Apparatus and method for mobility with AI/ML channel prediction
View Patent ↗A method performed by at least one processor of a user equipment (UE) having a connection with a serving cell in a wireless communication network includes receiving, from a target cell in the wireless communication network, a set of resources for a slot n. The method includes measuring, for the target cell, a network quality parameter based on the set of resources. The method includes reporting, to the serving cell, results corresponding to the measuring of the network quality parameter. The method includes receiving, based on (i) a prediction from an artificial intelligence machine learning (AI/ML) model using the measuring of the network quality parameter as an input into the AI/ML model and (ii) a determination that a cell switching condition is satisfied, a command to switch from the serving cell to the target cell at a time corresponding to slot n+k.
1 . A method performed by at least one processor of a user equipment (UE) having a connection with a serving cell in a wireless communication network, the method comprising:
receiving, from a target cell in the wireless communication network, a set of resources for a slot n;
measuring, for the target cell, a network quality parameter based on the set of resources;
reporting, to the serving cell, results corresponding to the measuring of the network quality parameter; and
receiving, based on (i) a prediction from an artificial intelligence machine learning (AI/ML) model using the measuring of the network quality parameter as an input into the AI/ML model and (ii) a determination that a cell switching condition is satisfied, a command to switch from the serving cell to the target cell at a time corresponding to slot n+k,
wherein n and k are integers greater than zero,
wherein the prediction from the AI/ML model is a predicted network quality parameter of the target cell at slot n+k, and
wherein the network quality parameter is a channel impulse response (CIR).
2 . The method according to claim 1 , wherein the set of resources include a plurality of reference signals, and wherein the network quality parameter is a reference signal resource power (RSRP) of each reference signal.
3 . The method of claim 2 , wherein the cell switching condition specifies that the AI/ML model predicts that at the time corresponding to slot n+k, the target cell has a RSRP that is greater than a threshold.
4 . The method of claim 2 , wherein the cell switching condition specifies that the AI/ML model predicts that at the time corresponding to slot n+k, the serving cell has a RSRP that is less than a first threshold and the target cell has a RSRP that is greater than a second threshold.
5 . The method according to claim 1 , wherein the AI/ML model is located at the UE, and the reporting the results corresponding to the measuring of the network quality parameter includes the prediction.
6 . The method according to claim 5 , wherein the UE receives the AI/ML model from the serving cell.
7 . The method according to claim 1 , wherein the AI/ML model is located at a base station of the serving cell, and wherein the base station of the serving cell receives the AI/ML model from the target cell.
8 . The method of claim 1 , further comprising:
after receiving the command to switch from the serving cell to the target cell, performing downlink synchronization and uplink synchronization with the target cell before the time corresponding to slot n+k lapses.
9 . The method of claim 1 , further comprising:
after receiving the command to switch from the serving cell to the target cell, transmitting a request to the serving cell to cancel the switch based on a determination a condition for switching from the serving cell to the target cell is no longer valid.
10 . The method of claim 1 , further comprising:
increasing a periodicity of measuring one or more neighboring cells based on a determination the AI/ML model predicts that a network quality parameter for the serving cell will stay above a threshold for a predetermined time period.
11 . A user equipment (UE) having a connection with a serving cell in a wireless communication network, the UE comprising:
at least one memory configured to store computer program code; and
at least one processor configured to access said at least one memory and operate as instructed by the computer program code, the computer program code including:
first receiving code configured to cause at least one of said at least one processor to receive, from a target cell in the wireless communication network, a set of resources for a slot n,
measuring code configured to cause at least one of said at least one processor to measure, for the target cell, a network quality parameter based on the set of resources,
reporting code configured to cause at least one of said at least one processor to report, to the serving cell, results corresponding to the measuring of the network quality parameter, and
second receiving code configured to cause at least one of said at least one processor to receive, based on (i) a prediction from an artificial intelligence machine learning (AI/ML) model using the measuring of the network quality parameter as an input into the AI/ML model and (ii) a determination that a cell switching condition is satisfied, a command to switch from the serving cell to the target cell at a time corresponding to slot n+k,
wherein n and k are integers greater than zero,
wherein the prediction from the AI/ML model is a predicted network quality parameter of the target cell at slot n+k, and
wherein the network quality parameter is a channel impulse response (CIR).
12 . The UE according to claim 11 , wherein the set of resources include a plurality of reference signals, and wherein the network quality parameter is a reference signal resource power (RSRP) of each reference signal.
13 . The UE of claim 12 , wherein the cell switching condition specifies that the AI/ML model predicts that at the time corresponding to slot n+k, the target cell has a RSRP that is greater than a threshold.
14 . The UE of claim 12 , wherein the cell switching condition specifies that the AI/ML model predicts that at the time corresponding to slot n+k, the serving cell has a RSRP that is less than a first threshold and the target cell has a RSRP that is greater than a second threshold.
15 . The UE according to claim 11 , wherein the AI/ML model is located at the UE, and wherein the reported results corresponding to the measuring of the network quality parameter includes the prediction.
16 . The UE according to claim 15 , wherein the UE receives the AI/ML model from the serving cell.
17 . The UE according to claim 11 , wherein the AI/ML model is located at a base station of the serving cell, and wherein the base station of the serving cell receives the AI/ML model from the target cell.
18 . A non-transitory computer readable medium having instructions stored therein, which when executed by a processor in a user equipment (UE) having a connection with a serving cell in a wireless communication network, cause the processor to execute a method comprising:
receiving, from a target cell in the wireless communication network, a set of resources for a slot n;
measuring, for the target cell, a network quality parameter based on the set of resources;
reporting, to the serving cell, results corresponding to the measuring of the network quality parameter; and
receiving, based on (i) a prediction from an artificial intelligence machine learning (AI/ML) model using the measuring of the network quality parameter as an input into the AI/ML model and (ii) a determination that a cell switching condition is satisfied, a command to switch from the serving cell to the target cell at a time corresponding to slot n+k,
wherein n and k are integers greater than zero,
wherein the prediction from the AI/ML model is a predicted network quality parameter of the target cell at slot n+k, and
wherein the network quality parameter is a channel impulse response (CIR).