RUS  ENG
Full version
JOURNALS // Numerical methods and programming // Archive

Num. Meth. Prog., 2011 Volume 12, Issue 4, Pages 423–434 (Mi vmp211)

This article is cited in 3 papers

Вычислительные методы и приложения

A parallel algorithm for solving strong separability problem on the basis of Fejer mappings

A. V. Ershova, I. M. Sokolinskaya

South Ural State University, Chelyabinsk

Abstract: An approach to solving the problem of separating two convex nonintersecting polyhedrons by the layer of maximum thickness is proposed. A parallel algorithm based on a method of using Fejer mappings is described. This algorithm admits an efficient implementation on the massively parallel multiprocessor systems. The results of computing experiments confirming the efficiency of the proposed approach are discussed. This work was supported by the Russian Foundation for Basic Research (project N 09-01-00546а).

Keywords: strong separability; Fejer mappings; parallel programming; pseudoprojection; iterative process; pattern recognition.

UDC: 519.6



© Steklov Math. Inst. of RAS, 2026