This preview shows pages 1–5. Sign up to view the full content.
This preview has intentionally blurred sections. Sign up to view the full version.
View Full DocumentThis preview has intentionally blurred sections. Sign up to view the full version.
View Full Document
Unformatted text preview: EE362DSP I University of Saskatchewan a39 a38 Session 4 Frequency Response, Eigenfunctions, Suddenly Applied Complex Exponentials Brian Daku Page 51 EE362DSP I University of Saskatchewan a39 a38 Black Box Filter A black box LTI filter has been generated in MATLAB. The question is: What type of filter is in this black box? The filter is implemented as a MATLAB function, which is in the file bbfilter.p (this file is on the website). Given the input x the output of the filter can be generated using y=bbfilter(x); , as shown in the following figure. x y bbfilter(.) What approach should be used to determine the type of filter in the black box? Brian Daku Page 52 EE362DSP I University of Saskatchewan a39 a38 Magnitude/Phase Response of a LTI System If the input to a LTI system is a sinusoid, the output is also a sinusoid, having the same frequency, but the magnitude and phase could be altered. x [ n ] = cos( n ) y [ n ] = a cos( n + ) (3) Expanding the input and output gives x [ n ] = cos( n ) = 1 2 e jn + 1 2 e jn y [ n ] = a cos( n + ) = a 2 e j ( n + ) + a 2 e j ( n + ) Since this is a LTI system, if the input is x 1 [ n ] = e jn then the output will be y 1 [ n ] = ae j ( n + ) = ae jn e j The frequency term can be removed by multplying the output term by the complex conjugate, ( e jn ) = e jn , z 1 [ n ] = y 1 [ n ] e jn = ae jn e j e jn = ae j Thus  z 1  = a and z 1 = at frequency . e jn is referred to as an eigenfunction for the LTI system. Using this eigenfunction, the magnitude and phase for each frequency can determined (ie a frequency response). Example 4.1: Determine the magnitude and phase of the filter in bbfilter.p at the frequency f = 0 . 4 cycles/sample. Brian Daku Page 53 EE362DSP I University of Saskatchewan a39 a38 What are Eigenfunctions for LTI Systems? In general, any function f [ n ] is an eigenfunction if it satisfies the relationship, y [ n ] = T { f [ n ] } = f [ n ] , (4) where is the eigenvalue for the system, which does not depend on the time variable n . If the system is LTI with impulse response h [ n ] then equation (4) simplifies to, y [ n ] = f [ n ] h [ n ] = f [ n ] . Example 4.2: Is e jn (for < n < ) an eigenfunction of a LTI system with impulse response h [ n ] ? y [ n ] = h [ n ] e jn = summationdisplay k = h [ k ] e j ( n k ) (5) = parenleftBigg summationdisplay k = h [ k ] e jk parenrightBigg e jn = ( ) e jn (6) Yes, e jn is an eigenfunction, since y [ n ] is the product of the eigenvalue ( ) , which does not depend on...
View
Full
Document
 Fall '09
 Harris,J
 Frequency

Click to edit the document details