A new theoretical basis of higher-derivative optical differentiators

N. Q. Ngo, Siu Fung Yu, S. C. Tjin, C. H. Kam

Research output: Journal article publicationJournal articleAcademic researchpeer-review

129 Citations (Scopus)


This paper presents a new theoretical basis of higher-derivative finite-impulse response (FIR) optical differentiators for high-speed optical signal processing. A numerical differentiation algorithm, namely, the backward Taylor series expansion and the digital signal processing technique are employed, for the first time, to develop a generalized theory of a pth-derivative FIR optical differentiator. The proposed differentiators are synthesized using a simple integrated-optic transversal filter structure based on the silica-based planar lightwave circuit (PLC) technology, which is compact and compatible with fiber optics. The proposed pth-derivative differentiator only requires p+1 waveguide arms of the transversal filter. By means of computer simulations, the differentiators are shown to be capable of processing Gaussian pulses with high accuracy. We also show the application of a first-derivative differentiator for optical dark-soliton detection and for the generation of an ultrashort optical pulse train. Although the analysis is directed at integrated optical differentiators, the methodology and the results are applicable to other physical systems such as free-space and guided-wave optical signal processors.
Original languageEnglish
Pages (from-to)115-129
Number of pages15
JournalOptics Communications
Issue number1-3
Publication statusPublished - 15 Jan 2004
Externally publishedYes


  • Dark-soliton detection
  • Optical differentiators
  • Optical signal processing
  • Ultrashort pulse train generation

ASJC Scopus subject areas

  • Electronic, Optical and Magnetic Materials
  • Atomic and Molecular Physics, and Optics
  • Physical and Theoretical Chemistry
  • Electrical and Electronic Engineering


Dive into the research topics of 'A new theoretical basis of higher-derivative optical differentiators'. Together they form a unique fingerprint.

Cite this