The asymptotic number of zeros of exponential sums in critical strips (Q2225667)

From MaRDI portal

!

This is the item page for this Wikibase entity, intended for internal use and editing purposes. Please use the normal view instead:

scientific article; zbMATH DE number 7308732
Language Label Description Also known as
default for all languages
No label defined
    English
    The asymptotic number of zeros of exponential sums in critical strips
    scientific article; zbMATH DE number 7308732

      Statements

      The asymptotic number of zeros of exponential sums in critical strips (English)
      0 references
      0 references
      0 references
      10 February 2021
      0 references
      The authors consider the zero distribution of normalized exponential sums of the form \[f(z)=1+ H_1 e^{w_1 z}+\dots +H_n e^{w_n z}, \tag {1}\] where \(H_1, \dots, H_n \in \mathbb{C}\), \(0<w_1<\dots< w_n\). The article [\textit{R. E. Langer}, Bull. Am. Math. Soc. 37, 213--239 (1931, Zbl 0001.34403)] contains a survey of classical results on this topic. Any vertical strip of the form \[\left\{z=x+iy: |H_k| e^{w_k x} > \sum_{j\ne k} ^n |H_j| e^{w_j x} \right\} , \] where \(k\in \{0, \dots, n\} \), \(H_0=1\), \(w_0=0\), is zero free, provided that it is non-empty. There are at least two and at most \(n+1\) zero-free regions. The closed area between two consecutive zero-free regions with dominating terms \(H_je^{w_jz}\) and \(H_ke^{w_kz}\) is called a critical strip and is denoted by \(\Lambda(j,k)\). The following main result of the paper complements a result of \textit{C. J. Moreno} [Compos. Math. 26, 69--78 (1973, Zbl 0267.33001)]. Theorem 2.1. Let \(f\) be an exponential sum of the form (1). Then all zeros of \(f\) are in finitely many critical strips \(\Lambda(j,k)\). Moreover, let \(R\) be any rectangle cut from a critical strip \(\Lambda(j,k)\) by two horizontal lines \(\Im z= y_1\) and \(\Im z=y_2\) with \(y_2-y_1=r>0\). Then the number \(n(r, \Lambda(j,k))\) of zeros of \(f\) inside \(R\) satisfies \[ n(r, \Lambda(j,k))=\frac{|w_j-w_k|}{2\pi}r+O(1).\] The proof is based on Backlund's lemma [\textit{R. J. Backlund}, C. R. Acad. Sci., Paris 158, 1979--1981 (1914; JFM 45.0719.02)].
      0 references
      exponential sums
      0 references
      zero distribution
      0 references

      Identifiers