next up previous contents
Next: Impulse Response of LTI Up: Impulses Previous: Discrete Time Impulses: The

Continuous Time Impulses: The Unit Impulse

The continuous time unit impulse is a continuous time signal tex2html_wrap_inline1051 which is zero for all time tex2html_wrap_inline737 , but has the value tex2html_wrap_inline1103 at time tex2html_wrap_inline1105 , such that the "spike has an area of one," roughly speaking. More specifically, an approximation to tex2html_wrap_inline1051 is the signal tex2html_wrap_inline1109 defined as

displaymath1095

where tex2html_wrap_inline1113 is a parameter. Note that if tex2html_wrap_inline1115 is small, tex2html_wrap_inline1109 is simply a a tall spike of height tex2html_wrap_inline1119 and width tex2html_wrap_inline1115 centered at tex2html_wrap_inline1105 , and thus it has an area of one. As tex2html_wrap_inline1115 approaches zero from above, tex2html_wrap_inline1109 approaches tex2html_wrap_inline1051 , in a certain sense.

Given an arbitrary signal tex2html_wrap_inline721 , we can form the "staircase" approximation to tex2html_wrap_inline721 :

displaymath1096

This approximating signal tex2html_wrap_inline1135 is piecewise constant over intervals of length tex2html_wrap_inline1115 and agrees with tex2html_wrap_inline721 at all times tex2html_wrap_inline737 which are multiples of tex2html_wrap_inline1115 . You may recall that staircase approximations are used in basic integration theory.

As tex2html_wrap_inline1115 approaches zero, the right side of the equation above becomes an integral and the approximation to tex2html_wrap_inline721 becomes exact, so we obtain

eqnarray173

We call the first equation above the representation of tex2html_wrap_inline721 as a superposition of impulses. The second equation above is sometimes called the sifting property ( tex2html_wrap_inline721 is "sifted" at time tex2html_wrap_inline737 to obtain the value of tex2html_wrap_inline1091 at time tex2html_wrap_inline737 .) This is actually the formal definition of the unit impulse. Obviously, these two equations are really the same thing.



R. L. Cruz
Fri Dec 25 20:53:17 PST 1998