40m QIL Cryo_Lab CTN SUS_Lab TCS_Lab OMC_Lab CRIME_Lab FEA ENG_Labs OptContFac Mariner WBEEShop
  40m Log  Not logged in ELOG logo
Entry  Mon Dec 14 16:46:42 2015, ericq, Update, WienerFiltering, Noise Subtraction Puzzler subpuzz.pdf
    Reply  Thu Jan 7 15:29:54 2016, ericq, Update, WienerFiltering, Noise Subtraction Puzzler subpuzz2.pdfpuzzle.zip
       Reply  Thu Jan 14 15:08:37 2016, ericq, Update, WienerFiltering, Noise Subtraction Puzzler subpuzz_resolved.pdf
          Reply  Thu Jan 14 18:41:36 2016, rana, Update, WienerFiltering, Noise Subtraction Puzzler 
Message ID: 11918     Entry time: Thu Jan 7 15:29:54 2016     In reply to: 11880     Reply to this: 11933
Author: ericq 
Type: Update 
Category: WienerFiltering 
Subject: Noise Subtraction Puzzler 

The puzzle continues...

I found some reference for computing "multicoherence," which should properly estimate the potential MISO subtraction potential in situations where the witness channels themselves have nontrivial coherence. Specifically, I followed the derivations in LIGO-P990002. The underlying math is related to principal component analysis (PCA) or gram-schmidt orthogonalization. 

This produced the following results, wherein the Wiener subtraction is still below what the coherences predict. 

I've attached the data and code that produced this plot. 

Attachment 1: subpuzz2.pdf  24 kB  | Hide | Hide all
subpuzz2.pdf
Attachment 2: puzzle.zip  869 kB
ELOG V3.1.3-