Radar system and method for receiving and analyzing radar signals

1. A radar system system comprising:

at least one receiving element configured to detect reflected radar signals reflected by an object in a field-of-view of the system, wherein the receiving element is configured to generate detected signals indicative of the reflected radar signals detected by the receiving element;

an auto-encoding device, embedded in the receiving element, configured to perform an auto-encoding operation on the detected signals such that the receiving element outputs the detected signals as encoded detected signals, the auto-encoding operation minimizing data loss while compressing the detected signals into the encoded detected signals.

2. The radar system according to claim 1 , further comprising:

a transfer device in communication with the receiving element; and

a controller in communication with the transfer device, wherein:

the transfer device configured to transfer the encoded detected signals from the receiving element to the controller, and

the controller comprises a decoding device configured to perform a decoding operation to decode the encoded detected signals.

3. The radar system according to claim 2 , wherein

the auto-encoding device is further configured to apply a first non-linear function to the detected signals to encode the detected signals, and

the decoding device is further configured to apply a second non-linear function for decoding the encoded detected signals.

4. The radar system according claim 3 , wherein

the first non-linear function is different from the second non-linear function.

5. The radar system according claim 3 , wherein

a first neural network derives parameters of the first non-linear function being applied in the auto-encoding operation, and

a second neural network derives parameters of the second non-linear function being applied in the decoding operation.

6. The radar system according claim 5 , wherein

the controller comprises a discriminator configured to evaluate the parameters of the first non-linear function and the second non-linear function as part of a training procedure of the auto-encoding device and the decoding device.

7. The radar system according claim 6 , wherein

the controller comprises at least one application algorithm for analyzing the detected signals; and

the discriminator is further configured to use the application algorithm during the training procedure to evaluate the parameters for the first non-linear function and the second non-linear function.

8. The radar system according claim 7 , wherein

the at least one application algorithm comprises at least one algorithm for landmark classification, pedestrian classification, or bicycle classification.

9. A receiving element for detecting reflected radar signals, the receiving element comprising:

an embedded encoding device configured to encode detected signals into encoded detected signals via an auto-encoding operation that minimizes data loss while compressing the detected signals into the encoded detected signals.

10. A method for receiving and analyzing reflected radar signals, the method comprising:

detecting, via at least one receiving element, the reflected radar signals reflected by an object in a field-of-view of a radar system;

encoding the detected signals via an auto-encoding device that is embedded in the receiving element, the encoding minimizing loss of the detected signals while compressing the detected signals into encoded detected signals;

transferring the encoded detected signals to a controller;

decoding the encoded detected signals into decoded radar signals via a decoding device of the controller; and

analyzing the decoded radar signals via at least one application algorithm comprised by the controller.

11. The method according to claim 10 , wherein

the at least one application algorithm comprises at least one algorithm for landmark classification, pedestrian classification, or bicycle classification.

12. The method according to claim 10 , wherein:

the encoding comprises applying a first non-linear function to the detected signals;

the decoding comprises applying a second non-linear function to the encoded detected signals; and

the first non-linear function is different from the second non-linear function.

13. The method according to claim 12 , wherein:

the method is used as a part of a training procedure:

applying the first non-linear function comprises deriving parameters of the first non-linear function using a first neural network; and

applying the second non-linear function comprises deriving parameters of the second non-linear function using a second neural network.

14. The method according to claim 13 , wherein

deriving the parameters of the first non-linear function and the second non-linear function is performed by a discriminator comprised by the controller.

15. The method according to claim 14 , further comprising

using the application algorithm during the training procedure.

16. The radar system according to claim 7 , wherein

the application algorithm is further configured to analyze the decoded radar signals after the radar system has been implemented in an automated vehicle.

17. The radar system according to claim 7 , wherein the discriminator is further configured to:

receive the detected signals and the decoded radar signals; and

apply the application algorithm to the detected signals and the decoded radar signals.

18. The radar system according to claim 17 , wherein the discriminator is further configured to:

generate a first output of the application algorithm based on the detected signals;

generate a second output of the application algorithm based on the decoded radar signals; and

compare the first output and the second output.

19. The radar system according to claim 18 , wherein the discriminator is further configured to, based on the comparison of the first output and the second output, update the parameters for the first non-linear function and the second non-linear function, the updated parameters resulting in minimizing deviation between the first output and the second output.

20. The method according to claim 15 , wherein deriving the parameters of the first non-linear function and the second non-linear function further comprises:

receiving, by the discriminator, other detected signals and other decoded detected signals;

applying the application algorithm to the other detected signals and the other decoded detected signals;

generating a first output of the application algorithm based on the other detected signals;

generating a second output of the application algorithm based on the other decoded detected signals;

comparing the first output and the second output; and

based on the comparing, revising the parameters for the first non-linear function and the parameters for the second non-linear function, the revised parameters resulting in minimizing deviation between the first output and the second output.