Method and apparatus for handover management
Embodiments of the present disclosure provide methods and apparatus for handover management. A method performed by a handover management entity includes receiving a handover request from a network node. The handover request indicates that a first terminal device is to be handed over from a source cell to a target cell. The method further includes obtaining a position of the first terminal device. The method further includes predicting a handover result of the first terminal device by a classifier. The position of the first terminal device is used as an input of the classifier and the classifier is trained for the handover from the source cell to the target cell. The method further includes generating a handover decision based on the predicted handover result. The method further includes sending a handover response comprising the handover decision to the network node.
1 . A method performed by a handover management entity, comprising:
receiving a handover request from a network node, wherein the handover request indicates that a first terminal device is to be handed over from a source cell to a target cell;
obtaining a position of the first terminal device;
predicting a handover result of the first terminal device by a classifier, wherein the position of the first terminal device is used as an input of the classifier and the classifier is trained for the handover from the source cell to the target cell;
generating a handover decision based on the predicted handover result; and
sending a handover response comprising the handover decision to the network node.
2 . The method according to claim 1 , wherein the classifier is trained by a training set and the training set comprises history handover result data regarding the handover from the source cell to the target cell.
3 . The method according to claim 2 , wherein the training set further comprises handover result data regarding the handover from the source cell to the target cell generated by a generative adversarial network.
4 . The method according to claim 3 , wherein for a specific type of handover result, a corresponding type of generative adversarial network is trained by using the corresponding type of history handover result data regarding the handover from the source cell to the target cell.
5 . The method according to claim 2 , wherein the handover result data comprises a position of a terminal device, a handover result, a source cell identifier, a target cell identifier, a wrong cell identifier when the terminal device being handed over to a wrong cell.
6 . The method according to claim 5 , wherein the handover result data further comprises at least one of:
antenna information of the source cell,
antenna information of the target cell,
antenna information of the wrong cell,
a relative position of the terminal device to the antenna of a source cell,
a relative position of the terminal device to the antenna of the target cell, or
a relative position of the terminal device to the antenna of the wrong cell.
7 . The method according to claim 1 , wherein the handover result comprises at least one of:
too late handover;
too early handover;
handover to a wrong cell; or
handover success.
8 . The method according to claim 1 , wherein
when the predicted handover result indicates too late handover, the handover decision indicates the network node to decrease a handover triggering difficulty; or
when the predicted handover result indicates handover success, the handover decision indicates the network node to immediately perform the handover; or
when the predicted handover result indicates too early handover or handover to a wrong cell, the handover decision is generated further based on the movement information of the first terminal device.
9 . The method according to claim 8 , further comprising:
obtaining the movement information of the first terminal device.
10 . The method according to claim 9 , wherein the movement information of the first terminal device comprises at least one of:
a moving speed of the first terminal device;
a moving direction of the first terminal device; or
an acceleration of the first terminal device.
11 . The method according to claim 8 , wherein when the handover decision is generated further based on the movement information of the first terminal device,
when the first terminal device will enter into a handover success area at a particular time point, the handover decision indicates the network node the particular time point for performing the handover; or
when the first terminal device will enter into a handover success area at a particular time point, the handover decision indicates the network node to perform the handover, wherein the response comprising the handover decision is sent to the network node at or after the particular time point; or
when the first terminal device is moving far away from a handover success area, the handover decision comprises at least one recommended target cell and indicates the network node to perform cell reselection based on the at least one recommended target cell.
12 . The method according to claim 1 , further comprising:
determining at least one handover result area regarding the handover from the source cell to the target cell.
13 . The method according to claim 1 , further comprising:
receiving at least a part of history handover result data from the network node.
14 . The method according to claim 1 , wherein the handover management entity is deployed into an open radio access network.
15 . A method performed by a network node, comprising:
sending a handover request to a handover management entity, wherein the handover request indicates that a first terminal device is to be handed over from a source cell to a target cell; and
receiving a handover response comprising a handover decision from the handover management entity,
wherein the handover decision is generated based on a predicted handover result of the first terminal device and the predicted handover result of the first terminal device is predicted by a classifier, and
wherein a position of the first terminal device is used as an input of the classifier and the classifier is trained for the handover from the source cell to the target cell.
16 . The method according to claim 15 , wherein the classifier is trained by a training set and the training set comprises history handover result data regarding the handover from the source cell to the target cell.
17 . The method according to claim 16 , wherein the training set further comprises handover result data regarding the handover from the source cell to the target cell generated by a generative adversarial network.
18 . The method according to claim 17 , wherein for a specific type of handover result, a corresponding type of generative adversarial network is trained by using the corresponding type of history handover result data regarding the handover from the source cell to the target cell.
19 . The method according to claim 16 , wherein the handover result data comprises a position of a terminal device, a handover result, a source cell identifier, a target cell identifier, a wrong cell identifier when the terminal device being handed over to a wrong cell.
20 . The method according to claim 19 , wherein the handover result data further comprises at least one of:
antenna information of the source cell,
antenna information of the target cell,
antenna information of the wrong cell,
a relative position of the terminal device to the antenna of a source cell,
a relative position of the terminal device to the antenna of the target cell, or
a relative position of the terminal device to the antenna of the wrong cell.