<p>For a real distribution <InlineEquation ID="IEq1"> <EquationSource Format="TEX">\(\mathcal {D}\)</EquationSource> <EquationSource Format="MATHML"><math> <mi mathvariant="script">D</mi> </math></EquationSource> </InlineEquation> on the interval [0,&#xa0;<i>L</i>] with <InlineEquation ID="IEq2"> <EquationSource Format="TEX">\(\widetilde{\mathcal { D}}\)</EquationSource> <EquationSource Format="MATHML"><math> <mover accent="true"> <mi mathvariant="script">D</mi> <mo stretchy="true">~</mo> </mover> </math></EquationSource> </InlineEquation> the associated even distribution on the interval <InlineEquation ID="IEq3"> <EquationSource Format="TEX">\([-L, L]\)</EquationSource> <EquationSource Format="MATHML"><math> <mrow> <mo stretchy="false">[</mo> <mo>-</mo> <mi>L</mi> <mo>,</mo> <mi>L</mi> <mo stretchy="false">]</mo> </mrow> </math></EquationSource> </InlineEquation>, we prove that if the associated quadratic form with Schwartz kernel <InlineEquation ID="IEq4"> <EquationSource Format="TEX">\(\widetilde{\mathcal {D}}(x - y)\)</EquationSource> <EquationSource Format="MATHML"><math> <mrow> <mover accent="true"> <mi mathvariant="script">D</mi> <mo stretchy="true">~</mo> </mover> <mrow> <mo stretchy="false">(</mo> <mi>x</mi> <mo>-</mo> <mi>y</mi> <mo stretchy="false">)</mo> </mrow> </mrow> </math></EquationSource> </InlineEquation> defines a lower-bounded selfadjoint operator on <InlineEquation ID="IEq5"> <EquationSource Format="TEX">\(L^2([-\frac{L}{2}, \frac{L}{2}])\)</EquationSource> <EquationSource Format="MATHML"><math> <mrow> <msup> <mi>L</mi> <mn>2</mn> </msup> <mrow> <mo stretchy="false">(</mo> <mrow> <mo stretchy="false">[</mo> <mo>-</mo> <mfrac> <mi>L</mi> <mn>2</mn> </mfrac> <mo>,</mo> <mfrac> <mi>L</mi> <mn>2</mn> </mfrac> <mo stretchy="false">]</mo> </mrow> <mo stretchy="false">)</mo> </mrow> </mrow> </math></EquationSource> </InlineEquation>, whose lowest spectral value <InlineEquation ID="IEq6"> <EquationSource Format="TEX">\(\lambda \)</EquationSource> <EquationSource Format="MATHML"><math> <mi>λ</mi> </math></EquationSource> </InlineEquation> is a simple, isolated eigenvalue with even eigenfunction <InlineEquation ID="IEq7"> <EquationSource Format="TEX">\(\xi \)</EquationSource> <EquationSource Format="MATHML"><math> <mi>ξ</mi> </math></EquationSource> </InlineEquation>, then all the zeros of the entire function <InlineEquation ID="IEq8"> <EquationSource Format="TEX">\(\widehat{\xi }(z)\)</EquationSource> <EquationSource Format="MATHML"><math> <mrow> <mover accent="true"> <mi>ξ</mi> <mo stretchy="true">^</mo> </mover> <mrow> <mo stretchy="false">(</mo> <mi>z</mi> <mo stretchy="false">)</mo> </mrow> </mrow> </math></EquationSource> </InlineEquation>, the Fourier transform of <InlineEquation ID="IEq9"> <EquationSource Format="TEX">\(\xi \)</EquationSource> <EquationSource Format="MATHML"><math> <mi>ξ</mi> </math></EquationSource> </InlineEquation>, lie on the real line. The proof proceeds in five steps. (1) We give a <InlineEquation ID="IEq10"> <EquationSource Format="TEX">\(C^*\)</EquationSource> <EquationSource Format="MATHML"><math> <msup> <mi>C</mi> <mo>∗</mo> </msup> </math></EquationSource> </InlineEquation>-algebraic proof of a corollary of Carathéodory–Fejér’s 1911 structure theorem for Toeplitz matrices: if <InlineEquation ID="IEq11"> <EquationSource Format="TEX">\(T \in M_n(\mathbb {C})\)</EquationSource> <EquationSource Format="MATHML"><math> <mrow> <mi>T</mi> <mo>∈</mo> <msub> <mi>M</mi> <mi>n</mi> </msub> <mrow> <mo stretchy="false">(</mo> <mi mathvariant="double-struck">C</mi> <mo stretchy="false">)</mo> </mrow> </mrow> </math></EquationSource> </InlineEquation> is a Hermitian, positive semidefinite Toeplitz matrix of rank <InlineEquation ID="IEq12"> <EquationSource Format="TEX">\(n - 1\)</EquationSource> <EquationSource Format="MATHML"><math> <mrow> <mi>n</mi> <mo>-</mo> <mn>1</mn> </mrow> </math></EquationSource> </InlineEquation>, and <InlineEquation ID="IEq13"> <EquationSource Format="TEX">\(\xi \in \ker T\)</EquationSource> <EquationSource Format="MATHML"><math> <mrow> <mi>ξ</mi> <mo>∈</mo> <mo>ker</mo> <mi>T</mi> </mrow> </math></EquationSource> </InlineEquation>, then the polynomial <InlineEquation ID="IEq14"> <EquationSource Format="TEX">\(P(z) = \sum \xi _j z^j\)</EquationSource> <EquationSource Format="MATHML"><math> <mrow> <mi>P</mi> <mrow> <mo stretchy="false">(</mo> <mi>z</mi> <mo stretchy="false">)</mo> </mrow> <mo>=</mo> <mo>∑</mo> <msub> <mi>ξ</mi> <mi>j</mi> </msub> <msup> <mi>z</mi> <mi>j</mi> </msup> </mrow> </math></EquationSource> </InlineEquation> has all its zeros on the unit circle. (2) We formulate and prove a continuous analogue of this result, replacing the Toeplitz matrix with a convolution operator with continuous kernel <InlineEquation ID="IEq15"> <EquationSource Format="TEX">\(h(x - y)\)</EquationSource> <EquationSource Format="MATHML"><math> <mrow> <mi>h</mi> <mo stretchy="false">(</mo> <mi>x</mi> <mo>-</mo> <mi>y</mi> <mo stretchy="false">)</mo> </mrow> </math></EquationSource> </InlineEquation>, and the polynomial <i>P</i>(<i>z</i>) with the Fourier transform of the eigenfunction corresponding to the largest eigenvalue. (3) We analyze finite-dimensional truncations of the quadratic forms defined by real, even distributions <InlineEquation ID="IEq16"> <EquationSource Format="TEX">\(\mathcal {D}\)</EquationSource> <EquationSource Format="MATHML"><math> <mi mathvariant="script">D</mi> </math></EquationSource> </InlineEquation> on <InlineEquation ID="IEq17"> <EquationSource Format="TEX">\([-L, L]\)</EquationSource> <EquationSource Format="MATHML"><math> <mrow> <mo stretchy="false">[</mo> <mo>-</mo> <mi>L</mi> <mo>,</mo> <mi>L</mi> <mo stretchy="false">]</mo> </mrow> </math></EquationSource> </InlineEquation>, and observe that the resulting matrices exhibit a structure previously encountered in perturbative expansions of the spectral action. (4) We establish an analogue of Carathéodory–Fejér’s corollary for matrices of this specific structure, thereby extending the zero localization result beyond the classical Toeplitz setting. (5) Finally, we apply a classical theorem of Hurwitz concerning the zeros of uniform limits of holomorphic functions to deduce the general result stated above.</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

Quadratic Forms, Real Zeros and Echoes of the Spectral Action

  • Alain Connes,
  • Walter D. van Suijlekom

摘要

For a real distribution \(\mathcal {D}\) D on the interval [0, L] with \(\widetilde{\mathcal { D}}\) D ~ the associated even distribution on the interval \([-L, L]\) [ - L , L ] , we prove that if the associated quadratic form with Schwartz kernel \(\widetilde{\mathcal {D}}(x - y)\) D ~ ( x - y ) defines a lower-bounded selfadjoint operator on \(L^2([-\frac{L}{2}, \frac{L}{2}])\) L 2 ( [ - L 2 , L 2 ] ) , whose lowest spectral value \(\lambda \) λ is a simple, isolated eigenvalue with even eigenfunction \(\xi \) ξ , then all the zeros of the entire function \(\widehat{\xi }(z)\) ξ ^ ( z ) , the Fourier transform of \(\xi \) ξ , lie on the real line. The proof proceeds in five steps. (1) We give a \(C^*\) C -algebraic proof of a corollary of Carathéodory–Fejér’s 1911 structure theorem for Toeplitz matrices: if \(T \in M_n(\mathbb {C})\) T M n ( C ) is a Hermitian, positive semidefinite Toeplitz matrix of rank \(n - 1\) n - 1 , and \(\xi \in \ker T\) ξ ker T , then the polynomial \(P(z) = \sum \xi _j z^j\) P ( z ) = ξ j z j has all its zeros on the unit circle. (2) We formulate and prove a continuous analogue of this result, replacing the Toeplitz matrix with a convolution operator with continuous kernel \(h(x - y)\) h ( x - y ) , and the polynomial P(z) with the Fourier transform of the eigenfunction corresponding to the largest eigenvalue. (3) We analyze finite-dimensional truncations of the quadratic forms defined by real, even distributions \(\mathcal {D}\) D on \([-L, L]\) [ - L , L ] , and observe that the resulting matrices exhibit a structure previously encountered in perturbative expansions of the spectral action. (4) We establish an analogue of Carathéodory–Fejér’s corollary for matrices of this specific structure, thereby extending the zero localization result beyond the classical Toeplitz setting. (5) Finally, we apply a classical theorem of Hurwitz concerning the zeros of uniform limits of holomorphic functions to deduce the general result stated above.