Artificial intelligence (AI) for communication networks
A method and apparatus are provided for artificial intelligence (AI) for communication networks. In one embodiment, a method implemented in a core network node includes collecting location information, the collected location information associated with movement of at least one user equipment within a network; predicting based at least part of the collected location information; and communicating an indication of the prediction.
1. A method implemented in a core network node, the method comprising:
collecting location information, the collected location information associated with movement of at least one user equipment within a network;
predicting, according to a prediction algorithm and based at least in part on the collected location information, a movement path of the at least one user equipment relative to the network, the prediction algorithm associating each of a plurality of different time windows with a different weight, the predicting according to the prediction algorithm comprising determining at least one probability based at least in part on a weighted current prediction probability value and a weighted historical probability prediction value; and
communicating an indication of the prediction.
2. The method of claim 1 , wherein the plurality of different time windows includes at least one a time of day, a day-of-the-week, a week-of-the-year, a month-of-the-year and a season, each time window associated with a different weight in the prediction algorithm.
3. The method of claim 1 , wherein an outcome of a preceding prediction determines a weight value applied to the current probability prediction value and a weight value applied to the historical prediction value.
4. The method according to claim 1 , wherein:
collecting the location information further comprises:
collecting path information and corresponding time information, the path information and the time information indicating the movement of the at least one user equipment between at least two radio network areas in the network;
the predicting being based at least in part on the collected path information and the time information; and
communicating the indication further comprises:
communicating the indication of the predicted movement path of the at least one user equipment.
5. The method of claim 4 , wherein the at least two radio network areas include at least two tracking areas and at least two cells.
6. The method of claim 4 , wherein the indication of the predicted movement path includes at least one cell global identifier, CGI, the at least one CGI indicating at least one cell that the at least one user equipment is expected to move to according to the predicted movement path.
7. The method of claim 4 , wherein the predicting further comprises:
determining at least one probability of at least one movement path of the at least one user equipment, the at least one movement path including at least one of a cell and a neighboring cell of a base station.
8. The method of claim 4 , wherein the indication of the predicted movement path is configured to be included in a paging request for the at least one user equipment, the paging request indicating, for at least one cell in the paging request, a probability prediction that the at least one user equipment is in the at least one cell.
9. A core network node configured to facilitate communication of a user equipment in a communication system, the core network node comprising processing circuitry, the processing circuitry configured to cause the core network node to:
collect location information, the collected location information associated with movement of at least one user equipment within a network;
predict, according to a prediction algorithm and based at least in part on the collected location information, a movement path of the at least one user equipment relative to the network, the prediction algorithm associating each of a plurality of different time windows with a different weight, the predicting according to the prediction algorithm comprising determining at least one probability based at least in part on a weighted current prediction probability value and a weighted historical probability prediction value; and
communicate an indication of the prediction.
10. The core network node of claim 9 , wherein the plurality of different time windows includes at least one a time of day, a day-of-the-week, a week-of-the-year, a month-of-the-year and a season, each time window associated with a different weight in the prediction algorithm.
11. The core network node of claim 9 , wherein an outcome of a preceding prediction determines a weight value applied to the current probability prediction value and a weight value applied to the historical prediction value.
12. The core network node according to claim 9 , wherein the processing circuitry is further configured to cause the core network node to:
collect the location information by being configured to cause the core network node to:
collect path information and corresponding time information, the path information and the time information indicating the movement of the at least one user equipment between at least two radio network areas in the network;
the predicting being based at least in part on the collected path information and the time information; and
communicate the indication by being configured to cause the core network node to:
communicate the indication of the predicted movement path of the at least one user equipment.
13. The core network node of claim 12 , wherein the at least two radio network areas include at least two tracking areas and at least two cells.
14. The core network node of claim 12 , wherein the indication of the predicted movement path includes at least one cell global identifier, CGI, the at least one CGI indicating at least one cell that the at least one user equipment is expected to move to according to the predicted movement path.
15. The core network node of claim 12 , wherein the processing circuitry is further configured to cause the core network node to predict by being configured to cause the core network node to:
determine at least one probability of at least one movement path of the at least one user equipment, the at least one movement path including at least one of a cell and a neighboring cell of a base station.
16. The core network node of claim 12 , wherein the indication of the predicted movement path is configured to be included in a paging request for the at least one user equipment, the paging request indicating, for at least one cell in the paging request, a probability prediction that the at least one user equipment is in the at least one cell.