Searcharxiv⌕ Search

arXiv subjects

R. Eskandari

Publications and source records attributed to R. Eskandari.

5 recordsLinked to original sources
arXiv2024

Separated Pairs of Submodules in Hilbert $C^*$-modules

We introduce the notion of the separated pair of closed submodules in the setting of Hilbert $C^*$-modules. We demonstrate that even in the case of Hilbert spaces this concept has several nice characterizations enriching the theory of separated pairs of subspaces in Hilbert spaces. Let $\mathscr H$ and $\mathscr K$ be orthogonally complemented closed submodules of a Hilbert $C^*$-module $\mathscr E$. We establish that $ (\mathscr H,\mathscr K)$ is a separated pair in $\mathscr{E}$ if and only if there are idempotents $Π_1$ and $Π_2$ such that $Π_1Π_2=Π_2Π_1=0$ and $\mathscr R(Π_1)=\mathscr H$ and $\mathscr R(Π_2)=\mathscr K$. We show that $\mathscr R(Π_1+λΠ_2)$ is closed for each $λ\in \mathbb{C}$ if and only if $\mathscr R(Π_1+Π_2)$ is closed. We use the localization of Hilbert $C^*$-modules to define the angle between closed submodules. We prove that if $(\mathscr H^\perp,\mathscr K^\perp)$ is concordant, then $(\mathscr H^{\perp\perp},\mathscr K^{\perp\perp})$ is a separated pair if the cosine of this angle is less than one. We also present some surprising examples to illustrate our results.

math.OA↗
arXiv2021

Pedersen--Takesaki operator equation in Hilbert $C^*$-modules

We extend a work of Pedersen and Takesaki by giving some equivalent conditions for the existence of a positive solution of the so-called Pedersen--Takesaki operator equation $XHX=K$ in the setting of Hilbert $C^*$-modules. It is known that the Douglas lemma does not hold in the setting of Hilbert $C^*$-modules in its general form. In fact, if $\mathscr{E}$ is a Hilbert $C^*$-module and $A, B \in \mathcal{L}(\mathscr E)$, then the operator inequality $B B^*\le λAA^*$ with $λ>0$ does not ensure that the operator equation $AX=B$ has a solution, in general. We show that under a mild orthogonally complemented condition on the range of operators, $AX=B$ has a solution if and only if $BB^*\leq λAA^*$ and $\mathscr R(A) \supseteq \mathscr R(BB^*)$. Furthermore, we prove that if $\mathcal{L}(\mathscr E)$ is a $W^*$-algebra, $A,B\in \mathcal{L}(\mathscr E)$, and $\overline{\mathscr R(A^*)}=\mathscr E$, then $BB^*\leqλAA^*$ for some $λ>0$ if and only if $\mathscr R (B)\subseteq \mathscr R(A)$. Several examples are given to support the new findings.

math.OA↗
arXiv2021

Hilbert $C^*$-module independence

We introduce the notion of Hilbert $C^*$-module independence: Let $\mathscr{A}$ be a unital $C^*$-algebra and let $\mathscr{E}_i\subseteq \mathscr{E},\,\,i=1, 2$, be ternary subspaces of a Hilbert $\mathscr{A}$-module $\mathscr{E}$. Then $\mathscr{E}_1$ and $\mathscr{E}_2$ are said to be Hilbert $C^*$-module independent if there are positive constants $m$ and $M$ such that for every state $φ_i$ on $\langle \mathscr{E}_i,\mathscr{E}_i\rangle,\,\,i=1, 2$, there exists a state $φ$ on $\mathscr{A}$ such that \begin{align*} mφ_i(|x|)\leq φ(|x|) \leq Mφ_i(|x|^2)^{\frac{1}{2}},\qquad \mbox{for all~}x\in \mathscr{E}_i, i=1, 2. \end{align*} We show that it is a natural generalization of the notion of $C^*$-independence of $C^*$-algebras. Moreover, we demonstrate that even in case of $C^*$-algebras this concept of independence is new and has a nice characterization in terms of extensions. This enriches the theory of independence of $C^*$-algebras. We show that if $\langle \mathscr{E}_1,\mathscr{E}_1\rangle $ has the quasi extension property and $z\in \mathscr{E}_1\cap \mathscr{E}_2$ with $\|z\|=1$, then $|z|=1$. Several characterizations of Hilbert $C^*$-module independence and a new characterization of $C^*$-independence are given. One of characterizations states that if $z_0\in \mathscr{E}_1\cap \mathscr{E}_2$ is such that $\langle z_0,z_0\rangle=1$, then $\mathscr{E}_1$ and $\mathscr{E}_2$ are Hilbert $C^*$-module independent if and only if $\|\langle x,z_0\rangle\langle y,z_0\rangle\|=\|\langle x,z_0\rangle\|\,\|\langle y,z_0\rangle\|$ for all $x\in \mathscr{E}_1$ and $y\in \mathscr{E}_2$. We also provide some technical examples and counterexamples to illustrate our results.

math.OA↗
arXiv2019

Extensions of the Lax-Milgram theorem to Hilbert C*-modules

We present three versions of the Lax-Milgram theorem in the framework of Hilbert C*-modules, two for those over W*-algebras and one for those over C*-algebras of compact operators. It is remarkable that while the Riesz theorem is not valid for certain Hilbert C*-modules over C*-algebras of compact operators, our Lax-Milgram theorem turns out to be valid for all of them. We also give several examples to illustrate our results, in particular, we show that the main theorem is not true for Hilbert modules over arbitrary C*-algebras.

math.OA↗
arXiv2016

Operator equations $AX+YB=C$ and $AXA^*+BYB^*=C$ in Hilbert $C^*$-modules

Let $A,B$ and $C$ be adjointable operators on a Hilbert $C^*$-module $\mathscr{E}$. Giving a suitable version of the celebrated Douglas theorem in the context of Hilbert $C^*$-modules, we present the general solution of the equation $AX+YB=C$ when the ranges of $A,B$ and $C$ are not necessarily closed. We examine a result of Fillmore and Williams in the setting of Hilbert $C^*$-modules. Moreover, we obtain some necessary and sufficient conditions for existence of a solution for $AXA^*+BYB^*=C$. Finally, we deduce that there exist nonzero operators $X, Y\geq 0$ and $Z$ such that $AXA^*+BYB^*=CZ$, when $A, B$ and $C$ are given subject to some conditions.

math.OA↗