a1840_1.

pdf
OThE2.pdf

Network Management Solution for PS/PON, WDM/PON

and Hybrid PS/WDM/PON using DS-OCDM
Habib Fathallah and Leslie A. Rusch

Centre d Optique Photnique et Laser, Laval University, Qubec (Qc), G1V 4T8, Canada
E-mail: [email protected]

Abstract: A novel network management solution for PS/PON, WDM/PON and PS/WDM/PON is
proposed and developed. Modified DS-OCDM overcomes OTDR shortcomings and capacity limitation of
known PON management techniques for standard and advanced PON architectures.
2006 Optical Society of America

OCIS codes: (060.2330) Fiber optics communications; (060.4510) Optical communications

1. Introduction
Passive Optical Networks (PONs) seem to be the ultimate winning solution for tomorrow s last/first mile bottleneck.
Started from 1:1 (1 fiber to 1 customer) in the early 90s, passive splitters (PS) enabled up to 1:128 for the GPON
standard with forward error correction (FEC). PON managers, however, still lack efficient monitoring technology
appropriate for PS based PON, and its derivatives[1].
In this paper, we focus on some of the In-Service management problems and propose, for the first time to our
knowledge, a simple optical code division multiplexing (OCDM) based architectural scheme that overcomes not
only the PS/PON problems, but are also applicable to more advanced PON architectures like wavelength division
multiplexing WDM/PON and hybrid PS/WDM/PON (where a PS/PON like system is built over every wavelength in
a passive WDM network). Note that, in our application we use the OCDM codes to carry management information
and not the data itself. This makes the proposed OCDM scheme easy to implement, as low transmission rates
suffice.
2. An OCDM solution for In-Service PS/PON Management
Standard OTDR based on Rayleigh backscattering and power reflections, used to monitor point-to-point links, are
ineffective in PS/PONs [1-4]. Each branch termination connected to an optical network unit (ONU), as well as every
splice, connector and fiber default or break contributes to a reflection peak and/or power loss step. If two equidistant
branches experience failure they are indistinguishable. Even when a fiber-fault results in a clear event, the faulty
branch is not identified, requiring a track-roll tour and outside intervention of technicians. Every branch must be
checked separately from its end by means of upstream OTDR transmission in order to identify the faulty one [1].
Few solutions for In-Service PS/PON management have been proposed [2-3]; all of them impractical. For
instance, a unique signature (or identifier) may be assigned to each leg of the network, before the ONU. A discrete
Bragg grating with a unique wavelength is placed at each ONU. At the CO side, different interrogation techniques
have been proposed, including a broadband source, multi-wavelength laser and tunable laser and filters etc. [2-4].
These systems are impractical for high numbers of subscribers due to the very large spectrum to be sliced, i.e., one
slice for every network leg. For 32 home customers (respectively 128 in GPON with FEC), and using narrow slice
width of 0.8 nm, a total bandwidth of 25.6 nm (respectively 102.4 nm) is required [2-3].
Our OCDM management technique requires only one wavelength, preferably in the 1650 nm already reserved
for standard FTTH PS/PON monitoring. Our technique is also scalable for more advanced and complex PON
networks always using only a single wavelength. As illustrated in Fig. 1, every network leg is terminated by a
standard passive wavelength selector (WS), widely used in PONs, isolating the standard monitoring U band from the
other data bands at the front of every ONU and at the CO as well.

Fig. 1. OCDM based PS/PON monitoring system, every network leg is assigned an encoder, one tuneable decoder at the CO.

OSA 1-55752-830-6

a1840_1.pdf
OThE2.pdf

2 PSs of 1:2q
2 PSs of 1:2q, 1 circulator
1 PS of 1:22q
1 PS of 1:2q
1 multiple FBG
3+6q
3+6q
3(q+1)
7 to 10
6q+
(a)
(b)
(c)
(d)
(e)
Fig. 2. Encoding designs: (a) standard encoder, (b) modified encoder, (c) 2q+1 splitter encoder, (d) 2q splitter encoder and (e) MBG encoder.

Components
Loss in dB

The transmission section of the OCDM monitoring equipment at the CO (Fig. 1) consists of a U band pulsed
laser driven by a processor to transmit 1 ns (or shorter) pulses with a predetermined low frequency rate (few
megahertz or lower). Every pulse propagates through the tree network, split at the PS, is coded by the encoders
referred as Enci, i=1 to N, and is then reflected back to the CO. An encoder consists of a passive device that
fragments an incoming pulse to a number of p sub-pulses and distributes them in time according to a specific code,
i.e., direct sequence coding in the time domain. Every encoder at the leg termination implements a unique code. Our
proposed network management system is a modified form of direct sequence DS-OCDM, and our codes are also the
so-called optical orthogonal codes (OOC). The PS (coupler) combines the upstream encoded pulses together as in
standard OCDM. In the CO, a tunable DS-decoder, consisting here of an optical switch and a bank of fixed
decoders, (similar to encoders but introducing delays in reverse order) discriminates responses coming from
different branches of the tree network. Every healthy branch in the network contributes an auto-correlation peak. A
missing autocorrelation peak indicates the corresponding network leg is broken or exhibits abnormal power loss.
3. Encoder design
Our DS-OCDM network management technique is different from standard DS-OCDMA technique in various
aspects: 1) low required transmission rate, dramatically alleviating the known implementation challenges of
OCDMA, 2) no reception, no transmission or modulation at the ONU side, only the coding operation is needed, and
3) while standard DS encoders Fig. 2(a) work in transmission, in our case, reflection is the key encoder property.
We propose four designs of modified DS-encoder, all appropriate for our application, in Fig. 2(b) to 2(e). All
exploit splitters and delay lines to fragment the incident pulse into sub-pulses, disperse them, gather them back to
the fiber and return them to the CO. Each of these encoder designs has its own advantages and drawbacks. The first,
Fig. 2(b), is the straightforward modification of the standard one of Fig. 2(a), entails 2 x 1:2q splitters and one
circulator. The second implementation in Fig. 2(c) requires one 1:2q+1 splitter and saves the circulator. The
implementation in Fig. 2(d) requires only one 1:2q splitter and, finally, Fig. 2(e) is an implementation uses a multiple
Bragg grating, i.e., a series of discrete gratings at the same wavelength but with different reflection intensities and
physical locations [7].
We define three key design parameters for our system: 1) high coding capacity (the number of different codes
should be a minimum of 128 for standard GPON with FEC); 2) the modified DS-encoders to be located at the
network legs should be low cost and low component count; and 3) the power loss between the incident pulse and
reflected coded signal fed back to the fiber should be minimized.
The encoder in Fig. 2(b) is the most expensive because of the circulator (with 1.5 dB loss per pass) and exhibits
the highest power loss. The encoder in Fig. 2(d) is less expensive, however exhibits very high loss due to the
reflection coefficient , typically about 4% (13 dB) [8]. The encoder in Fig. 2(c) is less expensive than that in
Fig. 2(e) because of the Bragg gratings compared to PSs, however MBGs has less insertion loss.
1

Fig. 3. Implementation of OCDM management in advanced PONs: (a) WDM/PON and (b) PS/WDM/PON.

OSA 1-55752-830-6

a1840_1.pdf
OThE2.pdf

BPON, EPON, GPON (16 ONUs)

BPON, EPON, GPON (32 ONUs)
GPON (64 ONUs)
GPON with FEC (128 ONUs)

GPON (64 ONUs)

GPON with FEC (128 ONUs)
4 WDM, 1 GPON per (4128ONUs)
7 WDM, 1 GPON per (7128 ONUs)

Fig. 4. Coding system design tool: (a) PS/PON systems, (b) PS/WDM/PON systems.

4. Applicability to WDM/PON and PS/WDM/PON

In Fig. 3, we develop an architectural solution that allows our technique to manage more advanced and complex
PON architectures. In WDM/PON, every customer is served by dedicated wavelength [5-6]. Previously proposed
PON management uses half of wavelengths for data and the other half for monitoring. Every customer is dedicated
two wavelengths, one for data and another for monitoring. A wavelength selector is place before the WDM
demultiplexer in order to separate the monitoring band at the 1650 nm. Upstream and downstream data is assumed
here, to share the same fiber; other variants could be similarly derived. Our technique saves all the monitoring
wavelengths and makes them available for data, hence doubling the network capacity. Only a single monitoring
wavelength is split and distributed to all clients.
5. Coding system design
Let be a collection of binary n-tuplets generated by [9] from the specification (n, w, a, c) where n is the code
length, w is the Hamming weight, a and c are used as follows. For vectors X , Y , the expressions
R X ( ) =

n
i =1

X (i )X (i + ) a and R XY ( ) =

n
i =1

X (i )Y (i + ) c define respectively the autocorrelation and the

cross-correlation functions, both for n + 1 n 1 . Note that the code weight w (i.e., the number of ones per
code) is equal to the number of delay lines in the encoding devices of Fig. 2 (b) to 2(d), and the number of Bragg
gratings in the MBG encoder of Fig. 2 (e).
In [9], the authors recently developed an OOC generating technique, so called Outer-product matrix based
algorithm, that maximizes the number of codes for given weight w and length n and unit cross-correlation and autocorrelation side-lobes (a=c=1). We used this algorithm to develop a coding system. In Fig. 4(a) and 4(b) we show
the number of codes (i.e., clients) vs code length n for different code weights w. From Fig. 4(a), a designer could
determine the appropriate (w, n) pairs to support EPON, GPON etc. For w=4, four delay lines are required, the
encoder of Fig. 2(c) uses one 1:8 PS (i.e., 1:23 PS with q=2) and the MBG encoder of Fig. 2(e) entails four gratings.
Fig. 4 (b) presents codes that could support PS/WDM/PON of Fig. 3(b).
6. Conclusion
We proposed a novel PON management solution using modified DS-OCDM system that overcomes the OTDR
shortcomings for point to multipoint networks and the capacity limitation of previously proposed monitoring
techniques. Our solution easily scales up from PS/PON to WDM/PON and PS/WDM PON supporting close to a
thousand customers, all using only one monitoring wavelength. We developed different encoding settings for our
system and developed a design tool of appropriate codes.
8. References
[1]
[2]
[3]
[4]
[5]
[6]
[7]
[8]
[9]

Girard, Andr "FTTx PON Technology and Testing" EXFO Electro-Optical Engineering Inc., Canada 2005, ISBN 1-55342-006-3.
Ye, Chien-Hung, Chi, Sien, "Optical fiber-fault surveillance for passive optical ," OPTICS EXPRESS, vol. 13, No. 14, 11 July 2005.
Pfeiffer, et al., Monitoring and protecting the optical layer in FTTH networks, in Proceedings of the 2005 FTTH Conference, 2005.
Chan, et al., "Fiber-Fault Identification for Branched Access Networks...," IEEE Photonics Tech. Let., vol. 11, no. 5, May 1999.
Park, et al., "Optical fault monitoring method using broadband light source in WDM-PON," Electronic Let, vol. 42, no. 4, 16th Feb 2006.
Hann, et al., Monitoring technique for hybrid PSWDM-PON by using , Measurement Science and Technology, 17(2006) pp. 1070-74.
Castro, et al.,Novel Super Structured Bragg Gratings for Optical Encryption, J. of Lightwave Technology, vol. 24, no. 4, April 2006.
Derickson, Denis, Fiber Optic Test and Measurement, Prentice Hall PTR, New Jersey 07458. ISBN 0-13-534330-5.
Charmchi, et al., "Outer-Product Matrix representation of Optical Orthogonal Codes," IEEE Trans. on Comm., vol. 54, No. 6, June 2006.

OSA 1-55752-830-6