System and method for interference detection in a RF receiver

1. A method of detecting RF interference, comprising the steps of:

receiving an RF signal detected at a receiver, the received RF signal including a desired RF signal and potentially an intermittent interference signal occupying an interference bandwidth;

estimating thermal noise of the receiver, at an interference detection controller and/or processor, by statistically analyzing a plurality of time intervals of data of the received RF signal, including at least one time interval not including the intermittent interference signal;

estimating an intermittent-interference-plus-noise level, at an interference detection controller and/or processor, by statistically analyzing an extended time interval of the time intervals of data;

determining an interference metric, at an interference detection controller and/or processor, based on a ratio of the estimated intermittent-interference-plus-noise level to the estimated thermal noise; and

evaluating the interference metric, at an interference detection controller and/or processor, against one or more thresholds to detect the presence or absence of degrading RF interference.

2. The method of claim 1 , wherein statistically analyzing comprises application of order statistic filtering.

3. The method of claim 1 , wherein:

estimating the thermal noise of the receiver further comprises

obtaining a frequency domain representation of the plurality of time intervals, the frequency domain representation including a magnitude level for each of a plurality of frequencies sorted in an order statistic distribution,

determining a value associated with a thermal-noise reference percentile relative to the distribution as a raw thermal noise estimate, and

conditioning the raw thermal noise estimate to compensate for estimation bias in order to obtain the thermal noise estimate; and

estimating the intermittent-interference-plus-noise level further comprises

obtaining a frequency domain representation of the extended time interval, the frequency domain representation including a magnitude level for each of a plurality of frequencies sorted in an order statistic distribution,

determining a value associated with the reference percentile relative to the distribution as a raw intermittent-interference-plus-noise level estimate, and

conditioning the raw intermittent-interference-plus-noise level estimate to compensate for estimation bias to obtain the intermittent-interference-plus-noise level estimate.

4. The method of claim 3 , further comprising discarding a selected number of samples of the extended time interval prior to obtaining the frequency domain representation.

5. The method of claim 3 , wherein the thermal-noise reference percentile comprises a standard deviation percentile along a Rayleigh distribution selected to be lower than representations of RF object reflections and degrading interference in the Rayleigh distribution.

6. The method of claim 3 , wherein conditioning further comprises eliminating outlier and averaging remaining thermal noise or intermittent-interference-plus-noise level estimates, respectively, over several sampling cycles.

7. The method of claim 3 , wherein obtaining the frequency domain representations further comprise respectively reducing sidelobe energies of the frequency domain representations through application of a window approximation.

8. The method of claim 7 , further comprising respectively normalizing to compensate for time domain attenuation underestimations of the thermal noise and intermittent-interference-plus-noise level estimates resulting from the window approximation.

9. The method of claim 3 , wherein:

obtaining the frequency domain representations further comprises applying a FFT of respective lengths;

the FFT length associated with the thermal noise estimation is a fraction of the number of time intervals in the plurality of time intervals; and

the FFT length associated with the intermittent-interference-plus-noise level estimation is equal to a number of time domain samples of the extended time interval.

10. The method of claim 9 , further comprising normalizing to compensate for scaling changes in the thermal noise and intermittent-interference-plus-noise level estimates resulting from use of distinct FFT length.

11. The method of claim 1 , further comprising receiving one or more parameters specifying at least one of the number of time intervals in the plurality, a number of time intervals to be discarded prior to estimating the thermal noise, an FFT length to be used in estimating the thermal noise, an FFT length to be used in estimating the interference plus noise level, and a percentile for identifying a reference percentile relative to an order statistic distribution of the spectral magnitude data.

12. The method of claim 1 , further comprising issuing an interference alert or invoking an interference suppression or avoidance application if the interference metric exceeds an alert threshold for a predetermined number of sampling cycles.

13. The method of claim 1 , wherein the desired RF signal comprises an automotive radar signal and the receiver comprises an automotive radar receiver.

14. An interference detector for use with an RF receiver configured to receive an RF signal including a desired RF signal and potentially an intermittent interference signal occupying an interference bandwidth, comprising;

an interference detection controller and/or processor configured to:

estimate thermal noise of the receiver by statistically analyzing a plurality of time intervals of data of the received RF signal, including at least one time interval not including the intermittent interference signal;

estimate an intermittent-interference-plus-noise level by statistically analyzing an extended time interval of the at least one time intervals of data;

determine an interference metric based on a ratio of the estimated intermittent-interference-plus-noise level to thermal noise; and

evaluate the interference metric against one or more thresholds to detect the presence or absence of degrading RF interference.

15. The detector of claim 14 , wherein statistically analyzing comprises application of order statistic filtering.

16. The detector of claim 14 , wherein the controller and/or processor is further configured to estimate the thermal noise of the receiver by:

obtaining a frequency domain representation of the plurality of time intervals, the frequency domain representation including a magnitude level for each of a plurality of frequencies sorted in an order statistic distribution,

determining a value associated with a thermal-noise reference percentile relative to the distribution as a raw thermal noise estimate, and

conditioning the raw thermal noise estimate to compensate for estimation bias in order to obtain the thermal noise estimate; and

estimating the intermittent-interference-plus-noise level further comprises

obtaining a frequency domain representation of the extended time interval, the frequency domain representation including a magnitude level for each of a plurality of frequencies sorted in an order statistic distribution,

determining a value associated with the reference percentile relative to the distribution as a raw intermittent-interference-plus-noise level estimate, and

conditioning the raw intermittent-interference-plus-noise level estimate to compensate for estimation bias to obtain the intermittent-interference-plus-noise level estimate.

17. The detector of claim 16 , wherein the thermal-noise reference percentile comprises a standard deviation percentile along the distribution selected to be lower than representations of RF object reflections and degrading interference in the distribution.

18. The detector of claim 16 , wherein the controller and/or processor is further configured to obtain the frequency domain representations by:

respectively reducing sidelobe energies of the frequency domain representations through application of a window approximation; and

respectively normalizing to compensate for time domain attenuation underestimations of the thermal noise and intermittent-interference-plus-noise level estimates resulting from a window approximation.

19. The detector of claim 16 , wherein:

the controller and/or processor is further configured to obtain the frequency domain representations by applying a FFT of respective lengths;

the FFT length associated with the thermal noise estimation is a fraction of the number of time intervals in the plurality of time intervals; and

the FFT length associated with the intermittent-interference-plus-noise level estimation is equal to a number of time domain samples of the extended time interval.

20. The detector of claim 14 , wherein the controller and/or processor is further configured to issue an interference alert or invoke an interference suppression or avoidance application if the interference metric exceeds an alert threshold for a predetermined number of sampling cycles.

21. The detector of claim 20 , wherein the controller and/or processor is further configured to cease an alert or an invoked interference application if the interference metric falls below an alert off threshold.

22. The detector of claim 14 , wherein the RF receiver comprises an automotive radar receiver.

23. An RF receiver configured to detect RF interference, comprising:

a front end configured to receive an RF signal including a desired RF signal and potentially an intermittent interference signal occupying an interference bandwidth; and

an interference detection controller and/or processor configured to:

estimate thermal noise of the receiver by statistically analyzing a plurality of time intervals of data of the received RF signal, including at least one time interval not including the intermittent interference signal;

estimate an intermittent-interference-plus-noise level by statistically analyzing an extended time interval of the time intervals of data;

determine an interference metric based on a ratio of the estimated intermittent-interference-plus-noise level to thermal noise; and

evaluate the interference metric against one or more thresholds to detect the presence or absence of degrading RF interference.