I am testing my supervised NMF algorithm to extract signal from observation that have only one source inside. I am new here and I wonder this is very weak model or not? Is it acceptable in signal processing academic paper to start experimental result with factorization of only one source separation from the environment?
-
1$\begingroup$ I'd say that this is a very good starting point since there is nothing simpler that makes sense. By doing so you will already tackle enough problems. $\endgroup$– jojeck ♦Nov 1, 2017 at 11:25
-
$\begingroup$ @jojek thanks so much. Just want to ask that do you know some papers doing source separation on one source? May you please introduce them? $\endgroup$– JanNov 1, 2017 at 11:31
-
1$\begingroup$ Have you seen these two? 1, 2. Then it's just a matter of tracking the references $\endgroup$– jojeck ♦Nov 1, 2017 at 11:38
-
1$\begingroup$ You can also have a look at a recent answer, it might help you somehow. $\endgroup$– jojeck ♦Mar 27, 2018 at 10:24
1 Answer
Yes, that's a first sanity check. Basically, the idea of averaging multiple observations of a single signal buried in noise is a basic issue in signal processing. You can have several observations with multiple noise realizations and statistics, with different delays on the signal, with different data transformations (more or less linear).
If this does not work in your case, don't expect to much of NMF performance on multiple sources, and verify assumptions or test other source separation methods.