavelet RFI kernel to use as the RFI kernel, the wavelet RFI kernel derived from data that represents an actual RFI radial and neighboring radials; or

deriving an average wavelet RFI kernel to use as the RFI kernel, the average wavelet RFI kernel derived from data that represents multiple actual RFI radials and respective neighboring radials.

20. A method, comprising:

using a radar antenna to generate weather radar data;

computing p norms of radials of the weather radar data to construct an p norm profile of the weather radar data as a function of azimuth angle, the weather radar data comprising Level 2 or higher weather radar data in polar format, wherein:

each p norm comprises an p norm, a Manhattan Distance, or a Euclidian Norm;

the weather radar data in polar format includes, for each data point, a magnitude of the corresponding data point and a location of the corresponding data point, the location specified by a radial distance and azimuth angle; and

each radial includes all data points that have an azimuth angle within a given range of azimuth angles across a radial distance range for each azimuth angle within the given range of azimuth angles;

computing a derivative of the p norm profile with respect to azimuth angle;

obtaining a radio frequency interference (RFI) kernel, wherein obtaining the RFI kernel comprises one of:

computing a template RFI kernel to use as the RFI kernel, the template RFI kernel including an approximation of typical radials in an azimuthal neighborhood of RFI;

deriving a wavelet RFI kernel to use as the RFI kernel, the wavelet RFI kernel derived from data that represents an actual RFI radial and neighboring radials; or

deriving an average wavelet RFI kernel to use as the RFI kernel, the average wavelet RFI kernel derived from data that represents multiple actual RFI radials and respective neighboring radials;

computing a convolution of the derivative of the p norm profile with the RFI kernel;

determining that a given radial in the weather radar data includes an RFI radial based on the p norm profile of the weather radar data, including determining that the convolution exceeds a threshold at an azimuth angle of the given radial; and

displaying an image from the weather radar data in which at least one of:

the RFI radial is identified in the image as including RFI; or

the RFI radial is omitted from the image.