N-body simulation with hierarchical algorithm with principal component analysis support

• Grant number: 11/07624-4
• Support Opportunities: Scholarships in Brazil - Scientific Initiation
• Start date: June 01, 2011
• End date: November 30, 2011
• Field of knowledge: Physical Sciences and Mathematics - Computer Science
• Principal Investigator: Eraldo Pereira Marinho
• Grantee: Martha Dais Ferreira
• Host Institution: Instituto de Geociências e Ciências Exatas (IGCE). Universidade Estadual Paulista (UNESP). Campus de Rio Claro. Rio Claro , SP, Brazil

Abstract

The project is intended to initiate the student in hierarchical methods of spatial decomposition of data, applied to N-body simulations of selfgravitating systems. The decomposition techniques are based on a new extension of octal trees, whose directions of spatial tessellation are determined by the principal components of the covariance tensor. Unlike the traditional techniques of well-known tree-codes, which generally require cubic division of space, this new technique allows the division of the space into convex cells, formed by the intersection of the planes defined by the covariance eigenvectors, ensuring greater entropy of the configuration elements volumes collected as sources of collective gravitational field.