US 12,035,155 B2
Communications and measurement systems for characterizing radio propagation channels
Timothy J. O'Shea, Arlington, VA (US); Ben Hilburn, Reston, VA (US); Nathan West, Washington, DC (US); and Tamoghna Roy, Arlington, VA (US)
Assigned to DeepSig Inc., Arlington, VA (US)
Filed by DeepSig Inc., Arlington, VA (US)
Filed on Feb. 1, 2022, as Appl. No. 17/589,979.
Application 17/589,979 is a continuation of application No. 16/676,600, filed on Nov. 7, 2019, granted, now 11,284,277.
Claims priority of provisional application 62/756,681, filed on Nov. 7, 2018.
Prior Publication US 2022/0264328 A1, Aug. 18, 2022
Int. Cl. H04W 24/02 (2009.01); H04W 24/08 (2009.01)
CPC H04W 24/02 (2013.01) [H04W 24/08 (2013.01)] 19 Claims
OG exemplary drawing
 
1. A computer-implemented method comprising:
obtaining, at a first node, first communications over a radio frequency communication link from a second node;
extracting, by the first node, one or more fields in the first communications, the extracted fields used to analyze the radio frequency communication link;
training, by the first node and using the extracted fields from the first communications, a channel model comprising one or more machine learning models, wherein the channel model is trained to reproduce channel conditions of the radio frequency communication link, wherein the training comprises:
applying a set of desired channel effects to the one or more machine learning models;
in response to training the channel model, generating, using the trained model, a simulated set of channel approximations associated with the radio frequency communication link;
training, by the first node, at least one of encoder or decoder model using the generated simulated set of channel approximations; and
deploying, by the first node and over a network, at least one of the trained encoder or decoder model to a receiver modem hosted by the second node for radio frequency communications.