DESIGN-RLS - RLS Algorithm : Guidelines Memory parameter...

Info iconThis preview shows page 1. Sign up to view the full content.

View Full Document Right Arrow Icon
RLS Algorithm: Design guidelines RLS Algorithm: Design guidelines ± For high SNR (> 30 dB), stationary environments present information reliable : choose λ 1. ± High SNR, slow non-stationary environment choose λ small & weight present observations more. ± For low SNR (< 10 dB), present observations unreliable & prior information better: choose λ large & de-emphasize noisy observations. RLS Algorithm : Guidelines
Background image of page 1
This is the end of the preview. Sign up to access the rest of the document.

Unformatted text preview: RLS Algorithm : Guidelines Memory parameter provides trade-off between time/frequency resolution. For low SNR, initialize inverse recursion with large matrix norm: P[0] = I, large. For high SNR, initialize the inverse recursion with small matrix norm: P[0] = I, small. If environment is highly non-stationary retraining needed....
View Full Document

This note was uploaded on 12/02/2011 for the course AR 107 taught by Professor Gracegraham during the Fall '11 term at Montgomery College.

Ask a homework question - tutors are online