Approximation by periodic multivariate quasi-projection operators (Q2183638)

Approximation properties of non-periodic quasi-projection operators are under an active study. This family of operators is very large. It includes classical sampling and sampling-type expansions, Kantorovich-Kotelnikov operators and their generalizations, scaling expansions associated with wavelet constructions, and others. References are given in the Introduction, as well as a brief presentation of the pertinent definitions and results. The goal of the paper is to obtain periodic analogs of some results obtained by the authors in [J. Fourier Anal. Appl. 24, No. 5, 1276--1305 (2018; Zbl 1402.41008); J. Math. Anal. Appl. 456, No. 1, 195--213 (2017; Zbl 1371.41036); Anal. Appl., Singap. 15, No. 4, 521--542 (2017; Zbl 1366.41020)]. In this sense, approximation properties of periodic quasi-projection operators with matrix dilations are studied. Such operators are generated by a sequence of functions and a sequence of distributions/functions. Error estimates for sampling-type quasi-projection operators are obtained under the periodic Strang-Fix conditions for the functions and the compatibility conditions for functions and distributions. The error estimates are given in terms of the Fourier coefficients of approximated functions and provide analogs of some known non-periodic results. Several improvements of the error estimates are given in the particular case of strict compatibility of functions and distributions: the error estimate is given only in terms of the best approximation. If the compatibility is weak, the corresponding estimates are given simultaneously in terms of Fourier coefficients and the best approximation. The case of Kantorovich-type quasi-projection operators (i.e., the distributions are integrable functions) is also considered. The last section is devoted to examples.