The construction of quadrature formulae (Q4334945)

This paper develops a method for effectively constructing quadrature formulas based on the theory of divisors in algebraic number fields that has been proposed by \textit{S. M. Voronin} and \textit{N. Temirgaliev} [Math. Notes 46, No. 2, 597-602 (1989); translation from Mat. Zametki 46, No. 2, 34-41 (1989; Zbl 0715.41043)], \textit{N. Temirgaliev} [Math. USSR, Sb. 69, No. 2, 527-542 (1991); translation from Mat. Sb. 181, No. 4, 490-505 (1990; Zbl 0727.41032), and \textit{S. M. Voronin} [Russ. Acad. Sci., Izv., Math. 45, No. 2, 417-422 (1995); translation from Izv. Russ. Akad. Nauk, Ser. Mat. 58, No. 5, 189-194 (1994; Zbl 0841.41029)]. Let \(f(\vec x) =\sum_{\vec n \in {\mathcal M}} c_{\vec n} \exp (2\pi i(\vec n, \vec x))\) be a Fourier polynomial, where \({\mathcal M} \subset \mathbb{Z}^r\) and \(\vec x\in \mathbb{R}^n\). Using the arithmetic of cyclotomic fields, the author constructs a quadrature formula which is exact on classes of Fourier polynomials, i.e. \(\int^1_0 \cdots \int^1_0 f(\vec x)d \vec x= {1\over p} \sum_{\nu \bmod p} f({\vec a^* \nu \over p})\), where \(p\) is prime, and \(\vec a^* \in \mathbb{Z}^r\) can be effectively determined. Differing from the previous works mentioned avove, the present method can be used in practice for comparatively large values of \(r\).