US 12,126,467 B2
Low resolution OFDM receivers via deep learning
Jeffrey Andrews, Austin, TX (US); and Eren Balevi, Austin, TX (US)
Assigned to Board of Regents, The University of Texas System, Austin, TX (US)
Filed by Board of Regents, The University of Texas System, Austin, TX (US)
Filed on Feb. 3, 2023, as Appl. No. 18/164,428.
Application 18/164,428 is a continuation of application No. 17/289,555, granted, now 11,575,544, previously published as PCT/US2019/058595, filed on Oct. 29, 2019.
Claims priority of provisional application 62/752,187, filed on Oct. 29, 2018.
Prior Publication US 2023/0261910 A1, Aug. 17, 2023
Int. Cl. H04B 1/00 (2006.01); G06N 3/08 (2023.01); H04B 1/40 (2015.01); H04L 5/00 (2006.01); H04L 25/02 (2006.01)
CPC H04L 25/0254 (2013.01) [G06N 3/08 (2013.01); H04B 1/0003 (2013.01); H04B 1/40 (2013.01); H04L 5/0007 (2013.01)] 20 Claims
OG exemplary drawing
 
1. A communication system comprising:
a low resolution analog to digital convertor configured to quantize a first set of complex valued data symbols received over a communication channel and corresponding to a first set of pilot symbols, the low resolution analog to digital convertor having a quantization resolution of six bits or less;
one or more processors, and
a memory storing executable instructions, the executable instructions when executed by the one or more processors cause the one or more processors to:
determine, using the first set of pilot symbols and the quantized first set of complex valued data symbols, desired output data of a machine learning model to be trained to estimate the communication channel;
train the machine learning model using the quantized first set of complex valued data symbols and the desired output data; and
determine an estimate of the communication channel using the trained machine learning model.