This preview shows page 1. Sign up to view the full content.
Unformatted text preview: m ultipleinput,
m Ultipleoutput ( MIMO) systems.
A system's o utput for t ? 0 is t he r esult of two independent causes: t he initial
conditions of t he system (or t he s ystem s tate) a t t = 0 a nd t he i nput J (t) for t ? O.
I f a s ystem is t o b e linear, t he o utput m ust be t he sum of t he two components
resulting from t hese two causes: first, t he z eroinput r esponse c omponent t hat
results only from t he initial conditions a t t = 0 w ith t he i nput J (t) = 0 for t ? 0,
a nd t hen t he z erostate r esponse c omponent t hat results only from t he i nput J (t)
for t ? 0 when t he initial conditions ( at t = 0) a re assumed to be zero. When all
t he a ppropriate initial conditions are zero, t he s ystem is said t o b e in z ero s tate.
T he s ystem o utput is zero when the i nput is zero only if the system is in zero state.
In s ummary, a linear system response can be expressed as t he s um of a zeroinput a nd a z erostate component:
T otal r esponse = z eroinput r esponse + z erostate r esponse (1.41) This p roperty of linear systems which permits t he s eparation of a n o utput into
components resulting from t he initial conditions a nd from t he i nput is called t he
d ecomposition p roperty.
For t he R C circuit of Fig. 1.26, t he response y (t) was found t o b e [see Eq.
(1.36c)]
y (t) = v dO)  zi c omponent + RJ(t) , + 1 % 8 it . J(r)dr co, and
dY2
d t + 3Y2(t) = h (t) Multiplying the first equation by k l, the second with k2, and adding them yields But this equation is the system equation [Eq. (1.43)J with and
yet) Therefore, when the input is k dl(t) + k2!2(t), the system response is k1Yl(t) + k2Y2(t).
Consequently, the system is linear. Using this argument, we can readily generalize the
result to show t hat a system described by a differential equation of the form
(1.44)
is a linear system. The coefficients a i and bi in this equation can be constants or functions
of time . •
6 E xercise E 1.12 Show that the system described by the following equation is linear: (1.42) c omponent From Eq. (1.42), it is clear t hat if t he i nput J (t) = 0 for t ? 0, t he o utput y (t) =
v dO). Hence v c(O) is t he zeroinput component of t he response y (t). Similarly, if
t he s ystem s tate ( the voltage v c in this case) is zero a t t = 0, t he o utput is given
by t he second component on t he r ighthand side of Eq. (1.42). Clearly t his is t he
z erostate c omponent of t he response y (t).
I n addition to t he decomposition property, linearity implies t hat b oth t he zeroinput a nd z erostate components must obey t he principle of superposition with
respect t o each of their respective causes. For example, if we increase t he i nitial
condition kfold, t he zeroinput component must also increase kfold. Similarly, if
we increase t he i nput kfold, t he z erostate component must also increase kfold. = k lYl(t) + k2Y2(t) 6 E xercise E 1.13 ~ + t 2 y(t) = (2t + 3 )J(t)
dt 'V Show that a system described by the following equation is nonlinear: y(t)~ + 3 y(t) =
dt J (t...
View
Full
Document
 Spring '13
 Bayliss
 Signal Processing, The Land

Click to edit the document details