Lectremez PDF
Lectremez PDF
Lectremez PDF
-80
-100
0 2000 4000 6000 8000 10000
with L = M / 2
L
A( e j
) = h[0 ] + 2 h[ n ]cos( n )
n =1
k =0
O O Passband Ripple
O
O O
Maximum Boundary Points
error
Stopband Ripple
O O O
0
O O
L=7th order polynomial, number of extrema = 10
26/03/2003 ELEC4600/7602 Signal and Image Processing II Brian Lovell 8
The Method
We know where the boundary points are from the
band edge specifications. At least 3 of these points
must be extrema.
We know how many local extrema there are from the
estimated filter length (harris formula or similar) but
we dont know their positions.
Guess the positions of the extrema are evenly
spaced in the pass and stop bands.
Perform polynomial interpolation and reestimate
positions of local extrema.
Move extrema to new positions and iterate until the
extrema stop shifting.
26/03/2003 ELEC4600/7602 Signal and Image Processing II Brian Lovell 9
Comments
Given the positions of the extrema, there exists a
formula for the optimum . owever we dont know
the optimum nor the exact positions of the extrema.
Thus we need to iterate. Assume the positions of the
extrema, calculate , move the extrema, recalculate
, until stops changing.
The algorithm generally converges in about 12
iterations.
O O O
0
O O
L=7th order polynomial, number of extrema = 10
26/03/2003 ELEC4600/7602 Signal and Image Processing II Brian Lovell 11
Pole-Zero Diagram
X1/a a
H ( z ) = z +1 z 1
H ( f ) = e + j 2 f e j 2 f
= 2 j sin(2f )
which for small f is approximately
H ( f ) j 4f
0.3
0.2
0.1
-0.1
-0.2
-0.3
-0.4
0 5 10 15 20
x (t ) x$(t )
h(t)
h(t)
h( f ) = j sgn( f )
Im f
h(n)
2/
Ideal Impulse
response 2/3 2/5
2/7
n