Video based channel state information

1. A method of wireless communication performed by a user equipment (UE), comprising:

receiving a video-based channel state information (VCSI) configuration;

causing a camera to capture channel state information (CSI) video based at least in part on one or more VCSI measurement parameters indicated in the VCSI configuration;

deriving, from the CSI video, one or more VCSI measurements using a machine learning model indicated in the VCSI configuration; and

transmitting a VCSI report that includes the one or more VCSI measurements.

2. The method of claim 1 , further comprising:

transmitting, to a base station, information that relates to a capability to capture the CSI video, wherein the VCSI configuration is received from the base station based at least in part on the capability to capture the CSI video.

3. The method of claim 2 , wherein the information that relates to the capability to capture the CSI video includes a number of lenses available to capture the CSI video, a power constraint associated with the camera, a capability to track a direction of the base station, or a capability to perform on-demand direction tuning.

4. The method of claim 1 , wherein the VCSI configuration is received at a modem, and wherein causing the camera to capture the CSI video includes:

providing, from the modem to an application layer, a command that includes one or more camera control parameters associated with capturing the CSI video.

5. The method of claim 4 , wherein the one or more camera control parameters include one or more of a camera lens identifier, focus information, a beam identifier or beam direction, camera activation or deactivation timing information, or an address to which to upload the CSI video.

6. The method of claim 4 , further comprising:

receiving, at the modem from the application layer, a message acknowledging the command that includes the one or more camera control parameters.

7. The method of claim 1 , wherein the VCSI report is transmitted to a base station via a physical uplink control channel or a physical uplink shared channel.

8. The method of claim 1 , wherein the VCSI report is transmitted based at least in part on a triggering event or expiration of a periodic reporting interval.

9. The method of claim 1 , wherein the VCSI configuration is received from an application server, and wherein the VCSI report is transmitted to the application server.

10. The method of claim 1 , wherein the one or more VCSI measurements include a beam identifier, one or more object detection bits that indicate whether humans or objects are present in the CSI video, and one or more unstructured bits that are derived using the machine learning model indicated in the VCSI configuration.

11. A method of wireless communication performed by a base station, comprising:

transmitting a video-based channel state information (VCSI) configuration that indicates one or more VCSI measurement parameters and a machine learning model associated with processing channel state information (CSI) video that is based at least in part on the one or more VCSI measurement parameters;

receiving a VCSI report that includes one or more VCSI measurements associated with a user equipment (UE); and

scheduling one or more uplink or downlink transmissions for the UE based at least in part on the one or more VCSI measurements.

12. The method of claim 11 , further comprising:

receiving, from the UE, information that relates to a capability to capture the CSI video, wherein the VCSI configuration is transmitted to the UE based at least in part on the capability to capture the CSI video.

13. The method of claim 12 , wherein the information that relates to the capability to capture the CSI video includes a number of lenses available to capture the CSI video, a power constraint associated with a camera, a capability to track a direction of the base station, or a capability to perform on-demand direction tuning.

14. The method of claim 11 , wherein the VCSI report is received via a physical uplink control channel or a physical uplink shared channel.

15. The method of claim 11 , wherein the VCSI report is received based at least in part on a triggering event or expiration of a periodic reporting interval.

16. The method of claim 11 , wherein the VCSI configuration is transmitted to an application server associated with the UE, and wherein the VCSI report is received from the application server.

17. The method of claim 11 , wherein the one or more VCSI measurements include a beam identifier, one or more object detection bits that indicate whether humans or objects are present in the CSI video, and one or more unstructured bits that are derived using the machine learning model indicated in the VCSI configuration.

18. The method of claim 17 , further comprising:

deriving CSI from the one or more unstructured bits using the machine learning model.

19. The method of claim 17 , further comprising:

determining that the one or more object detection bits indicate that one or more humans are present in the CSI video; and

configuring one or more mitigation actions to avoid transmitting a high-power beam directly toward the one or more humans.

20. The method of claim 11 , further comprising:

using the one or more VCSI measurements for one or more of positioning, a measurement configuration for the UE, a mobility decision for the UE, traffic prediction, load balancing, or energy saving.

21. A method of wireless communication performed by a base station, comprising:

configuring a radio unit (RU) to capture channel state information (CSI) video;

causing the CSI video to be delivered from the RU to a distributed unit (DU) via a fronthaul interface;

deriving, by the DU, one or more video-based CSI (VCSI) measurements from the CSI video using a machine learning model; and

scheduling one or more uplink or downlink transmissions based at least in part on the one or more VCSI measurements.

22. The method of claim 21 , wherein the CSI video delivered to the DU is uncompressed.

23. The method of claim 21 , wherein the CSI video delivered to the DU is compressed using the machine learning model.

24. The method of claim 21 , wherein configuring the RU to capture the CSI video includes:

carrying, over the fronthaul interface, camera control signaling that includes one or more parameters to capture the CSI video using a camera associated with the RU.

25. The method of claim 24 , wherein the one or more parameters include one or more of a camera lens identifier, focus information, a beam identifier or beam direction, camera activation or deactivation timing information, or an address to which to upload the CSI video.

26. The method of claim 21 , wherein the one or more VCSI measurements include one or more object detection bits that indicate whether humans or objects are present in the CSI video.

27. The method of claim 26 , further comprising:

determining that the one or more object detection bits indicate that one or more humans are present in the CSI video; and

configuring one or more mitigation actions to avoid transmitting a high-power beam directly toward the one or more humans.

28. The method of claim 27 , wherein the one or more mitigation actions include configuring one or more beams to avoid transmission toward the one or more humans or reducing a power associated with one or more beams that are transmitted toward the one or more humans.

29. The method of claim 21 , further comprising:

reporting the one or more VCSI measurements to a central unit for one or more of positioning, a measurement configuration for the UE, a mobility decision for the UE, traffic prediction, load balancing, or energy saving.

30. A user equipment (UE) for wireless communication, comprising:

a memory; and

one or more processors, coupled to the memory, configured to:

receive a video-based channel state information (VCSI) configuration;

cause a camera to capture channel state information (CSI) video based at least in part on one or more VCSI measurement parameters indicated in the VCSI configuration;

derive, from the CSI video, one or more VCSI measurements using a machine learning model indicated in the VCSI configuration; and

transmit a VCSI report that includes the one or more VCSI measurements.