Techniques to disambiguate angle of arrival

1. A method for determining an angle of arrival for a signal received by a mobile device with a plurality of antennas, the method comprising:

receiving, from a transmitting device, the signal by the plurality of antennas;

measuring one or more phase differences among the signal received at the plurality of antennas;

determining a first set of possible values for the angle of arrival that are consistent with the one or more phase differences;

measuring one or more signal values using one or more sensors of the mobile device; for each of the first set of possible values, determining a confidence score based on the one or more signal values; and

selecting, based on the determined confidence score for each of the first set of possible values, one of the first set of possible values as the angle of arrival, wherein the selecting based on the confidence score for each of the first set of possible values comprises:

comparing the confidence score to a confirmation threshold, wherein a possible value with the confidence score above the confirmation threshold is in a confirmed state; and

selecting the possible value in the confirmed state as the angle of arrival, wherein for multiple possible values in the confirmed state, the possible value is selected if the possible value has a longer tracking time than other possible values in the confirmed state.

2. The method of claim 1 , wherein the one or more sensors are the plurality of antennas, the method further comprising:

determining, using a machine learning model and antenna measurements of the signal, a probability vector for the first set of possible values, each probability of the probability vector corresponding to a region of a plurality of regions around the mobile device, wherein the confidence score for a respective possible value is determined using the probability in the probability vector for the respective possible value.

3. The method of claim 2 , wherein each region of the plurality of regions is defined by an azimuth and an elevation with respect to a geometry of the mobile device.

4. The method of claim 1 , further comprising:

measuring a power level of the signal, the power level of the signal being one of the one or more signal values; and

comparing the power level of the signal to a threshold power level value, wherein the confidence score for at least one of plurality of possible values is determined based on the power level of the signal being below the threshold power level value.

5. The method of claim 1 , further comprising:

determining a motion of the mobile device over a period of time using the one or more signal values;

for each of the first set of possible values, predicting an angle of the signal at a first time based on the motion; and

determining a first angle of the signal at the first time based on one or more additional phase differences measured at the first time, wherein the confidence score for a respective possible value is determined based on a difference between the predicted angle and the first angle for the respective possible value.

6. The method of claim 1 , further comprising:

receiving a second signal by the plurality of antennas, wherein the second signal is transmitted over a second frequency;

measuring one or more second phase differences among the second signal received at the plurality of antennas, wherein the one or more second phase differences are the one or more signal values;

determining a second set of possible values for the angle of arrival that are consistent with the one or more second phase differences; and

for each of the first set of possible values, determining a difference between a respective possible value and one of the second set of possible values, wherein the confidence score for the respective possible value is determined based on the difference between the respective possible value and the one of the second set of possible values.

7. The method of claim 1 , wherein the angle of arrival is determined based in part on a first phase difference of arrival at a first antenna and a second phase difference of arrival at a second antenna.

8. The method of claim 1 , wherein the selecting based on the confidence score further comprises:

comparing the confidence score to a deletion threshold wherein a possible value with the confidence score below the deletion threshold is deleted.

9. The method of claim 1 , further comprising:

determining, using a machine learning model and antenna measurements of the signal, a probability vector for the first set of possible values, each probability of the probability vector corresponding to a region of a plurality of regions around the mobile device, wherein the confidence score for a respective possible value is determined using the probability in the probability vector for the respective possible value;

measuring a power level of the signal, the power level of the signal being one of the one or more signal values;

comparing the power level of the signal to a threshold power level value, wherein the confidence score for at least one of plurality of possible values is adjusted based on the power level of the signal being below the threshold power level value;

determining a motion of the mobile device over a period of time using the one or more signal values;

for each of the first set of possible values, predicting an angle of the signal at a first time based on the motion;

determining a first angle of the signal at the first time based on one or more additional phase differences measured at the first time, wherein the confidence score for a respective possible value is adjusted based on a difference between the predicted angle and the first angle for the respective possible value;

receiving a second signal by the plurality of antennas, wherein the second signal is transmitted over a second frequency;

measuring one or more second phase differences among the second signal received at the plurality of antennas, wherein the one or more second phase differences are the one or more signal values;

determining a second set of possible values for the angle of arrival that are consistent with the one or more second phase differences; and

for each of the first set of possible values, determining a difference between the respective possible value and one of the second set of possible values, wherein the confidence score for the respective possible value is adjusted based on the difference between the respective possible value and the one of the second set of possible values.

10. The method of claim 1 , further comprising:

determining, using a machine learning model and antenna measurements of the signal, a probability vector for the first set of possible values, each probability of the probability vector corresponding to a region of a plurality of regions around the mobile device, wherein the confidence score for a respective possible value is determined using the probability in the probability vector for the respective possible value;

measuring a power level of the signal, the power level of the signal being one of the one or more signal values;

comparing the power level of the signal to a threshold power level value, wherein the confidence score for at least one of plurality of possible values is adjusted based on the power level of the signal being below the threshold power level value;

determining a motion of the mobile device over a period of time using the one or more signal values;

for each of the first set of possible values, predicting an angle of the signal at a first time based on the motion;

determining a first angle of the signal at the first time based on one or more additional phase differences measured at the first time, wherein a first confidence score adjustment value for a respective possible value is determined based on a difference between the predicted angle and the first angle for the respective possible value;

determining a distance to an electronic device;

adjusting the first confidence score adjustment value based at least in part on the determined distance;

receiving a second signal by the plurality of antennas, wherein the second signal is transmitted over a second frequency;

measuring one or more second phase differences among the second signal received at the plurality of antennas, wherein the one or more second phase differences are the one or more signal values;

determining a second set of possible values for the angle of arrival that are consistent with the one or more second phase differences; and

for each of the first set of possible values, determining a difference between a respective possible value and one of the second set of possible values, wherein a second confidence score adjustment value for a respective possible value is determined based on the difference for the respective possible value;

determining a track score difference by adding the first confidence score adjustment value and the second confidence score adjustment value;

determining a final track score by adding an adjusted confidence score with the track score difference; and

determining the angle of arrival based at least in part on the final track score.

11. A method for determining an angle of arrival for a signal received by a mobile device with a plurality of antennas, the method comprising:

receiving, from a transmitting device, the signal by the plurality of antennas;

measuring one or more phase differences among the signal received at the plurality of antennas;

determining a first set of possible values for the angle of arrival that are consistent with the one or more phase differences;

determining, using a machine learning model and antenna measurements of the signal, a probability vector for the first set of possible values that are consistent with the one or more phase differences, each probability of the probability vector corresponding to a region of a plurality of regions around the mobile device;

determining a confidence score for a respective possible value using the probability in the probability vector for the respective possible value, wherein determining the confidence score comprises:

converting the one or more signal values, for each of a plurality of signal values, into a plurality of confidence values;

determining a delta track score comprising a sum of the plurality of confidence values; and

determining a final confidence score based on a track score and the delta track score;

measuring one or more signal values using one or more sensors of the mobile device;

determining a motion of the mobile device over a period of time using the one or more signal values;

for each of the first set of possible values, predicting an angle of the signal at a first time based on the motion; and

determining a first angle of the signal at the first time based on one or more additional phase differences measured at the first time, wherein the confidence score for the respective possible value is adjusted based on a difference between the predicted angle and the first angle for the respective possible value; and

selecting, based on the adjusted confidence score, one of the first set of possible values as the angle of arrival.

12. The method of claim 11 , further comprising:

receiving a second signal by the plurality of antennas, wherein the second signal is transmitted over a second frequency;

measuring one or more second phase differences among the second signal received at the plurality of antennas, wherein the one or more second phase differences are the one or more signal values;

determining a second set of possible values for the angle of arrival that are consistent with the one or more second phase differences; and

for each of the first set of possible values, determining a difference between a respective possible value and one of the second set of possible values, wherein the confidence score for a respective possible value is further determined based on the difference for the respective possible value.

13. The method of claim 11 , wherein the selecting based on the adjusted confidence score further comprises:

comparing the confidence score to a confirmation threshold and a deletion threshold wherein a possible value with the confidence score above the confirmation threshold is in a confirmed state and the possible value with the confidence score below the deletion threshold is deleted; and

selecting the possible value in the confirmed state as the angle of arrival wherein for multiple possible values in the confirmed state, the possible value is selected if the possible value has a longer tracking time than other possible values in the confirmed state.

14. A non-transitory computer-readable medium storing a plurality of instructions that, when executed by one or more processors of a mobile device with a plurality of antennas, cause the one or more processors to perform operations for determining an angle of arrival for a signal, comprising:

receiving, from a transmitting device, the signal by the plurality of antennas;

measuring one or more phase differences among the signal received at the plurality of antennas;

determining a first set of possible values for the angle of arrival that are consistent with the one or more phase differences;

measuring one or more signal values using one or more sensors of the mobile device;

for each of the first set of possible values, determining a confidence score based on the one or more signal values;

selecting, based on each determined confidence score for each of the first set of possible values, one of the first set of possible values as the angle of arrival;

receiving a second signal by the plurality of antennas, wherein the second signal is transmitted over a second frequency;

measuring one or more second phase differences among the second signal received at the plurality of antennas, wherein the one or more second phase differences are the one or more signal values;

determining a second set of possible values for the angle of arrival that are consistent with the one or more second phase differences; and

for each of the first set of possible values, determining a difference between a respective possible value and one of the second set of possible values,

wherein the confidence score for the respective possible value is determined based on the difference between the respective possible value and the one of the second set of possible values.

15. The non-transitory computer-readable medium of claim 14 , wherein the one or more sensors are the plurality of antennas, the operations further comprising:

determining, using a machine learning model and antenna measurements of the signal, a probability vector for the first set of possible values, each probability of the probability vector corresponding to a region of a plurality of regions around the mobile device, wherein the confidence score for a respective possible value is determined using the probability in the probability vector for the respective possible value.

16. The non-transitory computer-readable medium of claim 14 , the operations further comprising:

measuring a power level of the signal, the power level of the signal being one of the one or more signal values; and

comparing the power level of the signal to a threshold power level value, wherein the confidence score for at least one of plurality of possible values is determined based on the power level of the signal being below the threshold power level value.

17. The non-transitory computer-readable medium of claim 14 , further comprising:

determining a motion of the mobile device over a period of time using the one or more signal values;

for each of the first set of possible values, predicting an angle of the signal at a first time based on the motion; and

determining a first angle of the signal at the first time based on one or more additional phase differences measured at the first time,

wherein the confidence score for a respective possible value is determined based on a difference between the predicted angle and the first angle for the respective possible value.

18. The non-transitory computer-readable medium of claim 14 , the operations further comprising:

determining, using a machine learning model and antenna measurements of the signal, a probability vector for the first set of possible values, each probability of the probability vector corresponding to a region of a plurality of regions around the mobile device, wherein the confidence score for a respective possible value is determined using the probability in the probability vector for the respective possible value;

measuring a power level of the signal, the power level of the signal being one of the one or more signal values;

comparing the power level of the signal to a threshold power level value, wherein the confidence score for at least one of plurality of possible values is adjusted based on the power level of the signal being below the threshold power level value;

determining a motion of the mobile device over a period of time using the one or more signal values;

for each of the first set of possible values, predicting an angle of the signal at a first time based on the motion;

determining a first angle of the signal at the first time based on one or more additional phase differences measured at the first time, wherein the confidence score for a respective possible value is adjusted based on a difference between the predicted angle and the first angle for the respective possible value;

receiving a second signal by the plurality of antennas, wherein the second signal is transmitted over a second frequency;

measuring one or more second phase differences among the second signal received at the plurality of antennas, wherein the one or more second phase differences are the one or more signal values;

determining a second set of possible values for the angle of arrival that are consistent with the one or more second phase differences; and

for each of the first set of possible values, determining a difference between a respective possible value and one of the second set of possible values, wherein the confidence score for the respective possible value is determined based on the difference between the respective possible value and the one of the second set of possible values.