Scribd
Upload
169 views

Fir Kaiser Window Notes-3 PDF

This document discusses the design of FIR filters using the Kaiser window. It describes some disadvantages of fixed windows and introduces the Kaiser window as an adjustable window that allows controlling the sidelobe level. It provides the equations for the Kaiser window and outlines the design procedure, which includes specifying the design parameters, determining the Kaiser beta parameter, calculating the filter length, and computing the modified impulse response.

Uploaded by

prajwal
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
169 views

Fir Kaiser Window Notes-3 PDF

This document discusses the design of FIR filters using the Kaiser window. It describes some disadvantages of fixed windows and introduces the Kaiser window as an adjustable window that allows controlling the sidelobe level. It provides the equations for the Kaiser window and outlines the design procedure, which includes specifying the design parameters, determining the Kaiser beta parameter, calculating the filter length, and computing the modified impulse response.

Uploaded by

prajwal
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 19

FIR FILTER DESIGN

FIR FILTER DESIGN USING KAISER


WINDOW
DISADVANTAGES OF THE FIXED WINDOWS
• The Rectangular, Hanning, Hamming, Bartlett, Blackman windows, are
called fixed windows. These windows provide a fixed stop band
attenuation.

• There exists a trade-off between main lobe width and the side lobe
amplitude.

• The main lobe width is inversely proportional to N.

• An increase in window length decreases the transition band of the filter.

• The minimum stop band attenuation is independent of N and is a function


of selected window.

• Thus, in order to achieve the minimum stop band attenuation and pass
band ripple, the designer must find a window with an appropriate side
lobe level and then choose N, to achieve the prescribed transition width.
DISADVANTAGES OF THE FIXED WINDOWS
• In this process, the designer may often settle with a window with
undesirable design specifications, as a window function with minimum
stop band attenuation has the maximum main lobe width, and filter
length must be increased unnecessarily, to reduce main lobe width and
to achieve the desired transition band.

• For this reason, it is necessary to use an ADJUSTABLE WINDOW, which


allows controlling the side lobe level w.r.t main lobe by varying some
parameter of the window (so that the trade off can be adjusted by the
designer). This is possible with one of the window functions called the
KAISER WINDOW.

• A desirable property of the window function is that the function is of finite


duration in time domain and the Fourier Transform has maximum energy
in the main lobe or a given peak side lobe amplitude.

• The prolate spheroidal functions have this desirable property. However,


these functions are complicated and difficult to compute.
KAISER WINDOW
• The causal Kaiser window is given by

 1

  22
 n −α   
 oI  β 1 −   
   α  
  ; 0 ≤ n ≤ N-1
wk (n) =   
 I o [ β ] 
 
 
 
where, I o ( x) is the Bessel function of the first kind and zero order.
2
∞  x  1 
k
0.25 x 2 (0.25 x 2 ) 2 (0.25 x 2 )3
I o ( x ) = 1 + ∑    = 1+ 2
+ 2
+ 2
+ ...
 2  k !
k =1  (1!) (2!) (3!)
KAISER WINDOW
• The parameter β controls the side lobe level w.r.t main lobe (i.e, the
degree of taper towards the edge of the window). If β = 0, there is no
taper and we get a rectangular window; if β = 5.44, the window is similar
to Hamming function. Other values of β offer the designer a whole range
of trade-offs between main lobe and side lobe performance. The value of
α is given by α = (N-1)/2.

• β is determined from the equation

0; α s < 21 
 
β = 0.5842(α s − 21)0.4 + 0.07886(α s + 21); 21<α s ≤ 50 
0.1102(α − 8.7); α > 50 
 s s 
DESIGN SPECIFICATIONS
• Filter type: LPF, HPF, BPF, BSF

• Pass band and Stop band frequencies in Hz


For Low-pass / High-pass:--- f p (or ω p ) and f s (or ωs )
For Band-pass / Band-stop:--- f p1 (or ω p1 ) and f p 2 (or ω p 2 )
f s1 (or ωs1 ) and f s 2 (or ωs 2 )
'
• Pass band ripple and minimum stop band attenuation α 'p and α s

• Sampling Frequency (Fs)

• Filter Length (N). Choose N to be odd for Type 1 FIR filter


DESIGN SPECIFICATIONS FOR LPF

(ω p + ωs )
ωc = and ∆ω = ωs − ω p
2
DESIGN SPECIFICATIONS FOR HPF

(ω p + ωs )
ωc = and ∆ω = ω p − ωs
2
DESIGN SPECIFICATIONS FOR BPF
DESIGN SPECIFICATIONS FOR BSF
DESIGN PROCEDURE
• Determine hd (n) for an ideal frequency response H(ω).

• Choose δ according to equations δ = min{δ p , δ s }


0.05 α 'p
10 −1 −0.05α s'
δp =
10
0.05 α 'p
+1
and δ s = 10
• Calculate α s = −20 log10 δ

• Determine parameter β from Kaiser’s design equation given above.

• Calculate filter Length (N) for the lowest odd value of N given by
 ( Fs )0.9222  
  + 1; α s ≤ 21 
 ∆F  
N ≥ 
 ( Fs )(α s − 7.95) 
+ 1; α s > 21 

 14.36∆F  

DESIGN PROCEDURE
or equivalently,

 5.794  
 ∆ω  + 1; α s ≤ 21 
  
N ≥ 
  (α − 7.95) 

s
 + 1; α s > 21 
 2.285∆ω  

• The modified impulse response is computed using


h(n) = wk (n)hd (n) ; |n|≤ (N-1)/2
DESIGN PROCEDURE
 ( N −1)/2
−n 
• The TF is given by H ( z ) = z − ( N −1)/2
 h (0) + 2 ∑ h ( n )( z n
+ z )
 n =1 

• The magnitude response is given by


( N −1)/2
H (ω ) = ∑
n=0
a (n) cos ω n

 N −1 
where, a(0)=h  
 2 
 N −1 
a ( n) = 2h  − n
 2 

You might also like