Estimating angle measurements for source tracking using a phased array system

1. A device comprising:

a transceiver to transmit to or receive, from another device, radio frequency (RF) signals via a plurality of antenna elements; and

a processor coupled to the transceiver, the processor to execute a multi-angle source-tracking tool configured to determine and store a set of angle estimation values of an angle between the plurality of antenna elements and an antenna of the other device, a set of confidence measurements, and at least one of an Area-of-Arrival (ARoA) value or an Area-of-Departure (ARoD) value based on the RF signals, wherein each of the set of confidence measurements indicates a confidence of an angle estimation value of the set of angle estimation values.

2. The device of claim 1 , wherein the processor, to determine the set of angle estimation values and the set of confidence measurements, is to:

perform a source-tracking algorithm to obtain raw angle values based on digital samples of the RF signals, wherein the source-tracking algorithm comprises a cost function to weight the raw angle values to improve an accuracy of the source-tracking algorithm;

statistically model the raw angle values with a Gaussian Mixture Model (GMM) to obtain a plurality of mixtures, each mixture of the plurality of mixtures comprising GMM parameters including a weight, a mean, and a variance;

estimate an expected angle value for each of the plurality of mixtures from the GMM parameters to obtain the set of angle estimation values of the angle between the plurality of antenna elements and the antenna; and

calculate a confidence measurement for each of the set of angle estimation values of the plurality of mixtures from the GMM parameters to obtain the set of confidence measurements.

3. The device of claim 2 , wherein the processor is to determine the at least one of the ARoA value or the ARoD value from the set of angle estimation values and the set of confidence measurements, wherein the at least one of the ARoA value or the ARoD value represents an angled area between the plurality of antenna elements and the antenna, and wherein an angle estimation value of the set of angle estimation values is at least one of an Angle-of-Arrival (AoA) value, an Angle-of-Departure (AoD) value, a Time-Delay-of-Arrival (TDoA) value, a Time-Delay-of-Departure (TDoD) value, a Direction-of-Arrival (DoA) value, or a Direction-of-Departure (DoD) value.

4. The device of claim 2 , wherein the processor is further to determine an overall confidence measurement from the set of confidence measurements.

5. The device of claim 2 , wherein the source-tracking algorithm comprises a Generalized Cross Correlation (GCC) based algorithm that is modified by the cost function.

6. The device of claim 2 , wherein the source-tracking algorithm comprises a Generalized Cross Correlation with Phase Transform (GCC-PHAT) algorithm that is modified by the cost function.

7. The device of claim 2 , wherein the source-tracking algorithm comprises a MUltiple SIgnal Classification (MUSIC) algorithm that is modified by the cost function.

8. A System on Chip (SoC) comprising:

a transceiver configured to couple to a plurality of antenna elements to transmit to or receive from an antenna positioned at an angle relative to the plurality of antenna elements, radio frequency (RF) signals;

a processor coupled to the transceiver to process digital values representing the RF signals, wherein the processor is to:

obtain a raw angle sample;

update a statistical model with the raw angle sample;

identify one or more angle scenarios from the statistical model using a statistical classifier;

determine an estimated angle value for each of the one or more angle scenarios from the statistical model;

calculate a confidence value for the estimated angle value for each of the one or more angle scenarios from the statistical model; and

when the confidence value for the estimated angle value for the one or more angle scenarios is less than a confidence threshold,

estimate a number of multipaths between the plurality of antenna elements and the antenna from the one or more angle scenarios,

estimate an angled area between the plurality of antenna elements and the antenna from the estimated angle values and the corresponding confidence values to obtain an estimated angled area value, and

report the number of multipaths and the estimated angled area value.

9. The SoC of claim 8 , wherein the processor is further to report the estimated angle value when the confidence value for the estimated angle value for the one or more angle scenarios is equal to or greater than the confidence threshold.

10. The SoC of claim 8 , wherein the processor is further to:

obtain a second raw angle sample; and

update the statistical model with the second raw angle sample.

11. The SoC of claim 8 , wherein, to obtain the raw angle sample, the processor is to perform a source-tracking algorithm to obtain the raw angle sample, wherein the source-tracking algorithm comprises a cost function to weight the raw angle sample to improve an accuracy of the source-tracking algorithm, wherein, to update the statistical model with the raw angle sample and to identify the one or more angle scenarios from the statistical model using the statistical classifier, the processor is to update a Gaussian Mixture Model (GMM) to obtain a plurality of mixtures, each mixture of the plurality of mixtures comprising GMM parameters including a weight, a mean, and a variance.

12. The SoC of claim 11 , wherein, to determine the estimated angle value for each of the one or more angle scenarios from the statistical model, the processor is to estimate an expected angle value for each of the plurality of mixtures from the GMM parameters to obtain a set of angle estimation values corresponding to the number of multipaths between the plurality of antenna elements and the antenna, and wherein, to calculate the confidence value for the estimated angle value for each of the one or more angle scenarios from the statistical model, the processor is to calculate a confidence measurement for each of the plurality of mixtures from the GMM parameters to obtain a set of confidence measurements corresponding to the set of angle estimation values.

13. The SoC of claim 8 , wherein the estimated angle value is at least one of an Angle-of-Arrival (AoA) value, an Angle-of-Departure (AoD) value, a Time-Delay-of-Arrival (TDoA) value, a Time-Delay-of-Departure (TDoD) value, a Direction-of-Arrival (DoA) value, or a Direction-of-Departure (DoD) value, and wherein the estimated angled area value is at least one of an Area-of-Arrival (ARoA) value or an Area-of-Departure (ARoD) value.

14. A method comprising:

obtaining, by a processing device, raw angle samples from radio frequency (RF) signals transmitted or received by a plurality of antenna elements, wherein the RF signals are transmitted to or received from an antenna located at an angle relative to the plurality of antenna elements;

updating, by the processing device, a statistical model with the raw angle samples;

identifying, by the processing device, one or more angle scenarios from the statistical model using a statistical classifier;

determining, by the processing device, an estimated angle value for each of the one or more angle scenarios from the statistical model;

estimating, by the processing device, a confidence value for the estimated angle value for each of the one or more angle scenarios from the statistical model; and

when the confidence value for the estimated angle value for the one or more angle scenarios is less than a confidence threshold,

estimating, by the processing device, a number of multipaths between the plurality of antenna elements and the antenna from the one or more angle scenarios,

estimating, by the processing device, an angled area between the plurality of antenna elements and the antenna from the estimated angle values and the corresponding confidence values to obtain an estimated angled area value, and

reporting, by the processing device, the number of multipaths and the estimated angled area value.

15. The method of claim 14 , further comprising reporting, by the processing device, the estimated angle value when the confidence value for the estimated angle value for the one or more angle scenarios is equal to or greater than a confidence threshold.

16. The method of claim 14 , further comprising:

obtaining, by the processing device, a second raw angle sample; and

updating, by the processing device, the statistical model with the second raw angle sample.

17. The method of claim 14 , wherein the obtaining the raw angle samples comprises performing, by the processing device, a source-tracking algorithm to obtain the raw angle samples, wherein the source-tracking algorithm comprises a cost function to weight the raw angle sample to improve an accuracy of the source-tracking algorithm, wherein the updating the statistical model and the identifying the one or more angle scenarios comprises updating a Gaussian Mixture Model (GMM) to obtain a plurality of mixtures, each mixture of the plurality of mixtures comprising GMM parameters including a weight, a mean, and a variance.

18. The method of claim 14 , wherein the obtaining the raw angle samples comprises performing, by the processing device, a source-tracking algorithm to obtain the raw angle samples, wherein the source-tracking algorithm comprises at least one of:

a Generalized Cross Correlation (GCC) based algorithm that is modified by a cost function to weight the raw angle sample;

a Generalized Cross Correlation with Phase Transform (GCC-PHAT) algorithm that is modified by the cost function; or

a MUltiple SIgnal Classification (MUSIC) algorithm that is modified by the cost function.

19. The method of claim 14 , wherein the determining the estimated angle value comprises estimating, by the processing device, an expected angle value for each of a plurality of mixtures of a Gaussian Mixture Model (GMM) from GMM parameters to obtain a set of angle estimation values corresponding to the number of multipaths between the plurality of antenna elements and the antenna, and wherein the estimating the confidence value comprises estimating a confidence measurement for each of the plurality of mixtures from the GMM parameters to obtain a set of confidence measurements corresponding to the set of angle estimation values.

20. The method of claim 14 , wherein the estimated angle value is at least one of an Angle-of-Arrival (AoA) value, an Angle-of-Departure (AoD) value, a Time-Delay-of-Arrival (TDoA) value, a Time-Delay-of-Departure (TDoD) value, a Direction-of-Arrival (DoA) value, or a Direction-of-Departure (DoD) value, and wherein the estimated angled area value is at least one of an Area-of-Arrival (ARoA) value or an Area-of-Departure (ARoD) value.