Signal Processing and Linear Systems-B.P.Lathi copy

56b these are strikingly simple transformations which

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

View Full Document Right Arrow Icon
This is the end of the preview. Sign up to access the rest of the document.

Unformatted text preview: 75 w ith Op = - 2 d B a nd Os = - 11 d B. T he v alue o f n ( order o f t he filter) is found from E q. (7.39): %E lliptic wp' = 0.3753 n= [ n,Wn)=ellipord(Wp,Ws,-Gp,-Gs); [ b,a)=ellip(n,-Gp,-Gs,Wn) log[(10 Ll - 1)/(10°·2 - 1)] = 1.9405 210g(0.7975/0.3753) % P lotting A mplitude a nd P hase R esponse W =O:.OOl:pi;W = W'; H =freqz(b,a,W); w =W/T; m ag=abs(H); p hase=180 / pi*unwrap( a ngle(H»; s ubplot(2,1,1); p lot(w,mag);grid; s ubplot(2,1,2); p lot(w,phase);grid W e r ound u p t he v alue of n t o 2. Also from Eq. (7.41), we find t he 3 -dB c utoff frequency we' as , We = 0.7975 1 = 0 .4322 ( lOLl - 1)4 F rom T able 7.1, we find t he p rewarped filter t ransfer f unction Ha(s) for n = 2 a nd we' = 0.4322 as Ha(s) = 0.1868 1 --......,.-2------ (0.4~22) + V2 (0.4~22) + 1 S2 + 0.6112s + 0.1868 MATLAB returns b=O . 1039 0 .2078 0.1039 a nd a=1 - 0.9045 0 .3201 for Butterworth option. Therefore Hz _ [ I- F inally, we o btain H[z] from Ha(s), u sing t he s implified b ilinear t ransformation i n E q. ( 12.64b): Z2 - 0 .1039(z + 1)2 0.9045z + 0.3201 a result, which agrees with the answer found in Example 12.7. z -1 z+1 0 s=-- T herefore 0.1868 {:; I Z2 0 .1039(z + 1)2 + 0.904z + 0.3201 E xercise E 12.5 D esign a first-order lowpass B utterworth f ilter using t he p rewarping m ethod so t hat t he a nalog a nd d igital gains a re i dentical a t w = 0 a nd a t t he 3 -dB c utoff frequency W e' U se T = n j4we . Answer: H[z] = 0 .8284(z + 1) = 0.2929(z + 1) 2 .8284z - 1.1716 z - 0.4142 - -----------------~-------- 1 750 12 F requency R esponse a nd D igital F ilters 12.6 751 R ecursive F ilter d esign: T he B ilinear T ransformation M ethod 0.982614 'Jt p (8) = 82 (12.65) + 1.09788 + 1.1025 Next, t o o btain t he desired highpass transfer function, we replace 8 w ith wp' / 8 i n t he above p rototype t ransfer f unction [see Eq. (7.55)J. T o o btain t he d esired digital transfer function H[z], we t hen r eplace 8 with ;+~ [the bilinear transformation in Eq. (12.64b)J. T his t wo-step o peration m ay b e c ombined in a single-step transformation as 0.7 wp ' s=---= ( Z-I) z +1 0.3 (12.66) In this case wp ' = 0.24 so t hat we replace s w ith 0.2!~:1) i n t he p rototype t ransfer function in Eq. (12.65) t o o btain t he d esired digital transfer function (a) o w P'(z+l) ( z - 1) 0.6902(z - 1)2 H[zJ = z2 _ 1.4678z + 0.6298 \ 0 15 F ig. 1 2.15 C hebyshev highpass filter design using bilinear transformation w ith p rewarping m ethod. • E xample 1 2.7: H ighpass F ilter D esign Design a I -dB ripple Chebyshev highpass filter w ith t he following specifications (depicted b y t he brick walls in Fig. 12.15a): T he s topband g ain Gs ~ - 6.3 dB (G. ::; 0.484) o ver t he s topband 0 ~ w ::; 10 (ws = 10). T he r ipple f ::; 1 d B (G p ~ 0.891) over a p assband w ~ 15 (w p = 15). T he h ighest frequency t o b e p rocessed is W h = 80 radians/so I n o rder t o select a suitable value o f T , we use Eq. (12.58) t o avoid signal aliasing: =- Let us choose T he c ritical frequencies a re W s = 10 a nd a ccording t o E q. (12.64a), are w.' = t an wp' = t an wp T= ~ 100 = 15. T he p rewarped critical frequencies, 7.6-1 T he value o f n needed t o s atis...
View Full Document

Ask a homework question - tutors are online