	US 12,088,378 B2
	Low overhead and high accuracy channel state feedback using neural networks
Taesang Yoo, San Diego, CA (US); Pavan Kumar Vitthaladevuni, San Diego, CA (US); Naga Bhushan, San Diego, CA (US); Krishna Kiran Mukkavilli, San Diego, CA (US); Tingfang Ji, San Diego, CA (US); Hwan Joon Kwon, San Diego, CA (US); June Namgoong, San Diego, CA (US); Jay Kumar Sundararajan, San Diego, CA (US); Sooryanarayanan Gopalakrishnan, San Diego, CA (US); and Wanshi Chen, San Diego, CA (US)
Assigned to QUALCOMM Incorporated, San Diego, CA (US)
Filed by QUALCOMM Incorporated, San Diego, CA (US)
Filed on Sep. 15, 2021, as Appl. No. 17/447,778.
Claims priority of provisional application 63/080,109, filed on Sep. 18, 2020.
Prior Publication US 2022/0094411 A1, Mar. 24, 2022
Int. Cl. H04J 3/14 (2006.01); G06N 3/02 (2006.01); H04B 7/0417 (2017.01); H04B 7/06 (2006.01); H04L 5/00 (2006.01)

CPC H04B 7/0626 (2013.01) [G06N 3/02 (2013.01); H04B 7/0417 (2013.01); H04L 5/0051 (2013.01)]

24 Claims

13. A method of wireless communication performed by a user equipment (UE), comprising:

receiving, from a network entity, one or more indications of a first precoder or at least one second precoder;

receiving, from the network entity, a first reference signal corresponding to a first precoded channel associated with the first precoder and at least one second reference signal corresponding to at least one second precoded channel associated with the at least one second precoder;

measuring the first reference signal to obtain the first precoded channel;

measuring the at least one second reference signal to obtain the at least one second precoded channel; and

transmitting, to the network entity, first channel state feedback based at least in part on the first precoded channel and second channel state feedback based at least in part on the at least one second precoded channel, wherein the first channel state feedback and the second channel state feedback are based at least in part on outputs of at least one neural network.