US 12,355,480 B2
Machine learning-based radio frequency (RF) front-end calibration
Lindsey Makana Kostas, San Diego, CA (US); Rishubh Khurana, Bangalore (IN); Ahmed Youssef, San Diego, CA (US); Francisco Ledesma, Richardson, TX (US); Sergey Murashov, San Diego, CA (US); Viral Ranpara, San Diego, CA (US); Enrique De La Rosa, San Diego, CA (US); Ming Leung, San Diego, CA (US); Gurkanwal Singh Sahota, Rancho Santa Fe, CA (US); and Shahnaz Shirazi, San Diego, CA (US)
Assigned to QUALCOMM Incorporated, San Diego, CA (US)
Filed by QUALCOMM Incorporated, San Diego, CA (US)
Filed on Mar. 22, 2023, as Appl. No. 18/188,337.
Application 18/188,337 is a continuation of application No. 17/650,334, filed on Feb. 8, 2022, granted, now 11,637,582.
Prior Publication US 2023/0254003 A1, Aug. 10, 2023
Int. Cl. H04B 1/40 (2015.01); H04B 17/11 (2015.01); H04B 17/21 (2015.01)
CPC H04B 1/40 (2013.01) [H04B 17/11 (2015.01); H04B 17/22 (2023.05)] 30 Claims
OG exemplary drawing
 
1. A method for calibrating a radio frequency (RF) circuit, comprising:
generating, based on a machine learning model and a first subset of RF circuit calibration parameters, values for a second subset of RF circuit calibration parameters;
identifying the second subset of RF circuit calibration parameters based on a dropout gradient descent network; and
writing at least the first subset of RF circuit calibration parameters to a memory associated with the RF circuit.