US 12,218,804 B2
Method of communication traffic prediction via continual learning with knowledge distillation, and an apparatus for the same
Hang Li, Montreal (CA); Ju Wang, Brossard (CA); Chengming Hu, Montreal (CA); Xi Chen, Montreal (CA); Xue Liu, Montreal (CA); Seowoo Jang, Seoul (KR); and Gregory Lewis Dudek, Westmount (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. 17/957,499.
Claims priority of provisional application 63/254,363, filed on Oct. 11, 2021.
Prior Publication US 2023/0114810 A1, Apr. 13, 2023
Int. Cl. H04L 41/16 (2022.01); H04L 41/0816 (2022.01); H04L 41/082 (2022.01); H04L 41/147 (2022.01); H04L 43/0876 (2022.01); H04W 16/04 (2009.01); H04W 28/16 (2009.01)
CPC H04L 41/147 (2013.01) [H04L 43/0876 (2013.01)] 19 Claims
OG exemplary drawing
 
1. A server for predicting load in a communication system, the server comprising:
at least one memory storing computer-readable instructions; and
at least one processor configured to execute the computer-readable instructions to:
obtain a first traffic data set from a base station during a first period of time;
obtain a first artificial intelligence (AI) model trained to predict a traffic load based on the first traffic data set;
obtain a second traffic data set from the base station, during a second period time that follows the first period of time;
obtain a second AI model trained to predict a traffic load based on the second traffic data set; and
update the second AI model based on a difference between a first traffic load that is predicted by the first AI model using the second traffic data set, and a second traffic load that is predicted by the second AI model using the second traffic data set.