Few maps in the rich structure for the domains G E ( 3 , 3 , 1 , 1 , 1 ) G_{E(3;3;1,1,1)} and G E ( 3 , 2 , 1 , 2 ) G_{E(3;2;1,2)}
Dinesh Kumar Keshari, Shubhankar Mandal and Avijit Pal · 2025
Read Few maps in the rich structure for the domains G E ( 3 , 3 , 1 , 1 , 1 ) G_{E(3;3;1,1,1)} and G E ( 3 , 2 , 1 , 2 ) G_{E(3;2;1,2)} by Dinesh Kumar Keshari, Shubhankar Mandal and Avijit Pal (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2511-01648.txt.
Introduction
Dinesh Kumar Keshari, Shubhankar Mandal, and Avijit Pal work in multivariable operator theory, on two naturally occurring domains associated with 3-by-3 matrix problems. A rich structure is a package that links four natural objects of analysis for such domains. Joined with the classical realization formula and Agler-style Hilbert space models, it lets one build analytic maps into the domain effectively.
The authors describe several maps inside this structure for the two domains named in the title. They define an SE map between two Schur-type classes. They relate one class to analytic kernels on the tridisc. Then they construct Upper W, Upper E, Right S, and SE maps through a UW procedure. A payoff is that interpolation problems for these domains reduce to a standard matricial Nevanlinna-Pick problem, with direct use for solvability criteria in mu-synthesis.
Catalog record
| CREATOR | Dinesh Kumar Keshari, Shubhankar Mandal and Avijit Pal |
|---|---|
| YEAR | 2025 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 71 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2511-01648
Browse the full library · About the Tower of Babel
Full text
The opening of the work, from the library's hosted text:
Few maps in the rich structure for the domains G E ( 3 , 3 , 1 , 1 , 1 ) G_{E(3;3;1,1,1)} and G E ( 3 , 2 , 1 , 2 ) G_{E(3;2;1,2)} Dinesh Kumar Keshari, Shubhankar Mandal and Avijit Pal Abstract . The primary goal of a rich structure for some naturally occurring domains 𝒳 \mathcal{X} is to connect four naturally occurring objects of analysis in the context of 3 × 3 3\times 3 analytic matrix functions on 𝔻 \mathbb{D} . Combining this rich structure with the classical realisation formula and Hilbert space models in the sense of Agler, one can effectively construct functions in the space 𝒪 ( 𝔻 , 𝒳 ) \mathcal{O}(\mathbb{D},\mathcal{X}) of analytic maps from 𝔻 \mathbb{D} to 𝒳 \mathcal{X} . This allows one to obtain solvability criteria for two cases of the μ \mu -synthesis problem. We describe few maps in the rich structure. We define S E SE map between 𝒮 1 ( ℂ 3 , ℂ 3 ) \mathcal{S}_{1}(\mathbb{C}^{3},\mathbb{C}^{3}) and 𝒮 3 ( ℂ , ℂ ) \mathcal{S}_{3}(\mathbb{C},\mathbb{C}) and establish the relation between 𝒮 1 ( ℂ 3 , ℂ 3 ) \mathcal{S}_{1}(\mathbb{C}^{3},\mathbb{C}^{3}) and the set of analytic kernels on 𝔻 3 \mathbb{D}^{3} . We obtain the U W UW procedure and using the U W UW procedure we construct the U p p e r W Upper\,\,W and U p p e r E Upper\,\ E maps. We also construct R i g h t S Right~S and S E SE maps. We show how the interpolation problems for G E ( 3 , 3 , 1 , 1 , 1 ) G_{E(3;3;1,1,1)} and G E ( 3 , 2 , 1 , 2 ) G_{E(3;2;1,2)} can be reduced to a standard matricial Nevanlinna-Pick problem. 2010 Mathematics Subject Classification 30C80, 32A38, 47A13, 47A48, 47A56. 1. Introduction Let \Omega_{1},\Omega_{2}\subset\mathbb{C}^{n} be domains in \mathbb{C}^{n}. Suppose \mathcal{O}\left(\Omega_{1},\Omega_{2}\right) is used to represent the set of all holomorphic functions from \Omega_{1} to \Omega_{2} . We discuss the classic Nevanlinna-Pick problem (see for instance [ 9 , 10 ] ). The Nevanlinna-Pick problem: Assume that z_{1},\ldots,z_{n}\in\mathbb{D} and \lambda_{1},\ldots,\lambda_{n}\in\mathbb{D} , whether there is an analytic function F:\mathbb{D}\rightarrow\mathbb{C} such that F(z_{i})=\lambda_{i} , i=1,\dots,n and |F(z)|\leq 1 for all z\in\mathbb{D} ? A necessary and sufficient condition to solve this problem was found by G.Pick in 1916. He showed that this problem is solvable if and only if the following pick matrix {\left(\left(\frac{1-\bar{\lambda_{i}}\lambda_{j}}{1-\bar{z_{i}}z_{j}}\right)\right)_{i,j=1}^{n}} is positive semi-definite. Unaware of Pick’s work, R. Nevanlinna also considered the problem in 1919. Later, while dealing with robust stabilization problem John Doyle introduced the \mu -synthesis problem (see for instance [ 25 , 26 , 27 , 28 ] ). The \mu -synthesis problem is an interpolation problem for analytic matrix functions on the disc which are subject to a boundedness condition. It is a generalisation of the Nevanlinna- Pick problem. Let \mathcal{M}_{n\times n}(\mathbb{C}) be the set of all n\times n complex matrices and E be a linear subspace of \mathcal{M}_{n\times n}(\mathbb{C}). We define the structured singular value (see [ 27 ] ) \mu_{E}:\mathcal{M}_{n\times n}(\mathbb{C})\to[0,\infty) as follows: \mu_{E}(A):=\frac{1}{\inf\{\|X\|:\,\ \det(1-AX)=0,\,\,X\in E\}},\;\;A\in\mathcal{M}_{n\times n}(\mathbb{C}) (1.1) with the understanding that \mu_{E}(A):=0 if 1-AX is nonsingular for all X\in E [ 25 , 26 ] . Here \|\cdot\| denotes the operator norm. The \mu_{E} -synthesis problem (see [ 6 ] ) : Given distinct points \lambda_{j}\in\mathbb{D} and W_{j}\in\mathcal{M}_{n\times n}(\mathbb{C}) , j=1,\dots,k . Construct, if possible an analytic n\times n matrix-valued function F on \mathbb{D} such that F(\lambda_{j})=W_{j} for j=1,\dots,k and \mu_{E}(F(\lambda))\leq 1 for all \lambda\in\mathbb{D} . Note that for n=m=1 , the \mu_{E} -synthesis problem is the classical Nevanlinna-Pick problem. Moreover, in the case n=m and E=\{c\mathbb{I}_{n}:c\in\mathbb{C}\} , \mu_{E} is the spectral radius r and the \mu_{E} -synthesis problem is the spectral Nevanlinna-Pick problem (see [ 6 ] ). The spectral Nevanlinna-Pick problem(*): Given distinct points \lambda_{j}\in\mathbb{D} and W_{j}\in\mathcal{M}_{k\times k}(\mathbb{C}) , j=1,\dots,n . Construct, if possible an analytic k\times k matrix-valued function F on \mathbb{D} such that F(\lambda_{j})=W_{j} for j=1,\dots,n and r(F(\lambda))\leq 1 for all \lambda\in\mathbb{D} . Here r is the spectral radius. In the attempt to solve the two-by-two spectral Nevanlinna-Pick problem, Agler and Young introduced in [ 2 ] a domain in \mathbb{C}^{2} known as the symmetrised bidisc \Gamma_{E(2;1;2)} . Berocovici and Fioas characterized the existence of an interpolant in (\textasteriskcentered) . Their characterization involves a nontrivial search for m appropriate matrix in \text{GL}_{n}(\mathbb{C}). Agler and Young [ 1 ] , Costara [ 22 ] proved that in the case W_{1},\ldots,W_{m} are non-derogatory then (\textasteriskcentered) is equivalent to an interpolation problem from \mathbb{D} to symmetrized polydisc. The problem (\textasteriskcentered) descends to a region of much lower dimension with many pleasant properties. By taking E=\{\operatorname{diag}[z_{1},z_{2}]\,\,\in\mathbb{C}^{2\times 2}:z_{1},z_{2}\in\mathbb{C}\}, Abouhajar, White and Young [ 12 ] showed that (\textasteriskcentered) is equivalent to an interpolation problem from \mathbb{D} to tetrablock. By considering E=\{\operatorname{diag}[z_{1}I_{n-1},z_{2}]\,\,\in\mathbb{C}^{n\times n}:z_{1},z_{2}\in\mathbb{C}\}, Bharali in [ 14 ] , established that (\textasteriskcentered) is equivalent to an interpolation problem from \mathbb{D} to \mathbb{E}_{n}. For this case also, the problem (\textasteriskcentered) descends to a region of much lower dimension. In 2017, Brown et al. obtain, for certain naturally occuring domains \mathcal{X} (symmetrized bidisc and tetrablock), a rich structure of interrelations among four naturally arising objects of analysis associated with 2\times 2 analytic matrix functions on \mathbb{D}. This rich structure, together with the classical realization formula and Hilbert space model in the sense of Agler, provided an effective framework for constructing functions in the space \mathcal{O}(\mathbb{D},\bar{\mathcal{X}}) , and thereby yielding solvability criteria for \mu -synthesis problem. Rich structure for a domain \mathcal{X} is encapsulated in the diagram see( [ 18 , page 1706] ). Let E(n;s;r_{1},\dots,r_{s})\subset\mathcal{M}_{n\times n}(\mathbb{C}) be the vector subspace comprising block diagonal matrices, defined as follows: E=E(n;s;r_{1},...,r_{s}):=\{\operatorname{diag}[z_{1}I_{r_{1}},....,z_{s}I_{r_{s}}]\in\mathcal{M}_{n\times n}(\mathbb{C}):z_{1},...,z_{s}\in\mathbb{C}\}, (1.2) where \sum_{i=1}^{s}r_{i}=n. We recall the definitions of \Gamma_{E(3;3;1,1,1)} , \Gamma_{E(3;2;1,2)} and \Gamma_{E(2;2;1,1)} , [ 12 , 14 , 39 ] . The sets \Gamma_{E{(2;2;1,1)}} , \Gamma_{E(3;3;1,1,1)} and \Gamma_{E(3;2;1,2)} are defined as \displaystyle\Gamma_{E{(2;2;1,1)}}:= \displaystyle\Big\{\textbf{x}=(x_{1}=a_{11},x_{2}=a_{22},x_{3}=a_{11}a_{22}-a_{12}a_{21}=\det A)\in\mathbb{C}^{3}:A\in\mathcal{M}_{2\times 2}(\mathbb{C}) \displaystyle{\rm{\displaystyle and}}~\mu_{E(2;2;1,1)}\leq 1\Big\}, \displaystyle\Gamma_{E{(3;3;1,1,1)}}:=\Big\{\textbf{x}=(x_{1}=a_{11},x_{2}=a_{22},x_{3}=a_{11}a_{22}-a_{12}a_{21},x_{4}=a_{33},x_{5}=a_{11}a_{33}-a_{13}a_{31}, \displaystyle x_{6}=a_{22}a_{33}-a_{23}a_{32},x_{7}=\det A)\in\mathbb{C}^{7}:A\in\mathcal{M}_{3\times 3}(\mathbb{C})~{\rm{and}}~\mu_{E(3;3;1,1,1)}(A)\leq 1\Big\} {\rm{and}} \displaystyle\Gamma_{E(3;2;1,2)}:=\Big\{(x_{1}=a_{11},x_{2}=\det\left(\begin{smallmatrix}a_{11}&a_{12}\\ a_{21}&a_{22}\end{smallmatrix}\right)+\det\left(\begin{smallmatrix}a_{11}&a_{13}\\ a_{31}&a_{33}\end{smallmatrix}\right),x_{3}=\operatorname{det}A,y_{1}=a_{22}+a_{33}, \displaystyle y_{2}=\det\left(\begin{smallmatrix}a_{22}&a_{23}\\ a_{32}&a_{33}\end{smallmatrix})\right)\in\mathbb{C}^{5}:A\in\mathcal{M}_{3\times 3}(\mathbb{C})~{\rm{and}}~\mu_{E(3;2;1,2)}(A)\leq 1\Big\}. The domains \Gamma_{E(3;2;1,2)} and \Gamma_{E(2;2;1,1)} are referred to as as \mu_{1,3}- quotient and tetrablock, respectively [ 12 , 14 ] . We now describe these objects of the rich structure related to the domains \Gamma_{E(3;3;1,1,1)} and \Gamma_{E(3;2;1,2)} as follows: (1) \mathcal{S}_{1}(\mathbb{C}^{3},\mathbb{C}^{3}) denotes the 3\times 3 matricial Schur class of the disc, that is, the set of analytic 3\times 3 matrix function F on \mathbb{D} such that \|F(\lambda)\|\leq 1 for all \mathbb{D} ; (2) \mathcal{S}_{3}(\mathbb{C},\mathbb{C}) is the Schur class of tri-disc \mathbb{D}^{3} , that is, \mathcal{O}(\mathbb{D}^{3},\mathbb{D}) ; (3) \tilde{\mathcal{R}_{1}} (in the case of \Gamma_{E(3;3;1,1,1)} ) and \tilde{\mathcal{S}_{1}} (in the case of \Gamma_{E(3;2;1,2)} ). These are defined as follows: \displaystyle\tilde{\mathcal{R}_{1}}:= \displaystyle\{(N^{(1)},N^{(2)},N^{(3)}):N^{(1)},N^{(2)},N^{(3)},K_{(N^{(1)},N^{(2)},N^{(3)})}~\text{are the analytic kernel on } \displaystyle\mathbb{D}^{3}~\text{and}\,\,K_{(N^{(1)},N^{(2)},N^{(3)})}\,\,\text{is of rank 1}\}, \displaystyle\tilde{\mathcal{S}_{1}}:= \displaystyle\{(N^{(1)}+N^{(2)},N^{(3)}):N^{(1)},N^{(2)},N^{(3)},K_{(N^{(1)},N^{(2)},N^{(3)})}~\text{are the analytic kernel on } \displaystyle\mathbb{D}^{2}~\text{and}\,\,K_{(N^{(1)},N^{(2)},N^{(3)})}\,\,\text{is of rank 1}\};~\text{and} (4) \mathcal{O}(\mathbb{D},\bar{\mathcal{X}}) , where \mathcal{X} is either \Gamma_{E(3;3;1,1,1)} or \Gamma_{E(3;2;1,2)} . We aim to define a few maps in the rich structure. In Section 2, we define SE map between \mathcal{S}_{1}(\mathbb{C}^{3},\mathbb{C}^{3}) and \mathcal{S}_{3}(\mathbb{C},\mathbb{C}) . Whereas, in Section 3, we establish the relation between \mathcal{S}_{1}(\mathbb{C}^{3},\mathbb{C}^{3})