US 12,261,792 B2
Group-common reference signal for over-the-air aggregation in federated learning
Hamed Pezeshki, San Diego, CA (US); Tao Luo, San Diego, CA (US); Taesang Yoo, San Diego, CA (US); Mahmoud Taherzadeh Boroujeni, San Diego, CA (US); Eren Balevi, San Diego, CA (US); Srinivas Yerramalli, San Diego, CA (US); and Junyi Li, Fairless Hills, PA (US)
Assigned to QUALCOMM Incorporated, San Diego, CA (US)
Filed by QUALCOMM Incorporated, San Diego, CA (US)
Filed on Aug. 29, 2022, as Appl. No. 17/898,180.
Prior Publication US 2023/0084883 A1, Mar. 16, 2023
Int. Cl. H04L 5/00 (2006.01); H04W 24/02 (2009.01); H04W 72/12 (2023.01); H04W 72/121 (2023.01)
CPC H04L 5/0048 (2013.01) [H04W 24/02 (2013.01); H04W 72/121 (2013.01)] 30 Claims
OG exemplary drawing
 
1. An apparatus for wireless communication at a network entity, comprising:
a memory; and
at least one processor coupled to the memory and configured to:
transmit a configuration to a group of user equipments (UEs) that configures the group of UEs to simultaneously transmit one or more group-common reference signals and to simultaneously transmit one or more gradient vectors associated with a federated learning procedure;
receive the one or more group-common reference signals and the one or more gradient vectors from the group of UEs based on the configuration via multiple channels; and
calculate an average gradient vector based on the one or more group-common reference signals and the one or more gradient vectors.