US 11,748,657 B2
Source identification by non-negative matrix factorization combined with semi-supervised clustering
Boian S. Alexandrov, Santa Fe, NM (US); Ludmil B. Alexandrov, Santa Fe, NM (US); Filip L. Iliev, Oakland, CA (US); Valentin G. Stanev, Crofton, MD (US); and Velimir V. Vesselinov, Santa Fe, NM (US)
Assigned to Triad National Security, LLC, Los Alamos, NM (US)
Filed by Triad National Security, LLC, Los Alamos, NM (US)
Filed on Sep. 14, 2020, as Appl. No. 17/20,504.
Application 17/020,504 is a continuation of application No. 15/690,176, filed on Aug. 29, 2017, granted, now 10,776,718.
Claims priority of provisional application 62/381,486, filed on Aug. 30, 2016.
Prior Publication US 2021/0004724 A1, Jan. 7, 2021
This patent is subject to a terminal disclaimer.
Int. Cl. G06N 20/00 (2019.01); G06F 17/16 (2006.01); G06F 18/2115 (2023.01); G06F 18/2133 (2023.01); G06F 18/2135 (2023.01); G06N 7/01 (2023.01)
CPC G06N 20/00 (2019.01) [G06F 17/16 (2013.01); G06F 18/2115 (2023.01); G06F 18/2133 (2023.01); G06F 18/2135 (2023.01); G06N 7/01 (2023.01)] 28 Claims
OG exemplary drawing
 
1. A method, comprising:
with a computer system:
performing a plurality of non-negative matrix factorization (NMF) trials on mixed signals generated by one or more unidentified sources and detected by a plurality of sensors, each of the NMF trials producing a predetermined number b of candidate sources referred to each of the sensors according to a signal propagation model;
for each of the NMF trials, ranking the corresponding b candidate sources according respective magnitudes of the corresponding b candidate sources; and
determining b basis sources for the mixed signals by:
for each index d from 1 to b, determining the dth basis source as an average, across the plurality of NMF trials, of the candidate sources having dth rank.