Professional Documents
Culture Documents
Motivation
Some signals obviously have spectral characteristics that vary with time
time, t
time, t
Wavelet Transform
time-series
Wavelet Transform
time-series
Im going to ignore the complex conjugate from now on, assuming that were using real wavelets
Wavelet
normalization
shift in time
change in scale: big s means long wavelength wavelet with scale, s and time, t Mother wavelet
mother wavelet
t=5, s=2
time
frequency, w
Thus determining the wavelet coefficients at a fixed scale, s can be thought of as a filtering operation
g(s,t) = f(t) Y[(t-t)/s] dt = f(t) * Y(-t/s) where the filter Y(-t/s) is has a band-limited spectrum, so the filtering operation is a bandpass filter
Suppose the first n moments are zero (called the approximation order of the wavelet), then it can be shown that g(s,t)sn+2. So some effort has been put into finding wavelets with high approximation order.
tj,k = 2jkDt
Then g(sj,tj,k) = gjk
dyadic grid
The factor of two scaling means that the spectra of the wavelets divide up the frequency scale into octaves (frequency doubling intervals)
w
1/ 8wny
wny
wny
wny
As we showed previously, the coefficients of Y1 is just the band-passes filtered time-series, where Y1 is the wavelet, now viewed as a bandpass filter.
w
1/ 8wny
wny
wny
wny w
with
low-pass filter
wny
wny
time-series of length N HP LP
2
g(s1,t) HP 2 g(s2,t)
2
LP 2
HP
2
LP
2
Total: N coefficients
g(s3,t)
time, t
time, t
From http://en.wikipedia.org/wiki/Coiflet
Spectrum of wavelet
frequency, w
frequency, w
time-series
stage 1 - hi
stage 1 - lo
Stage 1 lo
stage 2 - hi
stage 2 - lo
Stage 2 lo
stage 3 - hi
stage 3 - lo
Stage 3 lo
stage 4 - hi
stage 4 - lo
Stage 4 lo
stage 5 - hi
stage 6 - lo
Stage 4 lo
stage 5 - hi
stage 6 - lo
Had enough?
short wavelengths
|g(sj,t)|2
long wavelengths
scale
time, t
stage 1 - hi
stage 1 - lo
short wavelengths
long wavelengths
scale
time, t