English

Quasi-Optimal Arithmetic for Quaternion Polynomials

Symbolic Computation 2007-05-23 v2

Abstract

Fast algorithms for arithmetic on real or complex polynomials are well-known and have proven to be not only asymptotically efficient but also very practical. Based on Fast Fourier Transform (FFT), they for instance multiply two polynomials of degree up to N or multi-evaluate one at N points simultaneously within quasi-linear time O(N.polylog N). An extension to (and in fact the mere definition of) polynomials over the skew-field H of quaternions is promising but still missing. The present work proposes three such definitions which in the commutative case coincide but for H turn out to differ, each one satisfying some desirable properties while lacking others. For each notion we devise algorithms for according arithmetic; these are quasi-optimal in that their running times match lower complexity bounds up to polylogarithmic factors.

Keywords

Cite

@article{arxiv.cs/0304004,
  title  = {Quasi-Optimal Arithmetic for Quaternion Polynomials},
  author = {Martin Ziegler},
  journal= {arXiv preprint arXiv:cs/0304004},
  year   = {2007}
}

Comments

published version (11 pages) plus appendix (2 pages)