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

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

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: I I I .J:r -20 -30 I i 0.05 0.01 ( u=-2) ~ I'-t- II . .-- _~rr . ,, ! " ............. I 0.1 I! 0.5 ~I I 1i I I! ! I ; I lea ! jI I ; , il 1 W- l~ r-J 00=1 (u 0 0_ II !i 100 50 10 (u = 0) ( u=-I) .. -, " 1I I I' !I II I I. , ' .-., j = I) (u = 2) (7.13) T he t erm 2 0Iog(KaIa2/bIb3) is a c onstant. We observe t hat t he log amplitude is a s um of four basic terms corresponding t o (i) a constant, (ii) a pole or zero a t t he origin (201og Ijwl)' (iii) a first-order pole or zero (20 log 11 + j w/all, a nd (iv) complex c onjugate poles o r zeros (20 log 11 + jwb2/b3 + (jw)2 /b31l· We can sketch these four basic t erms as functions of w and use t hem t o c onstruct t he log-amplitude plot of a ny d esired transfer function. Let us discuss each of t he terms. , 1 50"1==+=+=+++=I+I1C==+=+==l=rnr:m==+==+=+=mm==+=+=::r=m:m II . ..... I: II II 90"b_ . . _~.~~-;-1-;j;;m=±±hmfh-:dddd~!db=b!-::-:k~:Hj __ -i 1I --r 't ; T 11"--~ 5 0"·---+-+-I--HH+i+--- - (b ~ -,- --- -r ------ --- -- Il I !I r---+i~-+~rH~---+--~-+++H+--~--+-~+++H----~+-+4~++~---0 0_ I ~ - 50"f--·........... - --.. ~-t-L+ ~ 0• • "._ I + j I f -~"~.~.~.~+=~+=66~~~~~~~~~~~~~~~~~db~~~~~ .. - -. -I 1"( iI i! .. _ISO" ! ---..... , 0.05 0.01 ( u=-2) 0.1 0.5 ( u=-I) 00=1 ( u=O) + :ft·= 10 i --!- 50 ( u= I ) 100 (u = 2) Fig. 7 .3 Amplitude and phase response of a pole or a zero a t the origin. T he log a mplitude of this t erm is also a constant, 20 l og(K aIa2/bIb3). T he p hase c ontribution from this t erm is zero. 2. P ole (or Z ero) a t t he origin Log M agnitude t This p oint c an b e s hown as follows: Let Wj a nd W2 a long t he w-scale correspond t o Uj a nd along t he u ·scale. T hen l ogwj = Uj, a nd logw2 = U 2. T hen U 2 - Uj = loglO W 2 - logJQwj = logJQ(w2/Wj) T hus, if (w2/wIJ = 10 (which is a decade) t hen U 2 - Uj = loglO 10 = 1 a nd if (W2/WIJ 2 (which is a n o ctave) t hen U 2 - Uj = loglO 2 = 0.3010 = Such a pole gives rise t o t he t erm - 20 log Ijwl, which can be expressed as - 20logljwl = - 2010gw U2 7 Frequency Response and Analog Filters 4BO Note t hat equal increments in u are equivalent t o equal ratios on the w-scale. Thus, one unit along t he u-scale is t he same as one decade along the w-scale. This means t hat t he a mplitude plot has a slope of - 20 d B/decade or - 20(0.3010) = - 6.02 d B/octave ( commonly s tated as 6 dB/octave). Moreover, the amplitude plot crosses t he w-axis a t w = 1, since u = 10glQw = 0 when w = 1. For t he c ase of a zero a t t he origin, t he log-amplitude term is 20 log w. T his is a straight line passing through w = 1 a nd having a slope of 20 d B/decade (or 6 dB/octave). T his plot is a mirror image about t he w-axis of the plot for a pole a t t he origin a nd is shown dotted in Fig. 7.3a. Phase 7.2 Bode P lots 481 18 12 6 o - 12 T he phase function corresponding to the pole a t t he origin is - Ljw [see Eq. (7.12b)]. Thus -18 (7.15b) L H(jw) = - Ljw = - 90° O .Ola T he phase is c onstant ( - 90°) for all values of w , as depicted in Fig....
View Full Document

This note was uploaded on 04/14/2013 for the course ENG 350 taught by Professor Bayliss during the Spring '13 term at Northwestern.

Ask a homework question - tutors are online