3
$\begingroup$

Let $E$ be a complex Hilbert space, with inner product $\langle\cdot\;| \;\cdot\rangle$ and the norm $\|\cdot\|$ and let $\mathcal{B}(E)$ the algebra of all bounded linear operators from $E$ to $E$.

I want to show that for $(T_1,...,T_d) \in \mathcal{B}(E)^d$ we have: $$\displaystyle\frac{1}{2}\left\|\displaystyle\sum_{k=1}^dT_k^*T_k \right\|^{1/2}\leq \omega(T_1,\cdots,T_d) \leq \left\|\displaystyle\sum_{k=1}^dT_k^*T_k \right\|^{1/2},$$ where $$\omega(T_1,\cdots,T_d)=\displaystyle\sup_{\|x\|=1}\bigg(\displaystyle\sum_{k=1}^d|\langle T_kx\;|\;x\rangle|^2\bigg)^{1/2}.$$

And you for you help.

$\endgroup$
  • 1
    $\begingroup$ You should probably add some context as to why you want to know this. Is this for a class, for a research project, etc. $\endgroup$ – Chris Ramsey Nov 7 '17 at 15:14
10
$\begingroup$

The inequality you have written cannot be proven.

Gelu Popescu has a big paper on multivariable generalizations of the numerical radius (Memoirs of the AMS, arXiv). He calls what you have above the Euclidian operator radius of a tuple: $$ \omega_e(T_1,\dots, T_d) = \sup_{\|x\|=1} \left(\sum_{k=1}^d |\langle T_kx,x\rangle|^2\right)^{1/2} $$ and he proves many things about this in section 2 of his paper. Moreover, Proposition 2.4 establishes that $$ \frac{1}{2\sqrt n} \|[T_1, \dots, T_d]\| \leq \omega_e(T_1,\dots, T_d) \leq \|[T_1, \dots, T_d]\| $$ and the inequality is sharp, noting that $\|[T_1, \dots, T_d]\| = \|\sum_{k=1}^d T_k^*T_k\|^{1/2}$. The proof is developed throughout section 2 but is not too difficult to follow.

$\endgroup$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.