Three-dimensional angle of arrival capability in electronic devices

1. An electronic device, comprising:

a processor configured to:

obtain signal information based on wireless signals received from an external electronic device via a first antenna pair and a second antenna pair, wherein the first and second antenna pairs are aligned along different axes and wherein the signal information includes channel information, range information, a first angle of arrival (AoA) of the wireless signals based on the first antenna pair, and a second AoA of the wireless signals based on the second antenna pair;

generate an initial prediction of a presence of the external electronic device relative to a field of view (FoV) of the electronic device based on temporal patterns of the signal information, wherein the initial prediction includes an indication of whether the external electronic device is within the FoV or outside the FoV of the electronic device;

perform, using a tracking filter, a smoothing operation on the range information and the first and second AoA, the smoothing operation generating a predicted location of the external electronic device relative to the FoV of the electronic device; and

generate, based on the smoothing operation, confidence values associated with the predicted location.

2. The electronic device of claim 1 , wherein the processor is further configured to:

obtain motion information based on movement of the electronic device;

determine to reset the tracking filter based on a comparison of motion of the electronic device to a threshold; and

when the predicted presence of the external electronic device is within the FoV of the electronic device, perform a reset to the tracking filter.

3. The electronic device of claim 1 , wherein the processor is further configured to:

obtain motion information based on movement of the electronic device, the motion information including orientation of the electronic device;

perform the smoothing operation using the orientation of the electronic device;

determine whether the electronic device is orientated along a horizontal axis and a vertical axis;

in response to a determination that the electronic device is not orientated along the horizontal axis, modify the first AoA to align with the horizontal axis; and

in response to a determination that the electronic device is not orientated along the vertical axis, modify the second AoA to align with the vertical axis.

4. The electronic device of claim 1 , wherein:

the channel information comprises features of a first channel impulse response (CIR) and a second CIR of wireless communication channels from the first and second antenna pairs, respectively, wherein the first CIR and the second CIR are based on the wireless signals communicated between the electronic device and the external electronic device;

the processor is further configured to generate a feature vector based on the range information, the first and the second AoA, and the features; and

the initial prediction is generated based on a neural network operating on the feature vector.

5. The electronic device of claim 4 , wherein the neural network includes a long short term memory (LSTM) neural network.

6. The electronic device of claim 4 , wherein the feature vector comprises at least two of:

the range information, indicating a distance between the electronic device and the external electronic device;

the first AoA;

the second AoA;

phase difference of arrival from the first and second antenna pairs;

an index of the phase difference of arrival from the first and second antenna pairs;

strength of a first peak of the first CIR;

strength of a strongest peak of the first CIR;

an index of the first peak from the first CIR;

an index of the strongest peak from the first CIR;

strength of a first peak of the second CIR;

strength of a strongest peak of the second CIR;

an index of the first peak from the second CIR;

an index of the strongest peak from the second CIR;

a variance of the first AoA over a first time interval;

a variance of the first AoA over a second time interval that is longer than the first time interval;

a variance of the second AoA over the first time interval;

a variance of the second AoA over the second time interval;

a first time difference between the first peak strength of the first CIR and the strongest peak strength of the first CIR; or

a second time difference between the first peak strength of the second CIR and the strongest peak strength of the second CIR.

7. The electronic device of claim 1 , wherein the processor is further configured to:

determine whether the external electronic device is within a line of site (LoS) of the electronic device;

in response to a determination that the external electronic device is not within the LoS of the electronic device, generate the initial prediction based on a first classifier that is trained in non-LoS scenarios; and

in response to a determination that the external electronic device is within the LoS of the electronic device, generate the initial prediction based on a second classifier that is trained in LoS scenarios.

8. A method for operating an electronic device, the method comprising:

obtaining signal information based on wireless signals received from an external electronic device via a first antenna pair and a second antenna pair, wherein the first and second antenna pairs are aligned along different axes and wherein the signal information includes channel information, range information, a first angle of arrival (AoA) of the wireless signals based on the first antenna pair, and a second AoA of the wireless signals based on the second antenna pair;

generating an initial prediction of a presence of the external electronic device relative to a field of view (FoV) of the electronic device based on temporal patterns of the signal information, wherein the initial prediction includes an indication of whether the external electronic device is within the FoV or outside the FoV of the electronic device;

performing, using a tracking filter, a smoothing operation on the range information and the first and second AoA, the smoothing operation generating a predicted location of the external electronic device relative to the FoV of the electronic device; and

generating, based on the smoothing operation, confidence values associated with the predicted location.

9. The method of claim 8 , further comprising:

obtaining motion information based on movement of the electronic device;

determining to reset the tracking filter based on a comparison of motion of the electronic device to a threshold; and

when the predicted presence of the external electronic device is within the FoV of the electronic device, performing a reset to the tracking filter.

10. The method of claim 8 , further comprising:

obtaining motion information based on movement of the electronic device, the motion information including orientation of the electronic device;

performing the smoothing operation using the orientation of the electronic device;

determining whether the electronic device is orientated along a horizontal axis and a vertical axis;

in response to determining that the electronic device is not orientated along the horizontal axis, modifying the first AoA to align with the horizontal axis; and

in response to determining that the electronic device is not orientated along the vertical axis, modifying the second AoA to align with the vertical axis.

11. The method of claim 8 , wherein:

the channel information comprises features of a first channel impulse response (CIR) and a second CIR of wireless communication channels from the first and second antenna pairs, respectively, wherein the first CIR and the second CIR are based on the wireless signals communicated between the electronic device and the external electronic device;

the method further comprises generating a feature vector based on the range information, the first and the second AoA, and the features; and

generating the initial prediction is based on a neural network operating on the feature vector.

12. The method of claim 11 , wherein the neural network includes a long short term memory (LSTM) neural network.

13. The method of claim 11 , wherein the feature vector comprises at least two of:

the range information, indicating a distance between the electronic device and the external electronic device;

the first AoA;

the second AoA;

phase difference of arrival from the first and second antenna pairs;

an index of the phase difference of arrival from the first and second antenna pairs;

strength of a first peak of the first CIR;

strength of a strongest peak of the first CIR;

an index of the first peak from the first CIR;

an index of the strongest peak from the first CIR;

strength of a first peak of the second CIR;

strength of a strongest peak of the second CIR;

an index of the first peak from the second CIR;

an index of the strongest peak from the second CIR;

a variance of the first AoA over a first time interval;

a variance of the first AoA over a second time interval that is longer than the first time interval;

a variance of the second AoA over the first time interval;

a variance of the second AoA over the second time interval;

a first time difference between the first peak strength of the first CIR and the strongest peak strength of the first CIR; or

a second time difference between the first peak strength of the second CIR and the strongest peak strength of the second CIR.

14. The method of claim 8 , further comprising:

determining whether the external electronic device is within a line of site (LoS) of the electronic device;

in response to determining that the external electronic device is not within the LoS of the electronic device, generating the initial prediction based on a first classifier that is trained in non-LoS scenarios; and

in response to determining that the external electronic device is within the LoS of the electronic device, generating the initial prediction based on a second classifier that is trained in LoS scenarios.

15. A non-transitory computer readable medium containing instructions that when executed by a processor of an electronic device, cause the processor to:

obtain signal information based on wireless signals received from an external electronic device via a first antenna pair and a second antenna pair, wherein the first and second antenna pairs are aligned along different axes and wherein the signal information includes channel information, range information, a first angle of arrival (AoA) of the wireless signals based on the first antenna pair, and a second AoA of the wireless signals based on the second antenna pair;

generate an initial prediction of a presence of the external electronic device relative to a field of view (FoV) of the electronic device based on temporal patterns of the signal information, wherein the initial prediction includes an indication of whether the external electronic device is within the FoV or outside the FoV of the electronic device;

perform, using a tracking filter, a smoothing operation on the range information and the first and second AoA, the smoothing operation generating a predicted location of the external electronic device relative to the FoV of the electronic device; and

generate, based on the smoothing operation, confidence values associated with the predicted location.

16. The non-transitory computer readable medium of claim 15 , wherein the instructions that when executed further cause the processor to:

obtain motion information based on movement of the electronic device;

determine to reset the tracking filter based on a comparison of motion of the electronic device to a threshold; and

when the predicted presence of the external electronic device is within the FoV of the electronic device, perform a reset to the tracking filter.

17. The non-transitory computer readable medium of claim 15 , wherein the instructions that when executed further cause the processor to:

obtain motion information based on movement of the electronic device, the motion information including orientation of the electronic device;

perform the smoothing operation using the orientation of the electronic device;

determine whether the electronic device is orientated along a horizontal axis and a vertical axis;

in response to a determination that the electronic device is not orientated along the horizontal axis, modify the first AoA to align with the horizontal axis; and

in response to a determination that the electronic device is not orientated along the vertical axis, modify the second AoA to align with the vertical axis.

18. The non-transitory computer readable medium of claim 15 , wherein:

the channel information comprises features of a first channel impulse response (CIR) and a second CIR of wireless communication channels from the first and second antenna pairs, respectively, wherein the first CIR and the second CIR are based on the wireless signals communicated between the electronic device and the external electronic device;

the instructions that when executed further cause the processor to generate a feature vector based on the range information, the first and the second AoA, and the features; and

the initial prediction that is generated is based on a neural network operating on the feature vector.

19. The non-transitory computer readable medium of claim 18 , wherein the neural network includes a long short term memory (LSTM) neural network.

20. The non-transitory computer readable medium of claim 18 , wherein the feature vector comprises at least two of:

the range information, indicating a distance between the electronic device and the external electronic device;

the first AoA;

the second AoA;

phase difference of arrival from the first and second antenna pairs;

an index of the phase difference of arrival from the first and second antenna pairs;

strength of a first peak of the first CIR;

strength of a strongest peak of the first CIR;

an index of the first peak from the first CIR;

an index of the strongest peak from the first CIR;

strength of a first peak of the second CIR;

strength of a strongest peak of the second CIR;

an index of the first peak from the second CIR;

an index of the strongest peak from the second CIR;

a variance of the first AoA over a first time interval;

a variance of the first AoA over a second time interval that is longer than the first time interval;

a variance of the second AoA over the first time interval;

a variance of the second AoA over the second time interval;

a first time difference between the first peak strength of the first CIR and the strongest peak strength of the first CIR; or

a second time difference between the first peak strength of the second CIR and the strongest peak strength of the second CIR.