Efficient implementation of data binning in sectors
Year: 2012
Type: report
Abstract: We analyze the implementation of data binning of $\mathbb{R}^2$-vectors or complex numbers in sectors defined by ranges of vector/complex arguments. The binning problem is formalized and 4 different equivalent formulations are given resulting in 4 different solution methods. The 4 different methods have different implementation properties making them favorable on different platforms and under different circumstances. Binning with respect to the vector/complex number magnitude is also briefly covered. All methods are presented in a common bisection framework which make them easy to compare. The methods are given in basic arithmetic operations and logics and are directly emendable for implementation for both integer, fixed and floating point data. Detailed implementation aspects and optimizations are discussed and C-code snippets are given for the different methods.
Author: John-Olof Nilsson
Language: English
Subfield: Computational Theory and Mathematics
Field: Computer Science
Domain: Physical Sciences
Open Access status: green
OpenAlex ID: W2150437621