A Hilbert Transform for Matrix Functions on Fractal Domains

R. Abreu-Blaya, J. Bory-Reyes, F. Brackx, H. de Schepper, F. Sommen

Research output: Contribution to journalArticlepeer-review

5 Scopus citations

Abstract

We consider Hölder continuous circulant (2 × 2) matrix functions G 2 1 defined on the fractal boundary Γ of a Jordan domain Ω in ℝ 2n. The main goal is to establish a Hilbert transform for such functions, within the framework of Hermitian Clifford analysis. This is a higher dimensional function theory centered around the simultaneous null solutions of two first order vector valued differential operators, called Hermitian Dirac operators. In Brackx et al. (Bull Braz Math Soc 40(3): 395-416, 2009) a Hermitian Cauchy integral was constructed by means of a matrix approach using circulant (2 × 2) matrix functions, from which a Hilbert transform was derived in Brackx et al. (J Math Anal Appl 344: 1068-1078, 2008) for the case of domains with smooth boundary. However, crucial parts of the method are not extendable to the case where the boundary of the considered domain is fractal. At present we propose an alternative approach which will enable us to define a new Hermitian Hilbert transform in that case. As a consequence, we give necessary and sufficient conditions for the Hermitian monogenicity of a circulant matrix function G 2 1 in the interior and exterior of Ω, in terms of its boundary value g 2 1 = G 2 1{pipe}Γ, extending in this way also results of Abreu Blaya et al. (Bound. Value Probl. 2008: 2008) (article ID 425256), (article ID 385874), where Γ is required to be Ahlfors-David regular.

Original languageEnglish
Pages (from-to)359-372
Number of pages14
JournalComplex Analysis and Operator Theory
Volume6
Issue number2
DOIs
StatePublished - Apr 2012
Externally publishedYes

Keywords

  • Cauchy integral
  • Fractal geometry
  • Hermitian Clifford analysis
  • Hilbert transform

Fingerprint

Dive into the research topics of 'A Hilbert Transform for Matrix Functions on Fractal Domains'. Together they form a unique fingerprint.

Cite this