US 12,375,101 B1
Distributed system and method for adaptive neural network-based data compression
Zhu Li, Overland Park, KS (US); Paras Maharjan, Kansas City, MO (US); and Brian Galvin, Silverdale, WA (US)
Assigned to ATOMBEAM TECHNOLOGIES INC., Moraga, CA (US)
Filed by AtomBeam Technologies Inc., Moraga, CA (US)
Filed on Feb. 8, 2025, as Appl. No. 19/048,904.
Application 19/048,904 is a continuation in part of application No. 19/014,442, filed on Jan. 9, 2025.
Application 19/014,442 is a continuation of application No. 18/791,425, filed on Aug. 1, 2024, granted, now 12,199,643, issued on Jan. 14, 2025.
Application 18/791,425 is a continuation in part of application No. 18/623,018, filed on Mar. 31, 2024, granted, now 12,119,848, issued on Oct. 15, 2024.
Int. Cl. H03M 7/00 (2006.01); H03M 7/30 (2006.01)
CPC H03M 7/3059 (2013.01) [H03M 7/3082 (2013.01); H03M 7/6005 (2013.01); H03M 7/6011 (2013.01)] 20 Claims
OG exemplary drawing
 
1. A system for distributed data compression, comprising:
at least one edge computing device comprising at least one processor and at least one memory;
at least one central computing device comprising at least one processor and at least one memory;
a lightweight compression subsystem comprising programming instructions stored in the at least one memory of the at least one edge computing device that, when operating on the at least one processor, cause the at least one edge computing device to:
receive input data;
preprocess and encode the input data into a partially compressed representation; and
transmit the partially compressed representation to the at least one central computing device;
a central compression subsystem comprising programming instructions stored in the at least one memory of the at least one central computing device that, when operating on the at least one processor, cause the at least one central computing device to:
receive the partially compressed representation from the lightweight compression subsystem;
further process the partially compressed representation using adjustable compression parameters and a temporal modeling component;
generate reconstructed data from the processed representation; and
optimize the preprocessing, encoding, and reconstruction operations based on one or more optimization criteria.