scieee Science in your language
[en] (orig)

The soliton transmissions in optical fibers

Abstract

The objective of this paper is to familiarize readers with the basic analytical propagation model of short optical pulses in optical fiber. Based on this model simulation of propagation of the special type of pulse, called a soliton, will be carried out. A soliton transmission is especially attractive in the fiber optic telecommunication systems as it does not change a pulses shape during propagating right-down the fiber link to the receiver. The model of very short pulse propagation is based on the numerical solution of the nonlinear Schroedinger equation (NLSE), although in some specific cases it is possible to solve it analytically.

Read accessible full text

The soliton transmissions in optical fibers

Author: Boháč, Leoš
Publisher: Vysoká škola báňská - Technická univerzita Ostrava
Year: 2010
Source: https://dspace.vsb.cz/bitstreams/8b2a45c8-ac8e-48f8-92c0-8178b896b500/download
INFORMATION AND COMMUNICATION TECHNOLOGIES AND SERVICES, VOL. 8, NO. 5, DECEMBER 2010 107
© 2010 ADVANCES IN ELECTRICAL AND ELECTRONIC ENGINEERING ISSN 1804-3119
THE SOLITON TRANSMISSIONS IN OPTICAL FIBERS
Leoš BOHÁČ.1
1 Depa men o elecommunica ion enginee ing, Czech echnical uni e si y in P ague, P ague, Czech epublic Technicka
2, P ague 6, 16627
[email p o ec ed]
Abs ac . The objec i e o his pape is o amilia ize
eade s wi h he basic analy ical p opaga ion model o
sho op ical pulses in op ical ibe . Based on his model
simula ion o p opaga ion o he special ype o pulse,
called a soli on, will be ca ied ou . A soli on ansmission
is especially a ac i e in he ibe op ic elecommunica ion
sys ems as i does no change a pulses shape du ing
p opaga ing igh -down he ibe link o he ecei e . The
model o e y sho pulse p opaga ion is based on he
nume ical solu ion o he nonlinea Sch oedinge equa ion
(NLSE), al hough in some speci ic cases i is possible o
sol e i analy ically.
Keywo ds
Soli on, ul a sho pulses, NLSE, dispe sion,
Gaussian beam.
1. In oduc ion
Soli a y wa es, some imes simply called soli on, ha e
been a opic o heo e ical and expe imen al s udy o many
yea s. Though his pape is ela ed o ib e op ics, in eali y
soli a y wa es exis also in o he ield o science like
hyd odynamics, biology and plasma physics. His o ically
he one who i s obse ed a soli on wa e was James Sco
Russel in 1834 when he acciden ally no iced in he na ow
wa e canal a smoo hly shaped wa e heap ha o his
su p ise was able o p opaga e in he canal wi hou a
appa en change in i s shape a ew kilome es along. The
essence o p opaga ion o his soli a y wa e was no a long
ime unde s ood un il app op ia e ma hema ical model was
concei ed in he 1960’s oge he wi h a way o sol ing
nonlinea equa ion wi h he help o in e se sca e ing
me hod.
Now le us go back o he ield o op ics. Gene ally
speaking, he e exis wo o ms o soli a y wa es,
depending on whe he he ligh is being con ined in space
o ime. I he i s is he case wa e is e e ed o as spa ial
soli on o in he second case as empo al soli on. Soli on
o ming phenomenon s ems om nonlinea p ope ies o
medium whe e a pa icula wa e is p opaga ing. Namely,
in a ield o op ics i is Ke e ec ha is esponsible o
op ical nonlinea i ies. In he case o spa ial soli on he
na u al p ope y o ligh o dispe se in space is being
p oac i ely compensa ed by he nonlinea i y o he medium
in such a way ha highe in ensi y pa o an op ical beam
( ypically in he cen e o Gaussian beam) inc ease a alue
o e ac i e index o medium o ming de ac o a co e o
wa eguide ha is esponsible o con ine in e e se a
dispe sed ligh o he middle o he beam i sel . I can be
easily in ui i ely unde s ood ha i he sel induced
nonlinea i y is oo high he beam will ge ocused and on
he o he hand i i is e y small o none, beam will
dispe se in space – a p e ailing si ua ion in many cases
whe e a beam does no ha e enough powe densi y o
induce nonlinea i y in a medium.
2. Ma hema ical modeling o he
soli a y wa e
The p opaga ion o ligh can be p ecisely desc ibed
ma hema ically wi h Maxwell equa ions. When equa ions
o magne ic and elec ic ields a e combined oge he one
ge [1][2]:
2
2
2
0
2
2
2
211
P
c
E
c
E









(1)
whe e c is he speed o ligh in he acuum and 0 is he
acuum pe mi i i y. The induced pola iza ion P consis s
o wo pa s such ha [1][2]:
),(),(),( P P P NLL






 (2)
whe e he linea pa L
P

and nonlinea pa NL
P

a e ela ed
o he elec ic iled by he gene al ela ions [1][2] [3][4]:
108 INFORMATION AND COMMUNICATION TECHNOLOGIES AND SERVICES, VOL. 8, NO. 5, DECEMBER 2010
© 2010 ADVANCES IN ELECTRICAL AND ELECTRONIC ENGINEERING ISSN 1804-3119
')',().'(),( )1(
0d E PL




 


(3)

  ),,(),( 321
)3(
0 P
NL


 (4)
321321 ),(),(),( d d d E E E 






whe e (1) and (3) a e he i s - and hi d- o de
suscep ibili y enso s.
3. P opaga ion o soli on pulse in
op ical ibe s
To be e unde s and a soli on pulse p opaga ion in
op ical ibe i necessa y o se up ou modelling on he
ma hema ical exp ession (1). We will suppose, ha a
solu ion o elec ic iled E ha e a o m [1]:
)exp(),(),(),( 0ZiYXF ZA E

 (5)
whe e F(X,Y) is ans e se ield dis ibu ion ha
co esponds o he undamen al mode o single mode ib e.
A(Z, ) is along p opaga ion axis Z and on ime dependen
ampli ude o he mode. A e some ma h manipula ions
one can come o he equa ion ha go e ns pulse
p opaga ion in op ical ib es [1]:
AAi
A
i
A
Z
A2
2
2
2
12












(6)
The pa ame e s 1 a 2 include he e ec o dispe sion
o i s and second o de s, espec i ely. Physically,
1=1/ g, whe e g is g oup eloci y associa ed wi h he
pulse and 2 akes in o accoun he dispe sion o g oup
eloci y. Fo his eason, 2 is called he g oup eloci y
dispe sion (GVD) pa ame e .
Pa ame e  is nonlinea pa ame e ha akes in o
accoun he nonlinea p ope ies o a ibe medium.
Pa ame e 1 is in eal case always posi i e bu on he
o he hand pa ame e s ZD and  can be in some speci ic
case ei he posi i e o nega i e. The pa ame e 1 is closely
associa ed in p ac ice wi h be e known pa ame e called
dispe sion pa ame e – D (ps/nm/km). The ela ion
be ween hem is in he o m [1]:
2
2
21




c
d
d
D
g









 (7)
As we know, dispe sion pa ame e D is a mono onically
inc easing unc ion o wa eleng h, c ossing a ze o poin a
wa eleng h ZD, which is called a ze o ch oma ic dispe sion
wa eleng h. I a sys em ope a es wi h wa eleng hs abo e
ZD, whe e D is posi i e, 2 mus be nega i e and a ibe is
said o wo k in anomalous dispe sion mode. I a ibe is
ope a ed below ZD, he D is nega i e and 2 mus be
posi i e. In his case a ibe is said o ope a e in no mal
dispe sion mode. As ega ds he nonlinea pa ame e  i
can be gene ally ei he posi i e o nega i e, depending on
he ma e ial o he wa e guide. Fo silica ibe is pa ame e
 posi i e bu o some o he ma e ials i can be nega i e.
Mo e speci ically, equa ion (6) has only wo solu ion, in
he o m o ei he da k o b igh soli on. The b igh soli on
co esponds o he ligh pulse bu da k soli on is a he a
pulse shaped dip in CW ligh “backg ound”. In o he
wo ds, he da k soli on is in a ac nega ion o he b igh
soli on. Whe e he e is maximum o ligh in he b igh
soli on, he e is minimum o he ligh in he da k soli on
and ice e sa.
The b igh soli on can p opaga e in only such a
wa eguide whe e he e is ei he he posi i e nonlinea i y
pa ame e  and anomalous dispe sion o he nega i e
nonlinea pa ame e bu no mal dispe sion. Fo a classical
silica ibe he i s is he case.
4. The soli on pulse and he simula ion
o i s p opaga ion in he op ical ibe
Equa ion (6) can be no malized in he o m:
0
2
2
2
2




uu
us
z
u
i

(8)
using his ans o ms:
O
TZ /)( 1




, D
LZz /

, ALu D

 (9)
INFORMATION AND COMMUNICATION TECHNOLOGIES AND SERVICES, VOL. 8, NO. 5, DECEMBER 2010 109
© 2010 ADVANCES IN ELECTRICAL AND ELECTRONIC ENGINEERING ISSN 1804-3119
whe e T0 is mo ing ime window wid h ( e y o en se o
he pulse wid h) and 2
2
0/

TLD is dispe sion leng h.
Using in e se sca e ing me hod e eals ha solu ion o
abo e men ioned equa ion has a o m:
2/2/ )(sech
2
),( iziz eNe
ee
Nzu
 


 (10)
I N is in ege , i ep esen he o de o he soli on pulse.
Ve y in e es ing si ua ion comes when N=1. In his case o
i s o de soli on, he pulse does no change i s shape a all
as i p opaga es in op ical ibe . In con as when N is
highe hen one, pulse shape is no s able and change
pe iodically wi h soli on pe iod D
LZ 2
0

. A he end
o e e y pe iod Z0 he soli on esembles i s ini ial simple
pulse shape. I is e iden ha o elecommunica ion
pu poses is he soli on o i s o de mos sui able, because
in his applica ion is necessa y o keep a pulse shape s able.
Pa ame e N, which de ines he soli on o de can be
u he exp essed by:
2
0


P
TN O
 (11)
whe e T0 [s] co esponds o inpu pulse wid h, P0[W] is
pulse peak powe 2

[ s2/m] akes in o accoun g oup
eloci y dispe sion and [(Wm)-1] is nonlinea pa ame e o
he ibe ma e ial.
I ha e s udied a p opaga ion o soli on pulses wi hin he
op ical ibe using a simula ion ool Op sim o m ARTIS
In his case a equa ion (6) is sol ed nume ically using a
spli -s ep-Fou ie me hod. The scheme used is shown on
Fig.1. As can be seen, I ha e used a soli on gene a o ( he
mos le side icon) ha nume ically gene a es sequence o
sech( ) pulses. The pulse wid h was se o 10 ps wi h pe iod
o 400 ps. I ha e used a s anda d single mode ibe model
wi h a ibe leng h o 100 km. I ha e se up wo peak
powe s o he pulse, namely 100 mW and 166 mW. As
pulses sa is y a condi ion o he soli on shape, i was only
necessa y o adjus app op ia e pulse wid h and peak
powe . Acco ding o he
Fig.1 – Simula ion o soli on in ibe
o mula (11) and o he case o i s o de soli on, we
need:


2
0
2
0
1T
PN  (12)
when eal alues a e subs i u ed o he abo e equa ion, in
pa icula T0=10 ps, 2= -20 ps2/km and =1,2 W-1km-1 he
adequa e powe o each i s o de soli on egime is
app oxima ely 166 mW. In o he wo ds i a sech( ) pulse
ha e his peak powe one should a leas acco ding o
heo y ha e a i s o de soli on ha is ansmi ed h ough
a ibe unchanged in shape. To e i y his I ha e used in
abo e model a ibe wi hou a loss, i is possible o cou se
only in simula ions no in he p ac ice. In he eali y i
would be necessa y o deploy op ical ampli ie s (like
EDFA o Raman, o bo h in combina ion) o o e come he
loss in he eal ibe link and keep he powe o he soli on
in equi ed limi s, o he wise he pulse will s a o sp ead
again, he e ec o nonlinea i y will be no so s ong o
e ec i ely supp ess unwelcome in luence o he ibe
ch oma ic dispe sion.
To illus a e c ea ion o he soli on I ha e used he scheme
on Fig.1. The pulse shape can be seen on Fig. 2. The igh
couple o o e lapping (highe one abo e o he ) pulses
co esponds o he inpu . These pulses we e in sequence
sen o 100 km sec ion o he single mode ibe link.
110 INFORMATION AND COMMUNICATION TECHNOLOGIES AND SERVICES, VOL. 8, NO. 5, DECEMBER 2010
© 2010 ADVANCES IN ELECTRICAL AND ELECTRONIC ENGINEERING ISSN 1804-3119
Fig.2 – Resul s o simula ion
The one wi h he lowe powe (100 mW) ela es o he
lowe ou pu pulse ha is shown a he le side o he
pic u e. I is e iden ha in his case he pulse unde goes
sp eading caused by ch oma ic dispe sion and peak powe
is no ye high enough o compensa e dispe sion. On he
o he hand, i one inc ease op ical powe i is possible o
ge o he poin , whe e he nonlinea i y en i ely compensa e
dispe sion and he pulse p opaga es in he ibe link
wi hou any change in he shape, excep a small inc ease in
i s ampli ude. I is wo h o men ion igh now, ha his
inc ease in ampli ude is empo al phenomenon because i
one would ha e s udy he pulse p opaga ion mo e ca e ully
i would be e ealed ha i s peak oscilla es un il eaching a
poin o s eady ampli ude. This happens mos ly in cases
when he inpu pulse does no esemble he exac soli on
o m. The beau y o a soli on is among o he hings in i s
abili y o eassemble o iginal shape despi e o some
dis u bing ac o s ac ing upon i .
5. Conclusion
In his pape a e y basic analysis o soli ion
ansmission and i s dynamics in he op ical ibe was
pe o med. I has been shown by simula ion ha b igh
soli on can o m in classical op ical single mode ibe and
in he case o ze o loss i can p opaga e wi hou a change
o i s shape. As esul i is possible o o e come issue wi h
in luence o he dispe sion on pulse sp eading and achie e
a much longe ansmission dis ances and also inc ease line
capaci y. Howe e , he applica ion o soli on ansmission
is no en i ely wi hou p oblems. Soli on pulses should be
apa in ime conside ably o a oid excessi e o e lapping.
I soli on’s o e lap is no su icien ly supp essed soli ons
will ha e an e ec on he ansmission o hemsel es in a
way o cause indi idual soli on’s g oup eloci y o a y
along he line and esul ing in inc eased sys em ji e .
6. Acknowledgmen s
The wo k was suppo ed by g an The Resea ch in he
A ea o he P ospec i e In o ma ion and Communica ion
Technologies unde p ojec MSM6840770014.
Re e ences
[1] KIVSHAR, Yu i S., AGRAWAL, Go ind P. Op ical Soli ons :
F om Fibe s o Pho onic C ys als. San Diego : Academic P ess,
2003. 540 s. ISBN 0-12-410590-4.
[2] PORSEZIAN, K., KURIAKOSE, V.C. Op ical Soli ons :
The o ical and Expe imen ak challenges. 1s edi ion. Be lin :
Sp inge , 2003. 406 s. ISBN 3-540-00155-7.
[3] AGRAWAL, Go ind P. Applica ions on Nonline a Fibe Op ics.
San Diego: Academic P ess, 2001. 458 s. ISBN 0-12-045144-1.
[4] GAGLIARDI, Robe M., KARP, Sha man. Op ical
Communica ions. 2nd edi ion. New Yo k : Wiley, 1995. 347 s.
ISBN 0-471-54287-3.
Abou Au ho s ...
Leoš BOHÁČ ecei ed he M.S. and
Ph.D. deg ees in elec ical enginee ing
om he Czech echnical uni e si y in
P ague in 1992 and 2001, espec i ely.
F om 1992 he has been eaching Op ical
communica ion sys ems and Da a
ne wo ks a he same uni e si y. His
esea ch in e es is an applica ion o he
high speed op ical ansmission sys ems in a da a ne wo k.
He has also pa icipa ed in op ical esea ch p ojec s in
CESNET, he academic da a ne wo k p o ide , o help
implemen long haul high speed op ical esea ch ne wo k.