Beam-based machine learning-enabled RFFP positioning

1. An apparatus for wireless communication at a network entity, comprising:

a memory;

at least one transceiver; and

at least one processor communicatively connected to the memory and the at least one transceiver, the at least one processor configured to:

receive, from one or more wireless devices, a plurality of first radio frequency (RF) fingerprints, each of the plurality of first RF fingerprints being associated with at least one directional feature and a location;

receive a request to determine a position of a user equipment (UE) based on at least one second RF fingerprint associated with the UE or captured by the UE; and

estimate the position of the UE based at least in part on matching the at least one second RF fingerprint to at least one of the plurality of first RF fingerprints;

wherein the network entity is a location server, the UE, or a base station associated with a machine learning (ML) module;

wherein the at least one processor is further configured to:

train the ML module based on each of the plurality of first RF fingerprints and the associated at least one directional feature and the associated location to create a mapping between each of the plurality of first RF fingerprints and the associated at least one directional feature and the associated location, wherein the position of the UE is estimated based on the mapping between the at least one second RF fingerprint and the associated at least one directional feature and the associated location.

2. The apparatus of claim 1 , wherein the ML module is associated with a neural network (NN), a deep NN, or a random forest algorithm.

3. The apparatus of claim 1 , wherein the request is received from the UE, a location server, or a base station.

4. The apparatus of claim 1 , wherein the at least one directional feature corresponds to a channel impulse response (CIR) or a channel frequency response (CFR) captured from a positioning reference signal (PRS) or a sounding reference signal (SRS).

5. The apparatus of claim 1 , wherein the at least one directional feature corresponds to a channel impulse response (CIR) or a channel frequency response (CFR) captured at multiple antenna pairs, each of the multiple antenna pairs including a transmission antenna of a transmitting device and a reception antenna of a receiving device.

6. The apparatus of claim 1 , wherein the at least one directional feature corresponds to at least one of a normalized histogram of received signal strength measured over a period of time or a range of frequency captured at multiple antenna pairs, each of the multiple antenna pairs including a transmission antenna of a transmitting device and a reception antenna of a receiving device.

7. The apparatus of claim 1 , wherein the at least one directional feature corresponds to a complex-sum of channel impulse response (CIR) or channel frequency response (CFR) captured at multiple antenna pairs or multiple beam pairs, each of the multiple antenna pairs including a transmission antenna of a transmitting device and a reception antenna of a receiving device and each of the multiple beam pairs including a transmission beam of the transmitting device and a reception beam of the receiving device.

8. The apparatus of claim 1 , wherein the at least one directional feature corresponds to at least one of a normalized histogram of received signal strength measured over a period of time or a range of frequency captured at multiple antenna pairs and multiple beam pairs, each of the multiple antenna pairs including a transmission antenna of a transmitting device and a reception antenna of a receiving device and each of the multiple beam pairs including a transmission beam of the transmitting device and a reception beam of the receiving device.

9. The apparatus of claim 1 , wherein the at least one directional feature corresponds to a compressed channel impulse response (CIR) or a compressed channel frequency response (CFR) captured at multiple antenna pairs or multiple beam pairs, each of the multiple antenna pairs including a transmission antenna of a transmitting device and a reception antenna of a receiving device and each of the multiple beam pairs including a transmission beam of the transmitting device and a reception beam of the receiving device.

10. The apparatus of claim 1 , wherein the at least one directional feature corresponds to at least one of a compressed normalized histogram of received signal strength measured over a period of time or a range of frequency captured at multiple antenna pairs and multiple beam pairs, each of the multiple antenna pairs including a transmission antenna of a transmitting device and a reception antenna of a receiving device and each of the multiple beam pairs including a transmission beam of the transmitting device and a reception beam of the receiving device.

11. The apparatus of claim 1 , wherein the at least one directional feature corresponds to one or more strongest channel impulse responses (CIRs) or one or more strongest channel frequency response (CFRs) captured at multiple antenna pairs or multiple beam pairs, each of the multiple antenna pairs including a transmission antenna of a transmitting device and a reception antenna of a receiving device and each of the multiple beam pairs including a transmission beam of the transmitting device and a reception beam of the receiving device.

12. The apparatus of claim 1 , wherein the at least one directional feature corresponds to at least one of one or more strongest normalized histograms of received signal strength measured over a period of time or a range of frequency captured at multiple antenna pairs, each of the multiple antenna pairs including a transmission antenna of a transmitting device and a reception antenna of a receiving device.

13. The apparatus of claim 1 , wherein each of the plurality of first RF fingerprints is further associated with at least one beam feature, and wherein the position of the UE is estimated further based at least in part on the at least one beam feature.

14. The apparatus of claim 13 , wherein the at least one beam feature includes one or more of: a beam shape, a cumulative distribution function (CDF) or a probability density function (PDF) of beam-based spherical effective isotropic radiated power (EIRP) measurements, a beam configuration, a beam measurement, or a beam embedding based on a graph convolutional network (GCN).

15. The apparatus of claim 13 , wherein the at least one beam feature is associated with at least one of a transmission beam at a transmitting device or a reception beam at a receiving device.

16. The apparatus of claim 1 , wherein the at least one second RF fingerprint associated with the UE is based on one beam pair between one beam of the UE and one beam of a transmission reception point (TRP).

17. The apparatus of claim 1 , wherein the at least one second RF fingerprint associated with the UE is based on a beam sweeping between one or more beams of the UE and one or more beams of a transmission reception point (TRP).

18. The apparatus of claim 1 , wherein the network entity is associated with a machine learning (ML) module, and the at least one processor is further configured to:

train the ML module to map one directional feature captured from one antenna pair to one location, the one antenna pair being associated with one transmission reception point (TRP) and one UE.

19. The apparatus of claim 1 , wherein the network entity is associated with a machine learning (ML) module, and the at least one processor is further configured to:

train the ML module to map multiple directional features captured from multiple swept beam pairs to one location, the multiple swept beam pairs being associated with one transmission reception point (TRP) and one UE.

20. The apparatus of claim 1 , wherein the network entity is associated with a machine learning (ML) module, and the at least one processor is further configured to:

train the ML module to map multiple directional features captured from multiple beam pairs to one location, the multiple beam pairs being associated with one UE and multiple transmission reception points (TRPs).

21. The apparatus of claim 1 , wherein the network entity is associated with a machine learning (ML) module, and the at least one processor is further configured to:

train the ML module to map multiple directional features captured from multiple swept beam pairs to one location, the multiple swept beam pairs being associated with one UE and multiple transmission reception points (TRPs).

22. A method of wireless communication at a network entity, comprising:

receiving, from one or more wireless devices, a plurality of first radio frequency (RF) fingerprints, each of the plurality of first RF fingerprints being associated with at least one directional feature and a location;

receiving a request to determine a position of a user equipment (UE) based on at least one second RF fingerprint associated with the UE or captured by the UE; and

estimating the position of the UE based at least in part on matching the at least one second RF fingerprint to at least one of the plurality of first RF fingerprints;

wherein the network entity is a location server, the UE, or a base station associated with a machine learning (ML) module, and

the method further comprising:

training the ML module based on each of the plurality of first RF fingerprints and the associated at least one directional feature and the associated location to create a mapping between each of the plurality of first RF fingerprints and the associated at least one directional feature and the associated location, wherein the position of the UE is estimated based on the mapping between the at least one second RF fingerprint and the associated at least one directional feature and the associated location.

23. The method of claim 22 , wherein the ML module is associated with a neural network (NN), a deep NN, or a random forest algorithm.

24. The method of claim 22 , wherein each of the plurality of first RF fingerprints is further associated with at least one beam feature, and wherein the position of the UE is estimated further based at least in part on the at least one beam feature, and wherein the at least one beam feature includes one or more of: a beam shape, a cumulative distribution function (CDF) or a probability density function (PDF) of beam-based spherical effective isotropic radiated power (EIRP) measurements, a beam configuration, a beam measurement, or a beam embedding based on a graph convolutional network (GCN).

25. The method of claim 22 , wherein the at least one second RF fingerprint associated with the UE is based on one beam pair between one beam of the UE and one beam of a transmission reception point (TRP) or based on a beam sweeping between one or more beams of the UE and one or more beams of the TRP.

26. An apparatus for wireless communication at a network entity, comprising:

means for receiving, from one or more wireless devices, a plurality of first radio frequency (RF) fingerprints, each of the plurality of first RF fingerprints being associated with at least one directional feature and a location;

means for receiving a request to determine a position of a user equipment (UE) based on at least one second RF fingerprint associated with the UE or captured by the UE; and

means for estimating the position of the UE based at least in part on matching the at least one second RF fingerprint to at least one of the plurality of first RF fingerprints;

wherein the network entity is a location server, the UE, or a base station associated with a machine learning (ML) module, and

the apparatus further comprising:

means for training the ML module based on each of the plurality of first RF fingerprints and the associated at least one directional feature and the associated location to create a mapping between each of the plurality of first RF fingerprints and the associated at least one directional feature and the associated location, wherein the position of the UE is estimated based on the mapping between the at least one second RF fingerprint and the associated at least one directional feature and the associated location.

27. A non-transitory computer-readable medium storing computer executable code at a network entity, the code when executed by a processor causes the processor to:

receive, from one or more wireless devices, a plurality of first radio frequency (RF) fingerprints, each of the plurality of first RF fingerprints being associated with at least one directional feature and a location;

receive a request to determine a position of a user equipment (UE) based on at least one second RF fingerprint associated with the UE or captured by the UE; and

estimate the position of the UE based at least in part on matching the at least one second RF fingerprint to at least one of the plurality of first RF fingerprints;

wherein the network entity is a location server, the UE, or a base station associated with a machine learning (ML) module, and

the code when executed by the processor further causes the processor to:

train the ML module based on each of the plurality of first RF fingerprints and the associated at least one directional feature and the associated location to create a mapping between each of the plurality of first RF fingerprints and the associated at least one directional feature and the associated location, wherein the position of the UE is estimated based on the mapping between the at least one second RF fingerprint and the associated at least one directional feature and the associated location.