US 12,341,583 B2
Codebook-based beamforming with random forest on content addressable memory
Mahsa Salmani, Kanata (CA); Sreenil Saha, Kanata (CA); and Armaghan Eshaghi, Kanata (CA)
Assigned to HUAWEI TECHNOLOGIES CO., LTD., Shenzhen (CN)
Filed by HUAWEI TECHNOLOGIES CO., LTD., Shenzhen (CN)
Filed on Sep. 27, 2023, as Appl. No. 18/373,500.
Claims priority of provisional application 63/419,270, filed on Oct. 25, 2022.
Prior Publication US 2024/0137083 A1, Apr. 25, 2024
Prior Publication US 2024/0235629 A9, Jul. 11, 2024
Int. Cl. H04B 7/0456 (2017.01); H04B 7/06 (2006.01)
CPC H04B 7/0482 (2013.01) [H04B 7/0617 (2013.01)] 20 Claims
OG exemplary drawing
 
1. A method of data collection from a wireless channel, the method comprising:
receiving, by user equipment (UE), from an access point (AP), a plurality of pilot signals over the wireless channel;
determining, by the UE, an index identifying a codeword selected from a pre-defined codebook in response to the UE receiving the plurality of pilot signals, the codeword for use as a beamforming vector for use in a data transmission phase between the AP and the UE; and
using the index as a label for a data set including the plurality of pilot signals, the label and the data set forming a labeled data for training a random forest for use in selecting a data transmission beamforming codeword given a plurality of features derived from a second plurality of pilot signals received by the UE during a channel training phase prior to the data transmission phase.