IEEE JOURNAL OF QUANTUM ELECTRONICS, VOL. 39, NO. 1, JANUARY 2003 91
Optimization of Refractive Index Sampling for
Multichannel Fiber Bragg Gratings
Alexander V. Buryak, Kazimir Y. Kolossovski, and Dmitrii Yu. Stepanov
Abstract—A comprehensive analysis of multichannel grating op-
timization strategies is presented. The central idea is in dephasing
of partial gratings with respect to each other. This dephasing al-
lows the utilization of ultraviolet-induced refractive index changes
with maximum efficiency. The dependence of group delay ripple on
optimization strategy, number of channels, and other grating char-
acteristics is also briefly discussed. Finally, we discuss further gen-
eralization of the dephasing approach to the case of multichannel
gratings with nonidentical spectral characteristics.
Index Terms—Fiber Bragg grating, multichannel.
I. INTRODUCTION
S
INCE early experiments in the 1970s [1], fiber Bragg grat-
ings (FBGs) have become an essential part of fiber optics
communications. They are used as filters, strain and tempera-
ture sensors, laser source tuners, stabilizers, etc. [2]. We specif-
ically emphasize the importance of FBG applications in wave-
length-division-multiplexing (WDM) systems which have at-
tracted growing attention over recent years because of their ob-
vious advantages when upgrading system capacity. Such sys-
tems may rely on multichannel Bragg gratings—optical com-
ponents that provide a series of wavelength channels with ac-
curate inter-channel separations and customized, e.g., identical
in-band specifications [3]–[6]. Recently, the problem of multi-
channel FBG optimization has been put on the agenda for laser
applications [7], [8] and for passive FBG-based devices [9],
[10]. Additionally, a conceptually similar problem has attracted
considerable attention in radiophysics (see, e.g., [11] and [12]).
For FBG applications, the main scope of such an optimization is
to reduce the refractive index change required for manufacturing
a multichannel FBG. In this work, we significantly extend the
analysis of [9], [10] and present a comprehensive description of
different optimization strategies.
Before going into a detailed description of our findings, we
summarize some basic results for FBGs and define our nota-
tions. The fundamental system of equations describing light
propagation in FBGs is
(1)
Manuscript received April 19, 2002; revised August 19, 2002.
A. V. Buryak is with Redfern Optical Components Pty. Ltd., Eveleigh, NSW
1430, Australia.
D. Yu. Stepanov was with Redfern Optical Components Pty. Ltd., Eveleigh,
NSW 1430, Australia. He is now with Bandwidth Foundry Pty. Ltd., Eveleigh,
NSW 1430, Australia.
K. Y. Kolossovski is with the School of Mathematics, University of New
South Wales at ADFA, ACT 2600, Australia.
Digital Object Identifier 10.1109/JQE.2002.806202
where and are the amplitudes of the forward and
backward propagating fields, respectively, is a normalized
frequency detuning from the central Bragg reflection frequency,
is a local distance along FBG, and is a spatial profile of
the FBG coupling coefficient. For a reciprocal and lossless FBG
of length (see, e.g., [13] for definitions), we may find complex
reflection and transmission coefficients from a transfer matrix
, which relates field values at the grating ends
(2)
and is given by
(3)
where is the reflection coefficient measured at
and is the complex transmission coeffi-
cient. Conventional reflection and group delay (
depending on grating side) are related to the complex reflection
coefficients as , where is the
light wave frequency. Similar relations exist for transmission ,
group delay in transmission and the complex transmission co-
efficient .
II. FORMULATION OF THE PROBLEM
Before one designs any multichannel grating with equal
inter-channel separations and identical in-band specifications,
a corresponding single-channel grating design should be
constructed. Below we refer to this initial single channel
grating as a seeding grating. For any physically viable spectral
response in reflection of a single-channel FBG device, a
so-called inverse scattering problem should be solved, e.g., by
applying a layer-peeling algorithm (LPA) [14]–[16] to system
(1). As a result of solving the inverse scattering problem, we
obtain a grating design , where
is the grating amplitude and is the grating phase. The
grating amplitude is normalized to be measured in cm
and is related to the FBG effective refractive index modulation
amplitude as , where is
the FBG period and is the average refractive index. In the
literature, FBG grating amplitude and phase are often referred
to as apodization and chirp profiles, respectively.
In a vast majority of previously reported work on multi-
channel gratings, a so-called sinc-sampling approach has been
used. In this method, the amplitude (but not the phase) of a
given single-channel grating is periodically modulated. For
0018-9197/03$17.00 © 2003 IEEE