English

Real Sparse Fast DCT for Vectors with Short Support

Numerical Analysis 2018-07-24 v2

Abstract

In this paper we present a new fast and deterministic algorithm for the inverse discrete cosine transform of type II for reconstructing the input vector xRN\mathbf x\in\mathbb R^N, N=2JN=2^J, with short support of length mm from its discrete cosine transform xII^=CNIIx\mathbf x^{\widehat{\mathrm{II}}}=C^{\mathrm{II}}_N\mathbf x if an upper bound MmM\geq m is known. The resulting algorithm only uses real arithmetic, has a runtime of O(MlogM+mlog2NM)\mathcal{O}\left(M\log M+m\log_2\frac{N}{M}\right) and requires O(M+mlog2NM)\mathcal{O}\left(M+m\log_2\frac{N}{M}\right) samples of xII^\mathbf x^{\widehat{\mathrm{II}}}. For m,MNm,M\rightarrow N the runtime and sampling requirements approach those of a regular IDCT-II for vectors with full support. The algorithm presented hereafter does not employ inverse FFT algorithms to recover x\mathbf x.

Keywords

Cite

@article{arxiv.1807.07397,
  title  = {Real Sparse Fast DCT for Vectors with Short Support},
  author = {Sina Bittens and Gerlind Plonka},
  journal= {arXiv preprint arXiv:1807.07397},
  year   = {2018}
}

Comments

28 pages, 5 figures

R2 v1 2026-06-23T03:07:21.680Z