(iii)→(ii)
Suppose that ∣∣T(x)∣∣=∣∣x∣∣ for all x∈V.
Now for any x,y∈V:<x,y>=41∑k=14ik∣∣x+iky∣∣2
if F=C,
And if F=R ,
we have similar result: <x,y>=41∑k=14(∣∣x+y∣∣2−∣∣x−y∣∣2)
Using the above relation for the complex version, we have
<T(x),T(y)>=41∑k=14ik∣∣T(x)+ikT(y)∣∣2=41∑k=14ik∣∣T(x+iky)∣∣2=41∑k=14ik∣∣x+iky∣∣2=<x,y>
(ii)→(iii)
Suppose that <T(x),T(y)>=<x,y>∀x,y∈V.
Substituting y=x, we have the desired result.
(ii)→(i)
Suppose T is unitary.
Then T has an inverse T−1.
So, for x,y∈V ,
<T(x),y>=<T(x),TT−1(y)>=<x,T−1(y)> .
So, T∗ exists and T∗=T−1
⇒TT∗=I
(i)→(ii)
Assume the adjoint T∗ exists
and TT∗=T∗T=I
Now for x,y∈V, we have <T(x),T(y))=<x,T∗T(y)>=<x,y>
Therefore from above discussion, we can conclude that,
(iii)⟺(ii)and(i)⟺(ii)
So that implies (i)⟺(ii)⟺(iii)
∴ (i),(ii),(iii)(i),(ii),(iii) are equivalent.
Comments
Leave a comment