Method and apparatus for determining prediction for status of wireless network

1 . An apparatus A first node for wireless communication, comprising:

at least one memory; and

at least one processor coupled with the at least one memory and operable to cause the first node to:

transmit a request associated with a prediction for a status of a wireless network to a second node and a third node;

receive information of a trained artificial intelligence (AI) model from the third node;

transmit, to the second node and based at least in part on the information, data for input to the trained AI model; and

receive the prediction from the second node, wherein the prediction is based at least in part on the trained AI model and the data.

2 . The first node of claim 1 , wherein the request comprises at least one of:

an identity of a user equipment (UE) associated with the trained AI model;

an identity of a radio access network (RAN) node associated with the trained AI model;

an identity of a cell associated with the trained AI model;

an accuracy of the prediction;

an improvement requirement associated with the accuracy;

time information;

first information of the input; or

a function of the trained AI model.

3 . The first node of claim 1 , wherein the information comprises at least one of:

an identity of the trained AI model;

an identity of a user equipment (UE) associated with the trained AI model;

an identity of a radio access network (RAN) node associated with the trained AI model;

an identity of a cell associated with the trained AI model;

a list of predictions associated with the trained AI model;

an accuracy associated with each item in the list of predictions;

an improvement associated with each item in the list of predictions;

second information of the input;

a function of the trained AI model; or

third information of feedback.

4 . The first node of claim 3 , wherein the at least one processor is further operable to cause the first node to:

transmit a feedback for the trained AI model to the third node based at least in part on the third information of the feedback, wherein the feedback comprises at least one of:

an identity of the trained AI model;

a cause for transmitting the feedback; or

additional data for AI model training.

5 . The first node of claim 4 , wherein the feedback for the trained AI model is transmitted periodically with a predetermined period or in one shot with a format, and wherein the cause for transmitting the feedback comprises the accuracy or the improvement being below a threshold.

6 . The first node of claim 1 , wherein the at least one processor is further operable to cause the first node to:

receive an AI capability of a radio access network (RAN) node via at least one of:

an Xn interface from the RAN node during a Xn interface setup procedure, a SN addition procedure, or a SN modification procedure; or

an N2 interface from operations, administration and maintenance (OAM) or access and mobility management function (AMF).

7 . The first node of claim 6 , wherein the AI capability comprises at least one of inference, training an AI model, providing an AI model, updating an AI model, or providing the prediction.

8 . The first node of claim 1 , wherein the at least one processor is further operable to cause the first node to:

receive an additional request from a radio access network (RAN) node or transmit the additional request to the RAN node, wherein the additional request comprises at least one of:

a measurement of the status of the wireless network and time information associated with the measurement;

the prediction and time information associated with the prediction;

a period for an update of the status of the wireless network;

an improvement requirement;

a period of the measurement; or

a period of the prediction.

9 . The first node of claim 1 , wherein the information comprises one or more UE identifiers of one or more user equipment (UEs), and wherein the trained AI model is configured to generate the prediction for the one or more UEs.

10 . The first node of claim 9 , wherein the at least one processor is further operable to cause the first node to:

select, based at least in part on the one or more UE identifiers, the data for input to the trained AI model, wherein the data comprises measurement data associated with the one or more UEs.

11 . The first node of claim 1 , wherein the first node comprises at least one of a radio access network (RAN) node or a data node configured to request and receive the prediction, wherein the second node comprises an AI inference node configured to generate the prediction based at least in part on the data, and wherein the third node comprises an AI training node configured to train the AI model.

12 . The first node of claim 1 , wherein the at least one processor is further operable to cause the first node to:

receive, from at least one of a user equipment (UE) or a radio access network (RAN) node, the data for input to the trained AI model, wherein the data comprises at least one of a measurement of the status of the wireless network, time information associated with the measurement, a period for an update of the status of the wireless network, or a period of the measurement.

13 . The first node of claim 1 , wherein the information requests training data associated with the trained AI model, and wherein the at least one processor is further operable to cause the first node to:

transmit, to the third node, feedback for the trained AI model, wherein the feedback comprises at least one of an identity of the trained AI model, a cause for transmitting the feedback, or the training data.

14 . A third node for wireless communication, comprising:

at least one memory; and

at least one processor coupled with the at least one memory and operable to cause the third node to:

receive a request associated with a prediction for a status of a wireless network from a first node or from a second node;

transmit a trained artificial intelligence (AI) model to the second node; and

transmit information of the trained AI model to the first node, wherein the prediction is based at least in part on the trained AI model and data associated with the information.

15 . The third node of claim 14 , wherein the request comprises at least one of:

an identity of a user equipment (UE) associated with the trained AI model;

an identity of a radio access network (RAN) node associated with the trained AI model;

an identity of a cell associated with the trained AI model;

an accuracy of the prediction;

an improvement requirement associated with the accuracy;

time information;

first information of an input for the trained AI model; or

a function of the trained AI model.

16 . The third node of claim 14 , wherein the information comprises at least one of:

an identity of the trained AI model;

an identity of a user equipment (UE) associated with the trained AI model;

an identity of a radio access network (RAN) node associated with the trained AI model;

an identity of a cell associated with the trained AI model;

a list of predictions;

an accuracy associated with each item in the list of predictions;

an improvement associated with each item in the list of predictions;

second information of an input for the trained AI model;

a function of the trained AI model; or

third information of feedback.

17 . The third node of claim 16 , wherein the at least one processor is further operable to cause the third node to:

transmit, to the first node, at least one of a predetermined period or a one shot format associated with the feedback for the trained AI model; and

receive, based at least in part on the at least one of the predetermined period or the one shot format, the feedback for the trained AI model from the first node based at least in part on the third information of the feedback, wherein the feedback comprises at least one of an identity of the trained AI model, a cause for transmitting the feedback, or additional data for AI model training.

18 . The third node of claim 17 , wherein the feedback for the trained AI model is received periodically with the predetermined period or in one shot with the one shot format, and wherein the cause for transmitting the feedback comprises an accuracy or an improvement of the trained AI model being below a threshold.

19 . A second node for wireless communication, comprising:

at least one memory; and

at least one processor coupled with the at least one memory and operable to cause the second node to:

receive, from a first node, a request associated with a prediction for a status of a wireless network;

receive, from a third node, a trained artificial intelligence (AI) model;

receive, from the first node and based at least in part on information associated with the trained AI model, data for input to the trained AI model;

generate the prediction as output from the trained AI model based at least in part on the request; and

transmit the prediction to the first node.

20 . The second node of claim 19 , wherein the request comprises at least one of:

an identity of a user equipment (UE) associated with the trained AI model;

an identity of a radio access network (RAN) node associated with the trained AI model;

an identity of a cell associated with the trained AI model;

an accuracy of the prediction;

an improvement requirement associated with the accuracy;

time information;

first information of the input; or

a function of the trained AI model.