Time of arrival information for positioning

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

training a machine learning model at the UE based on signals received from one or more transmission reception points (TRPs);

obtaining one or more measurements of one or more Positioning Reference Signal (PRS) resources from the one or more TRPs;

applying the machine learning model to the one or more measurements of one or more PRS resources to obtain ToA information for the one or more PRS resources;

selecting a reference timing mark based on the ToA information for the one or more PRS resources, and

reporting Time of Arrival (ToA) information for each of the one or more PRS resources and the reference timing mark to a network entity, wherein the ToA information reported is relative to the reference timing mark.

2 . The method of claim 1 , further comprising:

assigning, by the UE, the reference timing mark; and

transmitting an indication of the reference timing mark to the network entity.

3 . The method of claim 1 , wherein obtaining one or more measurements of one or more PRS resources comprises:

obtaining a first PRS resource measurement at a first time;

obtaining a second PRS resource measurement at a second later time, wherein the first PRS resource and the second PRS resource are received from a same TRP; and

reporting ToA information for the second PRS resource measurement relative to ToA information for the first PRS resource measurement.

4 . The method of claim 1 , wherein obtaining one or more measurements of one or more PRS resources comprises obtaining information indicative of a floating ToA for a first PRS resource, and wherein the reference timing mark is based on floating ToA information.

5 . The method of claim 1 , wherein the ToA information obtained using the machine learning model comprises Time of Flight (ToF) information based on time of arrival of the one or more PRS resources transmitted by a TRP, wherein the ToF information is based on a line of sight path between the UE and the TRP, or ToF information based on a first arrival path between the UE and the TRP.

6 . The method of claim 1 , further comprising:

updating the reference timing mark.

7 . The method of claim 6 , wherein updating the reference timing mark comprises selecting an updated reference timing mark at the UE.

8 . The method of claim 6 , wherein updating the reference timing mark comprises updating the reference timing mark based on an update timer.

9 . The method of claim 6 , wherein updating the reference timing mark comprises updating the reference timing mark based on an instruction received from the network entity.

10 . The method of claim 1 , wherein the reference timing mark is network-initiated.

11 . The method of claim 10 , further comprising:

receiving an indication of the reference timing mark from the network entity.

12 . The method of claim 10 , wherein the reference timing mark is based on a downlink signal.

13 . The method of claim 12 , wherein the downlink signal is a synchronization signal block (SSB) signal, channel state information reference signal (CSI-RS), tracking reference signal (TRS), or a previously-received PRS resource.

14 . The method of claim 1 , further comprising:

transmitting capability information indicating a UE capability to report the ToA information.

15 . The method of claim 1 , further comprising:

receiving configuration information from the network entity, the configuration information indicating at least resources for reporting the ToA information.

16 . The method of claim 1 , wherein obtaining the one or more measurements of the one or more PRS resources comprises obtaining a plurality of measurements of PRS resources each from a different TRP and associated with a measurement occasion, wherein reporting ToA information for each of the one or more PRS resources comprises reporting the ToA information relative to the reference timing mark, and wherein the reference timing mark is common for each of the PRS resources from the different TRPs.

17 . The method of claim 1 , wherein training the machine learning model at the UE comprises training the machine learning model to reduce a timing difference between a clock at the UE and a clock at the network entity.

18 . The method of claim 17 , further comprising, subsequent to obtaining the measurements of the one or more PRS resources:

selecting an updated reference timing mark based on training the model to reduce an updated timing difference between a clock at the UE and a clock at the network entity.

19 . The method of claim 1 , wherein the network entity is a location server in communication with the UE.

20 . A user equipment (UE), comprising:

one or more memories;

one or more transceivers; and

one or more processors communicatively coupled to the one or more memories and the one or more transceivers, the one or more processors, either alone or in combination, configured to:

train a machine learning model at the UE based on signals received from one or more transmission reception points (TRPs);

obtain one or more measurements of one or more PRS resources from the one or more TRPs;

apply the machine learning model to the one or more measurements of the one or more PRS resources to obtain ToA information for the one or more PRS resources;

select a reference timing mark based on the ToA information for the one or more PRS resources;

report, via the one or more transceivers, ToA information for each of the one or more PRS resources and the reference timing mark to a network entity, wherein the ToA information reported is relative to the reference timing mark.

21 . The UE of claim 20 , wherein the one or more processors, either alone or in combination, are further configured to:

assign the reference timing mark; and

transmit, via the one or more transceivers, an indication of the reference timing mark to the network entity.

22 . The UE of claim 20 , wherein the one or more processors configured to obtain one or more measurements of one or more PRS resources comprises the one or more processors, either alone or in combination, configured to:

obtain a first PRS resource measurement at a first time;

obtain a second PRS resource measurement at a second later time, wherein the first PRS resource and the second PRS resource are received from a same TRP; and

reporting ToA information for the second PRS resource measurement relative to ToA information for the first PRS resource measurement.

23 . The UE of claim 20 , wherein the one or more processors, either alone or in combination, are further configured to:

transmit, via the one or more transceivers, capability information indicating a UE capability to report the ToA information.

24 . A method of communication at a network entity, the method comprising:

transmitting one or more Positioning Reference Signal (PRS) resources; and

receiving Time of Arrival (ToA) information for each of the one or more PRS resources, wherein the ToA information reported is relative to a reference timing mark and the ToA information reported is based on a machine learning model applied to one or more measurements of the one or more PRS resources, wherein the machine learning model is trained based on signals prior to the one or more PRS resources and the reference timing mark is selected based on the ToA information for the one or more PRS resources.

25 . The method of claim 24 , further comprising:

transmitting an indication of the reference timing mark, wherein the reference timing mark is based on a downlink signal selected from a group consisting of a signal included in a synch signal block, a channel state information reference signal (CSI-RS), a tracking reference signal (TRS), and a previously transmitted PRS resource.

26 . The method of claim 24 , further comprising:

transmitting configuration information indicating at least resources for reporting the ToA information.

27 . A network entity, comprising:

one or more memories;

one or more transceivers; and

one or more processors communicatively coupled to the one or more memories and the one or more transceivers, the one or more processors, either alone or in combination, configured to:

transmit, via the one or more transceivers, one or more PRS resources; and

receive, via the one or more transceivers, Time of Arrival (ToA) information for each of the one or more PRS resources, wherein the ToA information reported is relative to a reference timing mark and the ToA information reported is based on a machine learning model applied to the one or more PRS resources, wherein the machine learning model is trained based on signals prior to the one or more PRS resources and the reference timing mark is selected based on the ToA information for the one or more PRS resources.