Continuously Frame-Convertible Sequences
Chad Berner · 2026
Read Continuously Frame-Convertible Sequences by Chad Berner (2026) 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-2605-02042.txt.
Introduction
Chad Berner studies sequences in a Hilbert space that can be continuously mapped to Parseval frames. Frame theory recovers a vector from inner product data, but its reconstruction formula can be costly to compute. Berner relaxes the frame condition and keeps a similar recovery formula through the continuous map.
He characterises such sequences purely through their analysis operators, with no reference to the map itself. Examples include sequences that are incomplete and sequences with no frame inside them at all. Norm-based tests tell when certain Schauder sequences, and finite unions of them, have the property. He also classifies measures on the torus for which the standard exponential system qualifies and forms a Riesz Fischer sequence.
Catalog record
| CREATOR | Chad Berner |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 41 KB |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2605.02042
Browse the full library · About the Tower of Babel
Full text
The opening of the work, from the library's hosted text:
Continuously Frame-Convertible Sequences Chad Berner Abstract . Frame theory provides a robust method for recovering vectors in a Hilbert space from inner product data, though the associated decomposition formula can be computationally demanding. We relax the frame condition by studying sequences that can be continuously mapped to Parseval frames, yielding a similar reconstruction formula. We characterize such sequences in terms of their analysis operators, without reference to any continuous mapping. We present examples, including sequences that are not complete and those containing no frame sequence. We also give norm-based criteria for when unconditional Schauder sequences and finite unions of bounded unconditional Schauder sequences admit this property. Finally, we classify finite Borel measures on the torus for which the standard exponential system has this property and forms a Riesz–Fischer sequence. 1. Introduction Frame theory, originating from Duffin and Schaeffer [ 9 ] , provides useful and stable tools to recover vectors in a Hilbert space \mathcal{H} from inner product data. Specifically, the frame decomposition formula for a frame \{f_{n}\}_{n=0}^{\infty}\subseteq\mathcal{H} provides a representation of any vector f\in\mathcal{H} via its inner product data as follows: f=\sum_{n=0}^{\infty}\langle f,f_{n}\rangle S^{-1}f_{n}=\sum_{n=0}^{\infty}\langle f,S^{-1}f_{n}\rangle f_{n}=\sum_{n=0}^{\infty}\langle S^{-1}f,f_{n}\rangle f_{n} where S\in B(\mathcal{H}) denotes the positive and invertible frame operator corresponding to \{f_{n}\} . Moreover, it is known that for a frame \{f_{n}\}_{n=0}^{\infty}\subseteq\mathcal{H} , for all f\in\mathcal{H} , f=\sum_{n=0}^{\infty}\left\langle f,S^{-\frac{1}{2}}f_{n}\right\rangle S^{-\frac{1}{2}}f_{n}. That is, \{S^{-\frac{1}{2}}f_{n}\} is a Parseval frame. However, in practice, computing the inverse (or inverse square root) of the frame operator S can be difficult. Since the frame decomposition formula already can require applying a nontrivial invertible operator to the sequence, it is natural to relax the condition and ask: which sequences \{f_{n}\}\subseteq\mathcal{H} have the property that there exists an invertible B\in B(\mathcal{H}) such that \{Bf_{n}\} is a Parseval frame. Yet, it is easy to see that this is an equivalent definition of a frame. Therefore, in this paper, we instead investigate which sequences \{f_{n}\}\subseteq\mathcal{H} have the property that there exists a B\in B(\mathcal{H}) such that \{Bf_{n}\} is a Parseval frame. A sequence with this property will be called a continuously frame-convertible sequence (CFC sequence). We provide a complete characterization of CFC sequences in terms of the analysis operator, without reference to a bounded operator. Our analysis shows that, although CFC sequences share some properties with frames, they differ from them in substantial ways. In particular, we will show that CFC sequences, although they admit frame-like reconstruction properties, need not be Bessel sequences, need not be complete, and may not even contain frame sequences. There is substantial research on relaxing the frame condition for sequences in a Hilbert space in order to understand the consequences for stable signal recovery. For instance, Li and Ogawa studied frame series and reconstruction in the weak sense [ 19 ] . Additionally, Balazs, Antoine, Grybos, and others have investigated weighted frames, that is, sequences that can be converted into frames via a sequence of scalars; see [ 2 , 11 , 3 ] . Furthermore, Christensen provided results on frame expansions utilizing bounded operators [ 8 ] . Antoine, Corso, and Trapani also studied metric operators in relation to frames [ 1 ] , considering when \{Bf_{n}\} forms a frame for positive definite (possibly unbounded) operators B . Their setting is more general in the sense that it allows unbounded positive operators, whereas we restrict attention to bounded operators without a positivity assumption and obtain a more explicit characterization. In [ 4 ] , sequences with a frame-like decomposition were investigated; this condition is strictly stronger than the CFC property considered here. Finally, the Kaczmarz algorithm offers an independent approach to recovering vectors in a Hilbert space from inner product data; see [ 16 , 13 ] . In this work, we provide examples of weighted frames and sequences arising from the Kaczmarz algorithm, and identify specific cases in which they are, or fail to be, CFC sequences. The primary aim of this paper is to establish a classification of CFC sequences in terms of the analysis operator, without reference to any bounded operator. A secondary objective is to determine when specific sequences, including finite unions of bounded Schauder sequences and exponential systems, are CFC sequences. The outline of this paper is as follows. In Section 3 , we classify CFC sequences in terms of the analysis operator and provide examples of CFC sequences that fail to exhibit properties typical of frames. In Section 4 , we show that Riesz–Fischer sequences are CFC sequences and classify them in terms of the analysis operator. In Sections 5 , 6 , we provide a norm-based classification for when unconditional Schauder sequences and finite unions of bounded unconditional Schauder sequences are CFC sequences. Finally, in Section 7 , we classify finite Borel measures on the torus for which the standard exponential system is a CFC sequence and, separately, those for which it is a Riesz–Fischer sequence. 2. Preliminaries We begin by providing the definition of a frame: Definition 2.1 . Let \{f_{n}\}_{n}\subseteq\mathcal{H} where \mathcal{H} is a Hilbert space. (1) If there exists A>0 such that A\|f\|^{2}\leq\sum_{n}|\langle f,f_{n}\rangle|^{2} for all f\in\mathcal{H} , then \{f_{n}\}_{n} is called a lower semi-frame . (2) If there exists B>0 such that \sum_{n}|\langle f,f_{n}\rangle|^{2}\leq B\|f\|^{2} for all f\in\mathcal{H} , then \{f_{n}\}_{n} is called a Bessel sequence . (3) If there exists A,B>0 such that A\|f\|^{2}\leq\sum_{n}|\langle f,f_{n}\rangle|^{2}\leq B\|f\|^{2} for all f\in\mathcal{H} , then \{f_{n}\}_{n} is called a frame . Furthermore, if \mathcal{H}=\overline{\operatorname{span}\{f_{n}\}} , then \{f_{n}\}_{n} is called a frame sequence . (4) If \|f\|^{2}=\sum_{n}|\langle f,f_{n}\rangle|^{2} for all f\in\mathcal{H} , then \{f_{n}\}_{n} is called a Parseval frame . (5) If \{f_{n}\} is a frame, and there is a \{g_{n}\}_{n}\subseteq\mathcal{H} such that \langle f_{n},g_{k}\rangle=\delta_{nk}, then \{f_{n}\} is called a Riesz basis . Furthermore, if \mathcal{H}=\overline{\operatorname{span}\{f_{n}\}} , then \{f_{n}\}_{n} is called a Riesz sequence . Moreover, if \{f_{n}\}_{n} is a Bessel sequence in Hilbert space \mathcal{H} , the map on \mathcal{H} : f\to\{\langle f,f_{n}\rangle\}_{n} is called the analysis operator , its adjoint is called the synthesis operator , and we denote S:\mathcal{H}\to\mathcal{H} as the positive operator such that S(f)=\sum_{n}\langle f,f_{n}\rangle f_{n}, which is called the frame operator . Additionally, if \{f_{n}\}_{n} is a frame, then its associated frame operator is invertible, and \{S^{-1}f_{n}\}_{n} is a frame such that f=\sum_{n}\langle f,S^{-1}f_{n}\rangle f_{n}=\sum_{n}\langle f,f_{n}\rangle S^{-1}f_{n} with convergence in norm for all f\in\mathcal{H} . Additionally, if \{f_{n}\} and \{g_{n}\} are frames (Bessel sequences) such that for all f\in\mathcal{H} , f=\sum_{n}\langle f,f_{n}\rangle g_{n}, then \{f_{n}\} and \{g_{n}\} are called dual frames . In particular, many frames have multiple dual frames. Additionally, because the synthesis operator and analysis operator associated with both sequences are bounded, by an adjoint calculation, f=\sum_{n}\langle f,g_{n}\rangle f_{n} also holds for all f\in\mathcal{H} . Finally, the norm convergence of frame series is also unconditional. Furthermore, we define Schauder bases: Definition 2.2 . Let \{f_{n}\}_{n}\subseteq\mathcal{H} where \mathcal{H} is a Hilbert space. \{f_{n}\}_{n} is called a Schauder basis if there is a \{g_{n}\}_{n}\subseteq\mathcal{H} such that \langle f_{n},g_{k}\rangle=\delta_{nk} and (2.1) f=\sum_{n}\langle f,g_{n}\rangle f_{n} with convergence in norm for all f\in\mathcal{H} . If the convergence in equation ( 2.1 ) is unconditional for all f\in\mathcal{H} , then \{f_{n}\}_{n} is called an unconditional Schauder basis . Again, if \mathcal{H}=\overline{\operatorname{span}\{f_{n}\}} , then \{f_{n}\}_{n} is called an (unconditional) Schauder sequence . Remark 2.3 . A