	US 11,750,719 B2
	Method of performing communication load balancing with multi-teacher reinforcement learning, and an apparatus for the same
Jikun Kang, Montreal (CA); Xi Chen, Montreal (CA); Chengming Hu, Montreal (CA); Ju Wang, Brossard (CA); Gregory Lewis Dudek, Westmount (CA); and Xue Liu, Montreal (CA)
Assigned to SAMSUNG ELECTRONICS CO., LTD., Suwon-si (KR)
Filed by SAMSUNG ELECTRONICS CO., LTD., Suwon-si (KR)
Filed on Sep. 30, 2022, as Appl. No. 17/957,811.
Claims priority of provisional application 63/253,089, filed on Oct. 6, 2021.
Prior Publication US 2023/0105719 A1, Apr. 6, 2023
Int. Cl. H04L 67/5682 (2022.01); H04L 67/1004 (2022.01); H04L 41/16 (2022.01)

CPC H04L 67/5682 (2022.05) [H04L 41/16 (2013.01); H04L 67/1004 (2013.01)]

20 Claims

1. A server for obtaining a load balancing artificial intelligence (AI) model for a plurality of base stations in a communication system, the server comprising:

at least one memory storing instructions; and

at least one processor configured to execute the instructions to:

obtain a plurality of teacher models based on a plurality of traffic data sets collected from the plurality of base stations, respectively;

obtain a plurality of student models based on knowledge distillation from the plurality of teacher models;

obtain an ensemble student model by ensembling the plurality of student models; and

transmit the ensemble student model to the plurality of base stations, respectively;

receive feedback information of the ensemble student model from the plurality of base stations, and

update the ensemble student model based on the received feedback information.