Radar data denoising systems and methods

1. A method comprising:

receiving radar data;

beamforming the radar data to obtain beamformed radar data associated with a first plurality of channels;

performing a first transform associated with the beamformed radar data to obtain transformed radar data, wherein the transformed radar data is associated with a location parameter and a variance that is independent of the location parameter;

performing a second transform of the transformed radar data to obtain dimensionality-reduced radar data associated with a second plurality of channels, wherein the number of channels in the second plurality of channels is less than the number of channels in the first plurality of channels, and wherein each channel of the first and second plurality of channels is associated with a respective angle; and

filtering, by a multichannel denoising filter, the dimensionality-reduced radar data to obtain denoised dimensionality-reduced radar data.

2. The method of claim 1 , further comprising:

performing a third transform of the denoised dimensionality-reduced radar data to obtain denoised transformed radar data, wherein the third transform comprises an inverse transform of the second transform; and

performing a fourth transform of the denoised transformed radar data to obtain denoised radar data, wherein the fourth transform comprises an inverse transform of the first transform.

3. The method of claim 1 , further comprising determining a magnitude of the beamformed radar data to obtain magnitude images, wherein the first transform is associated with the magnitude images.

4. The method of claim 3 , wherein the magnitude images have a Rician distribution, and wherein the first transform comprises a variance-stabilizing transformation for the Rician distribution.

5. The method of claim 1 , wherein the first transform comprises a variance-stabilizing transformation.

6. The method of claim 5 , wherein the radar data has a variance that depends on a location parameter of the radar data.

7. The method of claim 1 , wherein the transformed radar data has unit variance.

8. The method of claim 1 , wherein the second transform comprises a singular value decomposition.

9. The method of claim 1 , further comprising determining a respective power spectral density of each channel of the second plurality of channels, wherein the filtering is based on the power spectral density of each channel.

10. The method of claim 1 , wherein the filtering comprises:

performing block matching on a first channel of the second plurality of channels to obtain a set of block coordinates; and

denoising each of the second plurality of channels based on the set of block coordinates.

11. The method of claim 1 , wherein the beamformed radar data is associated with a scale parameter indicative of a tensor product of a range-wise standard deviation and a channel-wise standard deviation.

12. The method of claim 1 , wherein the first transform is based on a range-wise standard deviation and a channel-wise standard deviation.

13. A system comprising:

a receiver configured to receive radar data;

a processor coupled to the receiver; and

a non-transitory machine-readable medium comprising instructions stored therein, which when executed by the processor, cause the processor to perform operations comprising:

beamforming the radar data to obtain beamformed radar data associated with a first plurality of channels;

performing a first transform associated with the beamformed radar data to obtain transformed radar data, wherein the transformed radar data is associated with a location parameter and a variance that is independent of the location parameter;

performing a second transform of the transformed radar data to obtain dimensionality-reduced radar data associated with a second plurality of channels, wherein the number of channels in the second plurality of channels is less than the number of channels in the first plurality of channels, and wherein each channel of the first and second plurality of channels is associated with a respective angle;

determining a respective power spectral density of each channel of the second plurality of channels; and

filtering the dimensionality-reduced radar data based on the power spectral density of each channel to obtain denoised dimensionality-reduced radar data.

14. The system of claim 13 , wherein the operations further comprise:

performing a third transform of the denoised dimensionality-reduced radar data to obtain denoised transformed radar data, wherein the third transform comprises an inverse transform of the second transform; and

performing a fourth transform of the denoised transformed radar data to obtain denoised radar data, wherein the fourth transform comprises an inverse transform of the first transform.

15. The system of claim 13 , wherein the operations further comprise determining a magnitude of the beamformed radar data to obtain magnitude images, and wherein the first transform is associated with the magnitude images.

16. The system of claim 15 , wherein the magnitude images have a Rician distribution, and wherein the first transform comprises a variance-stabilizing transformation for the Rician distribution.

17. The system of claim 13 , wherein the first transform comprises a variance-stabilizing transformation.

18. The system of claim 17 , wherein the radar data has a variance that depends on a location parameter of the radar data.

19. The system of claim 13 , wherein the transformed radar data has unit variance.

20. The system of claim 13 , wherein the second transform comprises a singular value decomposition.