Adaptive Noise Cancellation: A Panacea M.Evans, R. Adhikari

The Concept 5000 Channels 1 GW Channel Signal = DARM - noise

Wiener Filter Input Signal (PEM) Covariance Matrix Norbert Wiener, MIT Block Toeplitz Cross Correlation Matrix

Wiener Filter Sim H1 & L1 DARM (S5) All Guralp & STS2 Vault not used Requires ~9 GB RAM Levinson-Durbin H1 L1 x10 x7

Adaptive Filtering / Neural Network Keenan Pepper Elena Gasparri LMS Algorithm (1st chapter of textbooks)

ATNF

Filtered-X LMS Edwards, Thesis (declassified) Naval Research Postgraduate School

Applications Seismic Noise ( Hz) Upconversion Lock Acquisition Bilinear (Crab shoulder) Acoustic Noise ( Hz) Magnetic (power lines) H1/H2 correlation MICH/PRC/SRC

From LMS to Filtered-X LMS FIR Adapt Noise Witness Noisy Signal Cleaner Signal

From LMS to Filtered-X Injecting into a high gain servo loop FIR Adapt Noise Witness Total Control Signal Servo Control Signal G

From LMS to Filtered-X Separated Actuators Special case of “Filtered X- LMS” FIR Adapt Noise Witness Total Control Signal Servo Control Signal A1 G A2 C

At the 40m Test on input mode- cleaner FIR Adapt Seismic Sensors MC Length MC_L MC2 G MC1 C

At the 40m

And more !