<p>We develop an alternative approach to the study of Fourier series, based on the Short-Time-Fourier Transform (STFT) acting on <InlineEquation ID="IEq1"> <InlineMediaObject> <ImageObject Color="BlackWhite" FileRef="41_2025_10196_Article_IEq1.gif" Format="GIF" Height="20" Rendition="HTML" Resolution="72" Type="Linedraw" Width="56" /> </InlineMediaObject> <EquationSource Format="TEX">\(L_{\nu }^{2}(0,1)\)</EquationSource> <EquationSource Format="MATHML"><math> <mrow> <msubsup> <mi>L</mi> <mrow> <mi>ν</mi> </mrow> <mn>2</mn> </msubsup> <mrow> <mo stretchy="false">(</mo> <mn>0</mn> <mo>,</mo> <mn>1</mn> <mo stretchy="false">)</mo> </mrow> </mrow> </math></EquationSource> </InlineEquation>, the space of measurable functions <i>f</i> in <InlineEquation ID="IEq2"> <InlineMediaObject> <ImageObject Color="BlackWhite" FileRef="41_2025_10196_Article_IEq2.gif" Format="GIF" Height="14" Rendition="HTML" Resolution="72" Type="Linedraw" Width="12" /> </InlineMediaObject> <EquationSource Format="TEX">\(\mathbb {R}\)</EquationSource> <EquationSource Format="MATHML"><math> <mi mathvariant="double-struck">R</mi> </math></EquationSource> </InlineEquation>, square-integrable in (0,&#xa0;1), and <i>time-periodic up to a phase factor</i>: for fixed <InlineEquation ID="IEq3"> <InlineMediaObject> <ImageObject Color="BlackWhite" FileRef="41_2025_10196_Article_IEq3.gif" Format="GIF" Height="14" Rendition="HTML" Resolution="72" Type="Linedraw" Width="43" /> </InlineMediaObject> <EquationSource Format="TEX">\(\nu \in \mathbb {R}\)</EquationSource> <EquationSource Format="MATHML"><math> <mrow> <mi>ν</mi> <mo>∈</mo> <mi mathvariant="double-struck">R</mi> </mrow> </math></EquationSource> </InlineEquation>, <Equation ID="Equ98"> <MediaObject> <ImageObject Color="BlackWhite" FileRef="41_2025_10196_Article_Equ98.gif" Format="GIF" Height="22" Rendition="HTML" Resolution="72" Type="Linedraw" Width="205" /> </MediaObject> <EquationSource Format="TEX">\(\begin{aligned} f(t+k)=e^{2\pi ik\nu }f(t){, \ }k\in \mathbb {Z}\text {.} \end{aligned}\)</EquationSource> <EquationSource Format="MATHML"><math display="block"> <mrow> <mtable> <mtr> <mtd columnalign="right"> <mrow> <mi>f</mi> <mrow> <mo stretchy="false">(</mo> <mi>t</mi> <mo>+</mo> <mi>k</mi> <mo stretchy="false">)</mo> </mrow> <mo>=</mo> <msup> <mi>e</mi> <mrow> <mn>2</mn> <mi>π</mi> <mi>i</mi> <mi>k</mi> <mi>ν</mi> </mrow> </msup> <mi>f</mi> <mrow> <mo stretchy="false">(</mo> <mi>t</mi> <mo stretchy="false">)</mo> </mrow> <mrow> <mo>,</mo> <mspace width="4pt" /> </mrow> <mi>k</mi> <mo>∈</mo> <mi mathvariant="double-struck">Z</mi> <mtext>.</mtext> </mrow> </mtd> </mtr> </mtable> </mrow> </math></EquationSource> </Equation>The resulting phase space is the vertical strip <InlineEquation ID="IEq4"> <InlineMediaObject> <ImageObject Color="BlackWhite" FileRef="41_2025_10196_Article_IEq4.gif" Format="GIF" Height="19" Rendition="HTML" Resolution="72" Type="Linedraw" Width="119" /> </InlineMediaObject> <EquationSource Format="TEX">\(\mathbb {C}/\mathbb {Z}=[0,1)\times \mathbb {R}\)</EquationSource> <EquationSource Format="MATHML"><math> <mrow> <mi mathvariant="double-struck">C</mi> <mo stretchy="false">/</mo> <mi mathvariant="double-struck">Z</mi> <mo>=</mo> <mo stretchy="false">[</mo> <mn>0</mn> <mo>,</mo> <mn>1</mn> <mo stretchy="false">)</mo> <mo>×</mo> <mi mathvariant="double-struck">R</mi> </mrow> </math></EquationSource> </InlineEquation>, a flat model of an infinite cylinder, which leads to Gabor frames with an interesting structure theory, allowing for a Janssen-type representation. As expected, a Gaussian window leads to a Fock space of entire functions, studied in the companion paper by the same authors [<i>Beurling-type density theorems for sampling and interpolation on the flat cylinder</i>]. When <i>g</i> is a Hermite function, we are lead to <i>true</i> Fock spaces of polyanalytic functions (<i>Landau level</i> eigenspaces) on the vertical strip <InlineEquation ID="IEq5"> <InlineMediaObject> <ImageObject Color="BlackWhite" FileRef="41_2025_10196_Article_IEq5.gif" Format="GIF" Height="19" Rendition="HTML" Resolution="72" Type="Linedraw" Width="67" /> </InlineMediaObject> <EquationSource Format="TEX">\([0,1)\times \mathbb {R}\)</EquationSource> <EquationSource Format="MATHML"><math> <mrow> <mo stretchy="false">[</mo> <mn>0</mn> <mo>,</mo> <mn>1</mn> <mo stretchy="false">)</mo> <mo>×</mo> <mi mathvariant="double-struck">R</mi> </mrow> </math></EquationSource> </InlineEquation>. We first prove a density condition for a lattice to be interpolating in this space. Furthermore, an analogue of the sufficient Wexler-Raz conditions is obtained which leads to new criteria for Gabor frames in <InlineEquation ID="IEq6"> <InlineMediaObject> <ImageObject Color="BlackWhite" FileRef="41_2025_10196_Article_IEq6.gif" Format="GIF" Height="20" Rendition="HTML" Resolution="72" Type="Linedraw" Width="42" /> </InlineMediaObject> <EquationSource Format="TEX">\(L^{2}(\mathbb {R})\)</EquationSource> <EquationSource Format="MATHML"><math> <mrow> <msup> <mi>L</mi> <mn>2</mn> </msup> <mrow> <mo stretchy="false">(</mo> <mi mathvariant="double-struck">R</mi> <mo stretchy="false">)</mo> </mrow> </mrow> </math></EquationSource> </InlineEquation>, and to sufficient conditions for Gabor frames in <InlineEquation ID="IEq7"> <InlineMediaObject> <ImageObject Color="BlackWhite" FileRef="41_2025_10196_Article_IEq1.gif" Format="GIF" Height="20" Rendition="HTML" Resolution="72" Type="Linedraw" Width="56" /> </InlineMediaObject> <EquationSource Format="TEX">\(L_{\nu }^{2}(0,1)\)</EquationSource> <EquationSource Format="MATHML"><math> <mrow> <msubsup> <mi>L</mi> <mrow> <mi>ν</mi> </mrow> <mn>2</mn> </msubsup> <mrow> <mo stretchy="false">(</mo> <mn>0</mn> <mo>,</mo> <mn>1</mn> <mo stretchy="false">)</mo> </mrow> </mrow> </math></EquationSource> </InlineEquation> with Hermite windows (an analogue of a theorem of Gröchenig and Lyubarskii about Gabor frames with Hermite windows) and with totally positive windows in the Feichtinger algebra (an analogue of a recent theorem of Gröchenig). We also consider a vectorial STFT in <InlineEquation ID="IEq8"> <InlineMediaObject> <ImageObject Color="BlackWhite" FileRef="41_2025_10196_Article_IEq1.gif" Format="GIF" Height="20" Rendition="HTML" Resolution="72" Type="Linedraw" Width="56" /> </InlineMediaObject> <EquationSource Format="TEX">\(L_{\nu }^{2}(0,1)\)</EquationSource> <EquationSource Format="MATHML"><math> <mrow> <msubsup> <mi>L</mi> <mrow> <mi>ν</mi> </mrow> <mn>2</mn> </msubsup> <mrow> <mo stretchy="false">(</mo> <mn>0</mn> <mo>,</mo> <mn>1</mn> <mo stretchy="false">)</mo> </mrow> </mrow> </math></EquationSource> </InlineEquation> and, using the vector with the first Hermite functions as window, we introduce the (full) Fock spaces of polyanalytic functions on <InlineEquation ID="IEq9"> <InlineMediaObject> <ImageObject Color="BlackWhite" FileRef="41_2025_10196_Article_IEq5.gif" Format="GIF" Height="19" Rendition="HTML" Resolution="72" Type="Linedraw" Width="67" /> </InlineMediaObject> <EquationSource Format="TEX">\([0,1)\times \mathbb {R}\)</EquationSource> <EquationSource Format="MATHML"><math> <mrow> <mo stretchy="false">[</mo> <mn>0</mn> <mo>,</mo> <mn>1</mn> <mo stretchy="false">)</mo> <mo>×</mo> <mi mathvariant="double-struck">R</mi> </mrow> </math></EquationSource> </InlineEquation> and their associated Bargmann-type transforms, and prove an analogue of Vasilevski’s orthogonal decomposition into true polyanalytic Fock spaces (Landau level eigenspaces on <InlineEquation ID="IEq10"> <InlineMediaObject> <ImageObject Color="BlackWhite" FileRef="41_2025_10196_Article_IEq5.gif" Format="GIF" Height="19" Rendition="HTML" Resolution="72" Type="Linedraw" Width="67" /> </InlineMediaObject> <EquationSource Format="TEX">\([0,1)\times \mathbb {R}\)</EquationSource> <EquationSource Format="MATHML"><math> <mrow> <mo stretchy="false">[</mo> <mn>0</mn> <mo>,</mo> <mn>1</mn> <mo stretchy="false">)</mo> <mo>×</mo> <mi mathvariant="double-struck">R</mi> </mrow> </math></EquationSource> </InlineEquation>). We conclude the paper with an analogue of Gröchenig-Lyubarskii’s sufficient condition for Gabor super-frames with Hermite functions, which is equivalent to a sufficient sampling condition on the full Fock space of polyanalytic functions on <InlineEquation ID="IEq11"> <InlineMediaObject> <ImageObject Color="BlackWhite" FileRef="41_2025_10196_Article_IEq5.gif" Format="GIF" Height="19" Rendition="HTML" Resolution="72" Type="Linedraw" Width="67" /> </InlineMediaObject> <EquationSource Format="TEX">\([0,1)\times \mathbb {R}\)</EquationSource> <EquationSource Format="MATHML"><math> <mrow> <mo stretchy="false">[</mo> <mn>0</mn> <mo>,</mo> <mn>1</mn> <mo stretchy="false">)</mo> <mo>×</mo> <mi mathvariant="double-struck">R</mi> </mrow> </math></EquationSource> </InlineEquation>. The proofs of the results about Gabor frames, involving some of Gröchenig’s most significant results of the past 25 years, are a clear indication of his influence on the field during this period.</p>

错误:搜索内容不能为空,请输入英文关键词
错误:关键词超出字数限制,请精简
高级检索

Gabor Frames for Quasi-Periodic Functions and Polyanalytic Spaces on the Flat Cylinder

  • Luís Daniel Abreu,
  • Franz Luef,
  • Mohammed Ziyat

摘要

We develop an alternative approach to the study of Fourier series, based on the Short-Time-Fourier Transform (STFT) acting on \(L_{\nu }^{2}(0,1)\) L ν 2 ( 0 , 1 ) , the space of measurable functions f in \(\mathbb {R}\) R , square-integrable in (0, 1), and time-periodic up to a phase factor: for fixed \(\nu \in \mathbb {R}\) ν R , \(\begin{aligned} f(t+k)=e^{2\pi ik\nu }f(t){, \ }k\in \mathbb {Z}\text {.} \end{aligned}\) f ( t + k ) = e 2 π i k ν f ( t ) , k Z . The resulting phase space is the vertical strip \(\mathbb {C}/\mathbb {Z}=[0,1)\times \mathbb {R}\) C / Z = [ 0 , 1 ) × R , a flat model of an infinite cylinder, which leads to Gabor frames with an interesting structure theory, allowing for a Janssen-type representation. As expected, a Gaussian window leads to a Fock space of entire functions, studied in the companion paper by the same authors [Beurling-type density theorems for sampling and interpolation on the flat cylinder]. When g is a Hermite function, we are lead to true Fock spaces of polyanalytic functions (Landau level eigenspaces) on the vertical strip \([0,1)\times \mathbb {R}\) [ 0 , 1 ) × R . We first prove a density condition for a lattice to be interpolating in this space. Furthermore, an analogue of the sufficient Wexler-Raz conditions is obtained which leads to new criteria for Gabor frames in \(L^{2}(\mathbb {R})\) L 2 ( R ) , and to sufficient conditions for Gabor frames in \(L_{\nu }^{2}(0,1)\) L ν 2 ( 0 , 1 ) with Hermite windows (an analogue of a theorem of Gröchenig and Lyubarskii about Gabor frames with Hermite windows) and with totally positive windows in the Feichtinger algebra (an analogue of a recent theorem of Gröchenig). We also consider a vectorial STFT in \(L_{\nu }^{2}(0,1)\) L ν 2 ( 0 , 1 ) and, using the vector with the first Hermite functions as window, we introduce the (full) Fock spaces of polyanalytic functions on \([0,1)\times \mathbb {R}\) [ 0 , 1 ) × R and their associated Bargmann-type transforms, and prove an analogue of Vasilevski’s orthogonal decomposition into true polyanalytic Fock spaces (Landau level eigenspaces on \([0,1)\times \mathbb {R}\) [ 0 , 1 ) × R ). We conclude the paper with an analogue of Gröchenig-Lyubarskii’s sufficient condition for Gabor super-frames with Hermite functions, which is equivalent to a sufficient sampling condition on the full Fock space of polyanalytic functions on \([0,1)\times \mathbb {R}\) [ 0 , 1 ) × R . The proofs of the results about Gabor frames, involving some of Gröchenig’s most significant results of the past 25 years, are a clear indication of his influence on the field during this period.