I need help. After spending uncounted hours studying tunable lasers and tunable filters, I'm still not sure how they interact. It appears Queensgate provides tunable filters. Are these used with Raman amplifiers/modules? If so, where do tunable lasers fit into the picture? I guess what I'm trying to figure out is whether this is a market SDL will enter, or if it's one they'll augment.
It appears that tunable lasers (not sure about filters) will lessen the need for massive cross-connects. There's reference to a "switchless mesh architecture."
Feeling a little like Hobbs:
weewave.mer.utexas.edu
Okay, here's what I've dug up so far:
I Papers from OFC2000:
<<<
Sprint Perspective on Next Generation Multi-service Transport and Switching Networks
Robert K. Butler, Ph.D.
Sprint
9300 Metcalf Ave, Mailstop: KSOPKB0803, Overland Park, KS 66212
Telephone: (913) 534-5838, Fax: (913) 534-3485
Email: robert.k.butler@mail.sprint.com
Introduction
Transmission requirements are evolving at a rapid pace as the network is expanded to increase customer base, extend into new markets, and to increase the bandwidth and products offered to current customers. Although efficiently creating the required bandwidth is a daunting
task, advances in transmission and switching technology are creating new options for deployment. The economics associated with consolidating existing networks into a single multi-service network is a significant driver for change.
Efficient multi-service platforms are being incorporated into the network and systems with unprecedented scalability, flexibility, and associated economics are required to meet Sprint?s needs. The optical transport networks carrying this traffic must support the requirements of voice, Asynchronous Transfer Mode (ATM), frame relay, Internet Protocol (IP), and video services. As advances in technology continually produce new implementation options, standards bodies struggle to keep pace. The carrier dilemma is whether to deploy proprietary solutions or wait until
standards are developed.
Multi-Service and Wavelength Based Networks
The objective of deploying multi-service networks is to transport TCP/IP, video, voice, frame relay, ATM and other signals on a common platform. This allows for the optimization of a single network and hopefully, reduction in the number of network elements, the simplification of management, and the reduction of cost. Sprint has been aggressively developing multi-service platforms.
The alternative to multi-service platforms is to implement overlay networks. Although this allows for the development of networks specifically optimized for a particular type of traffic, a larger number of network elements is required and more networks must be managed. The desire
to consolidate these networks has lead to the concept and deployment of the multi-service network.
As related to optical fiber communication, the utilization of multi-service networks means that many types of traffic with different transport requirements will be consolidated into a single wavelength. The system utilized for transport must meet the most stringent needs of any service
provided. Therefore, Sprint must necessarily provide high capacity, high reliability transmission with fast protection mechanisms for wavelengths carrying multi-service traffic.
Today, the network consists of many optical wavelengths operating in each transmission fiber. Instead of increasing the Time Division Multiplexing (TDM) rate to get more capacity,overlaid 2.5 Gb/s Synchronous Optical Network (SONET) rings supported by Wave Division Multiplexing (WDM) have been deployed. The current long haul transport network based on SONET Add-Drop Multiplexers (ADMs) cross-connects at STS-1 granularity. The amount of multi-service data being carried on each optical fiber continues to increase and contemplating a higher granularity than STS-1 is reasonable. Another reason to pursue alternatives is that overlaying independent SONET rings eventually results in scaling and management issues. The alternative solution is to deploy a wavelength-based network. Conceptually, this is a fiber optic communication network that provides Operation, Administration, Maintenance and Provisioning (OAM&P) and restoration functions at the granularity of the optical wavelength. Switches and routers producing the multi-service traffic with OC-12, OC-48 and eventually OC-192 interfaces must be accommodated. A large number of wavelengths must be supported even if the TDM rate of many of the wavelengths is increased in the future to 10 Gb/s or higher.
Therefore, the management of a large number of optical wavelengths has become a significant issue.
The attributes of these wavelength-based networks must be able to support the stringent requirements of the multi-service traffic carried in the Sprint network. The availability of the network must be in accordance with that of traditional voice networks, generally accepted to be
99.999%. The protection switching required to provide the required availability must be automatic and provide ~50 ms restoration. The OAM&P features must be advanced for the initial deployment as the network will quickly grow to maximum capacity. Accordingly, provisioning
new services across the network should be accomplished nearly instantaneously.
Transport Technology
Technological developments in transmission and wavelength switching are being driven by several factors. New applications and the growing utilization of existing applications are increasing the required bandwidth. The large bandwidth requirements can create congestion and
scaling issues, but also create opportunities for improved network efficiency and economy of scale. New technologies that provide increased capacity and superior economics are in demand and may be implemented before industry wide standards have been developed.
WDM is a prime example. In a 5-year period, WDM technology evolved rapidly and Sprint will have sequentially deployed 5 different, high capacity systems. Four channel systems were quickly superseded by 8 channel systems. Eight channel systems were superseded by 16 channel
systems, which fell to 40 channel systems. The deployment of 80 channel, 200 Gb/s or higher systems will begin shortly in strategic locations. Although the features and functionality of these systems are common to many carrier networks, none of these systems have been fully
standardized.
The rapid deployment of WDM equipment has created opportunities in the areas of optical networking. With the high number of wavelengths entering sites, the managed switching for provisioning and protection of wavelengths is becoming necessary. The creation of a new
wavelength-switching layer is underway and the ITU is standardizing optical networking for metropolitan and long haul applications. Although comprehensive WDM standards have lagged the deployment of equipment, the ITU is attempting to develop a standard for optical networking that is being called the Optical Transport Network (OTN). The OTN will include standards applicable to WDM technology and standards related to networks of interconnected wavelengths.
Sprint has embraced the rapid change in high bandwidth transmission technology and is working with vendors to develop full-featured optical networking solutions. Effective wavelength management systems, protections and routing algorithms and devices will be incorporated in the
network. The rate of development of high-speed transport solutions means that the technology deployed by different carriers will likely diverge significantly.
In this environment, standardizing these differing solutions at least for interoperability is critical. Signals are often passed between different carrier networks so interoperation is required. In addition, Sprint historically selects multiple vendors for each application and requires
interoperation among these vendors. As new technologies are rapidly introduced into the network, interoperability with existing and future solutions is desirable. Waiting for standards to be developed means the investment in the old technology must continue, competitors may pay less
per bit, market share may be lost, and aggressive equipment vendors may be locked into other networks. Proceeding without standards in place may lock the carrier in to proprietary solutions, reduces the likelihood of multiple vendors supporting the same application, and generally
increases the risk of deployment.
There is significant new technology under development to continue to stimulate the transport and wavelength switching market. Broadband and tunable sources are under development that may have a significant impact on economics, sparing, and network architecture. Routing
algorithms, wavelength cross-connects and wavelength routers, programmable optical add-drop multiplexers, optical wavelength conversion devices, and tunable filters may also allow network architecture to evolve for the better. Compensation devices to alleviate dispersion and
polarization-mode dispersion and advanced modulation techniques such as sub-carrier modulation, soliton transmission and forward error correction techniques will also have an impact on future systems.
Conclusions
The deployment of infrastructure providing high performance multi-service switching to support voice, ATM, frame relay, IP and video is well underway. The integration of service means that transport systems must provide the highest level of performance, restoration and protection.
Therefore, only high capacity transport systems meeting these requirements are being considered. Development cycles for new systems incorporating the latest technology continue to decrease. Carriers are always faced with new options in transmission equipment that are available
long before standards bodies are able to produce complete results. Waiting for standards to be developed means risking the loss of market share and economic benefits, while deploying the latest technology in the form of proprietary solutions entails other risks. Optical networking solutions for long haul and metropolitan applications are examples of systems being produced before comprehensive standards are in place. Significant differences between these applications exist, but solutions for both are focusing on the optical wavelength. Therefore, it can be said that a new layer of switching has developed with the granularity of the optical wavelength. Many new devices have yet to be incorporated in even the latest products f
networking. Tunable filters and sources, wavelength interchange devices, programmable optical add-drop multiplexers, large scale optical cross-connects and other devices are being studied and will continue to feed the assortment of possibilities. The pattern of rapid development of new transport and switching platforms is here to stay.
Another paper, this from scientists/engineers in Kista, Sweden, headquarters for Altitun:
Step-wise tunability of a DBR laser with a superimposed fiber grating external cavity
Mattias Andomat, Pierre-Yves Fonjallaz, Fredrik Olofsson
ACREO AB Isafjordsgatan 22, 164 44 Kista, Sweden
Tel: + 46 8 632 7753, fax: 46 8 632 7710, email: matado@imc.kth.se
Fiber Bragg gratings (FBG) have been used in many different ways in combination with semiconductor laser diodes to improve their emission properties [1-3]. One motivation for using FBGs is their good wavelength stability in comparison with simple Fabry-Perot cavities or even with DFB or DBR diode lasers. FBG can be used to stabilize the emission wavelength of diode lasers both actively and passively. In the first case, part of the laser emission is detected after reflection on and transmission through a fiber grating. The relative intensity of both signals is used to adjust the currents settings of the diode to maintain the wavelength constant [4]. In the second case, a pigtailed FBG is used as the out-coupling
reflector of the cavity. In that situation, the laser cavity suffers from a relatively high loss due to the imperfect coupling from the diode to the single-mode fiber. Nevertheless, since the cavity is made considerably longer (factor 10 to several 100), and thanks to the spectral selectivity of the grating which allows for the selection of a single longitudinal mode, spectral linewidths of the order of few hundreds of kHz are achievable [1].
n addition to the wavelength stability and reduced linewidth obtained with FBG external cavities, there is an increasing interest in being able to switch the wavelength between standardized channels. This has recently been reported and a sampled FBG was used in combination with a simple FP cavity allowing for the switching between eight 100 GHz-spaced channels [1]. In this paper, we report on the wavelength stabilization and step-wise tunablity of a DBR laser with the help of a superimposed fiber grating. We believe it is a promising configuration to obtain robust and high quality step-wise tunable external-cavity semiconductor lasers (ECSL).
Superimposed FBG (SIFBG) are fabricated by exposing the fiber with subsequent interference pattern of varying modulation period [5]. It corresponds to fabricating a first grating and re-exposing the same portion of fiber to generate other resonance wavelengths. When the second exposition is realized, the Bragg reflectivity of the first grating is reduced. When the third exposition occurs, the two first resonances are reduced and so on. By adjusting the dose of the subsequent expositions, it is possible to get constant reflectivity peaks without side lobes. Superimposed gratings are much easier to fabricate than sinc-sampled gratings [6]. The main problem is the wavelength control since the frequency separation is
not automatically constant, as it is the case for sampled gratings. For this experiment, a SIFBG has been fabricated, with three 40% reflectivity peaks and 100GHz spacing, using the multiple fringe printing technique [7]. The length is 40 mm and the apodization was chosen to be gaussian.
This SIFBG was combined with a DBR laser whose front facet was AR-coated to a residual reflectivity of about 1% (with a single layer of zirconium dioxide). The rear facet was left as cleaved. In order to increase the coupling efficiency to the DBR , the fiber had a lensed shaped extremity but was not AR-coated. The coupling efficiency was measured to be about 35%. The distance between the DBR and the center of the SIFBG was 25 mm.
In order to investigate the tunability of the proposed external cavity laser, the current of the Bragg (IBragg) and phase (IPhase) sections of the DBR semiconductor laser were both swept using a computer controlled current supply. During the currents sweep, at every new value of (IBragg, IPhase) the values of power and wavelength were measured. The results can be seen in Figures 1 and 2 (Igain = 50mA). Fig.1 wavelength mapping Fig.2 power mapping Fig.3 Cross-section of the wavelength mapping Fig.4 Cross-section of the wavelength mapping In figure 1, regions of constant wavelength (represented with grayscale) can be seen. This is even clearer in figures 3 and 4, which present sweeps of the Bragg current only. This shows that the wavelength tuning of the laser is robust with respect to tuning currents, which in turn makes it insensitive to degradation. Simultaneously, Figure 2 shows that those regions of constant emission wavelength also correspond to maximum output power. Lasing at other tuning currents is due to insufficient AR-coating. The maximum output power is 0.3mW, which is an order of magnitude lower than the power at the
same working point for the DBR-laser alone before AR-coating. This is due to the 35% coupling efficiency between DBR and SIFBG.
Fig.5 Laser spectrum ( Ibragg = 46.2mA Igain = 40mA Iphase = 2.5mA T = 20.2 øC ) In this experiment side mode supression ratio (SMSR) was measured for a number of operating points. Values above 30dB were easily obtained. The linewidth was determined to be 0.5MHz in a self
homodyne delayline measurement. The main benefit of this combination of a DBR and a SIFBG is the low sensitivity of the emission
wavelength to current variations. In particular, since the phase can vary rather much, such a device can probably be directly modulated with ultra low chirp [3]. The parasitic reflections inside the cavity should be reduced significantly in order to get a better insight of the maximal SMSR achievable. For direct modulation of such a device, the laser cavity has to be relatively short. Since the index profile of the SIFBG is not limited by the spatial resolution of the fabrication process, short multiple-peak FBGs can easily be fabricated. However, the maximum number of reflection peaks with a superimposed grating is only limited by the fiber?s photosensitivity to get reasonable reflectivities. Therefore, such a device might be improved considerably with respect to the number of stabilisation channels, output power and SMSR.
An external cavity DBR laser has been built with a superimposed FBG. Step-wise tunability over three wavelength channels has been demonstrated with excellent wavelength stability. Such a combination of a DBR and a SIFBG seems extremely promising regarding the relatively low
quality of both fiber-DBR coupling and AR coatings in the case presented here.
[1] J.-F. Lemieux et.al., ?Step-tunable (100GHz) hybrid laser based on Vernier effect between Fabry-Perot cavity and sampled
fibre Bragg grating?, Electronics Letters, vol. 35, no. 11, 1999
[2] D.M. Bird et.al., ?Narrow line semiconductor laser using fibre grating?, Electronics Letters, vol. 27, no. 13, pp. 1115-1116,
1991
[3] R. Paoletti et.al., ?15-GHz Modulation Bandwidth, Ultralow-Chirp 1.55-æm Directly Modulated Hybrid Distributed Bragg
Reflector (HDBR) laser Source, IEEE Photonics Technology Letters, vol. 10, no. 12, 1998
[4] M. Adomat et.al. ?Wavelength Stabilization of DBR Laser Using a Fiber With Multiple Gratings?, pp. 161-162, ECOC?98,
1998
[5] A. Othonos et al., ?Superimposed multiple Bragg gratings?, Electron. Lett. , vol. 30, pp. 1972-1974, 1994.
[6] M. Ibsen et al, ?Sinc-sampled fiber Bragg grating for identical multiwavelength operation?, OFC?98, pp. 5-6, 1998.
[7] P.-Y. Fonjallaz et.al. ?New Interferometer for Complete Design of UV-written Fibre Bragg Gratings?, pp. 36 38, OSA
Conference on Bragg Gratings, Photosensitivity, and Poling in Glass Fibers and Waveguides: Applications and
Fundamentals, 1997
Part of this work was financed by the Foundation for Knowledge and Competence Development. We
would like to thank Henrik hlfeldt and Adela Saavedra for useful discussions.
And, finally, a paper by Siemens, "Reconfigurable all-optical networking at terabit transmission rates"
P. Leisching, H. Bock, A. Richter, D. Stoll, and G. Fischer
SIEMENS ICN TR ON E T, Advanced Transport Systems, 81379 Mnchen, Germany
Phone: 0049-89-722-23062, FAX: -24510, patrick.leisching@icn.siemens.de
Future all-optical telecommunication networks for terabit transmission rates require the use of optical routing to cope with the increasing capacity demand due to growing internet traffic. Laboratory wavelength division multiplexing (WDM) demonstrations have shown up to 3.2 Tbit/s (80x40 Gbit/s) transmission rate [1]. In contrast to long-haul WDM systems, metropolitan networks will be optimized to offer flexible and scalable bandwidth for new services using transparent optical transmission. The highest static drop capacity reported so far for a metropolitan network is 80 Gbit/s [2]. The SIEMENS subnetwork of KOMNET (innovative communication networks for the future information society) investigates dynamic optical routing of up to 80 WDM channels. In this paper we are reporting on transparent all-optical metropolitan network experiments at 10 Gbit/s
line capacity for each channel, i.e. 0.8 Tbit/s in one direction. The system limitations imposed by the 50 GHz ITU grid at 10 and 20 Gbit/s for reconfigurable optical add/drop multiplexers (OADMs) are evaluated. The OADMs are composed of tunable fiber Bragg gratings (FBGs), an optical frequency interleaver (OFI) and optical switching matrices.
1. Experimental set-up
The experimental configuration is shown in Fig. 1. The WDM ring consists of three 20 km spans of standard single mode fiber (SSMF) and matches the upcoming field-trial KOMNET in Berlin. Dispersion compensating fiber (DCF) at the end of each span ensures fully compensated segments with a residual dispersion of less than 50 ps/nm. Channels can be added/dropped at each of the three OADMs. Fig. 1. Schematic experimental set-up of the field trial in Berlin. Fig. 2.: Wavelength comb of 80 channels added at the hub node at the input of OADM I. All OADM sites will be connected to adjacent networks: There is a long-haul transmission system and an access network supplied by SIEMENS. The optical cross connect is used to switch traffic to another ring network. The reconfigurable OADMs are composed of either tunable FBG technology using drop and park position [3,4] or a combination of an OFI and tunable FBGs [5]. Following the wavelength-selective drop process there is an optical switching matrix to enable for optical routing to different tributary equipment and/or optical interconnections. Thus, independent wavelength channels can be dropped an routed by remote software control steering all-optical components.To investigate the system performance, the signal wavelengths were all added at 50 GHz spaced wavelength comb after transmitting one span is shown in Fig. 2. In order to create the datastream of 800 Gbit/s, 80 laser wavelengths in the C-band (1530.66 to 1562.16 nm) are combined and modulated at 10 Gbit/s (PRBS: 231-1) by a LiNbO3 Mach-Zehnder modulator. Gain flattened boosterEDFAs are used to amplify the signals to a maximum channel power level of 0 dBm.2.
Results and Discussion
The optical parameters of a piezo-tunable FBG module optimized for the 100 GHz ITU grid are shown i Fig. 3. Tuning the inner FBG from 192.6 to 192.55 THz switches this channel from drop to park position,i.e., to a through state. Inter- and intrachannel crosstalk values are well below 25/35dB indicating the good quality of the fiber gratings. The interchannel crosstalk of 20 to 25 dB additionally required for a 80 channel system is supplied by a conventional dielectric demultiplexer. Tuning the center frequency of the gratings does not change the shape of reflection and transmission spectra. The group delay (not shown) is also unchanged in park position. In addition to their good optical properties, the piezo-tunable FBGs are
passively temperature compensated. A center wavelength stability of about 2 pm/øC can be achieved [4] thus enabling OADMs without active temperature stabilization. The piezo switching time from add to drop position is below 1 ms thus enabling optical protection purposes.
Fig. 3. Reflection and transmission of a FBG module in drop and park position.Fig. 4. BER versus received optical power of channels dropped at OADM I and OADM II. To set up a reconfigurable OADM for the 50 GHz ITU grid employing 100 GHz fiber grating technology an OFI is used [5]: The channels with 50 GHz spacing are converted into four streams of channels separated by 200 GHz. To demonstrate the system performance, the single channel power penalty is determined by a 10 Gbit/s and/or 20 Gbit/s bit error rate (BER) measurement (not shown). The
measurements at 10 Gbit/s of both piezo-tunable and thermooptically tunable FBGs reveal a power penalty in through position of < 0.5 dB. Penalties below 0.5 dB are also measured for the OFI. These results rule out excessive dispersion penalties and in combination with simultations demonstrate the cascadability of up to 7 OADMs in a metropolitan fiber ring system. Increasing the line rate up to 20 Gbit/s increases the power penalties to 2-3 dB indicating an upper limit for a system based on FBGs and 50/100 GHz technology. Typical results for the 50 GHz grid OADM network realized experimentally are shown in Fig. 4. An accumulated capacity of 0.8 Tbit/s is fed into the ring system. Error-free transmission is obtained for all wavelength channels that are re-routed by remote control. Dropping channel 7/11 out of 80 channels at OADM I leads to a penalty of about 1 dB. Electronically switching channel 7 or channel 11 from drop at OADM I to drop at OADM II increases the penalty to about 2 dB. Increasing the single channel power by 3 dB (channel 7/channel 11) decreases the penalty slightly at OADM I and significantly at OADM II.
These results and the single channel measurements confirm that the penalty increase from OADM I to OADM I is solely due to accumulation of EDFA noise and can be avoided using optimized EDFAs. To allow for optical routing following the wavelength selection process, the OADMs supplied in the field-trial are composed of a four FBG unit and a 4x4 optical switching matrix. System requirements for these key components are a low insertion loss (< 4 dB), high crosstalk values (> 40 dB), fast optical switching time (1 ms) for optical protection needs and an option for a latching state. The thermooptic 4x4 based on polymers fulfills all criteria besides the latching option. The microoptic solution is driven by
piezos with low power consumption, therefore enabling the latching state. The transmission and group delay properties of both alternatives do not affect a transmitted signal up to 10 Gbit/s. The stability of the traffic in the optically closed fiber ring loop is not affected by circulation of amplified spontaneous emission (ASE) from the EDFAs. The free spectral range of the demultiplexing component in the hub node sufficiently suppresses the ASE below 1530 nm and above 1562 nm. Thus out-of-band ring laser action is avoided. There is no in-band laser action, as long as there is one or more active channels present in the ring.
3. Conclusion
We demonstrated system operation of a 60 km metropolitan fiber-ring composed of reconfigurable OADMs capable of routing up to 80 optical channels using the 50 GHz ITU grid. Dynamic optical networking at an aggregate line capacity of 0.8 Tbit/s has been shown in this transparent WDM ring. Error free transmission was obtained for wavelength channels that were routed by remote control. The key components are an OFI, tunable FBGs and optical switching matrices. These components did not cause any significant signal distortions at 10 Gbit/s, i.e., FBG dispersion does not hamper cascadation of up to 7 OADMs. The distribution of dropped signals to a final destination can be performed by an optical switching matrix stage without affecting the signal quality. Limitations of the system operation are currently mainly due to the accumulation of EDFA noise. The optimum reconfigurable OADM for metropolitan purposes in the KOMNET field-trial includes an OFI stage to allow for 25/50 GHz channel spacing. To enable future reliable optical protection switching within 1 ms, piezo-tunable FBGs in combination with microoptic matrices are preferred. Using these OADMs, flexible drop capacities of several hundred Gbit/s are achievable, an order of magnitude larger than that of conventional SDH/SONET network elements.
4. References
[1] C. Scheerer, C. Glingener, A. F„rbert, J.-P. Elbers, A. Sch”pflin, E. Gottwald, and G. Fischer: ?3.2 Tbit/s (80 x 40 Gbit/s)
bidirectional WDM/ETDM transmission over 40 km standard single-mode fibre?, Electron. Lett., to be published, 1999.
[2] R.S. Vodhanel, F. Shehadeh, J.-C. Chiao, G.-K. Chang, C. Gibbons, and T. Suzaki: ?Performance of an 8-wavelength 8-node
WDM ring network experiment with 80 Gbit/s capacity?, OFC?96, postdeadline paper PD20, pp. 439 ? 442.
[3] P. Leisching, H. Bock, A. Richter, D. Stoll, and G. Fischer: ?Optical add/drop multiplexer for dynamic channel routing?,
Electron. Lett., vol. 35, pp. 591-592, 1999.
[4] A. Richter , T. Andritschke, H. Bock, P. Leisching, D. Stoll, L. Qu‚tel, and S. Aguy: ?Passive temperature compensation of
piezo-tunable fibre Bragg gratings?, Electron. Lett, vol. 35, pp. 1269-1270, 1999.
[5] H. Bock, A. Richter, P. Leisching, C. Glingener, D. Stoll, G. Fischer, P. Pace, J. Philipson, and Mark Farries: ?0.8 Tbit/s all-optical-
networking in a transparent WDM ring network?, submitted to ECOC?99
TuQ3-3
II Articles from LightWave:
Lightwave on March 12, 2000
lw.pennwellnet.com
--------------------------------------------------------------------------------
Article Date: March, 2000
Magazine Volume: 17
Issue: 3
--------------------------------------------------------------------------------
Emerging tunable-laser applications in optical networks
Robert Plastow Altitun AB
The Internet as we know it today exists because of the high-capacity transport infrastructure provided by dense wavelength-division multiplexing (DWDM) optical-fiber links. DWDM technology allows network operators to send multiple wavelength signals down a single fiber, enabling huge growth in network capacity at reasonable cost. With the expected exponential traffic growth, it is clear that optical-network capacity must continue to scale dramatically.
To enable this growth without repeatedly hitting bottlenecks, service providers must be able to assign, and re-assign, wavelengths as needed throughout the network. Laser transmitters that can tune wavelengths to any desired channel are a key element in these future network architectures. Network operators such as Sprint and Telenor are already working with tunable lasers to overcome the capacity barriers of existing systems.
Future network requirements
The need to flexibly provide more bandwidth is the driving force behind future optical-network requirements. Service providers will require scalability of factors of 10 beyond the network's initial installed capacity. Rapid service activation is another requisite to allow carriers to provide high-capacity links as needed-moving from the current provisioning times of months for OC-192 (10-Gbit/sec) circuits to true bandwidth on demand. Future optical networks must also handle multiple traffic types, offer the lowest cost of ownership, and provide seamless restoration (survivability).
All of these requirements will drive network functionality into the optical layer. The use of reconfigurable multiwavelength transport will allow service providers to offer bandwidth on demand, but also provide the survivability and quality of service needed.
Dynamic capacity allocation
Tunable-laser transmitters can provide a way to achieve network functionality in the optical layer in a reliable and flexible manner, using commercially available technology. In particular, the physical route taken by an optical signal can be altered at the transmitter by simply changing the wavelength and using either passive or tunable WDM filters at each node to determine the signal route.
This design can be implemented in many ways. Examples include star, ring, and mesh networks. In general, star and ring networks are more suitable for access and metropolitan networks. Figure 1 shows a simple, ring architecture using tunable lasers and either fixed or tunable filters. The node where a signal is dropped is determined by the wavelength that the transmission laser is set to.
Mesh architectures are often used in the backbone network to provide short connection and protection paths and make efficient use of both the available bandwidth and the fiber infrastructure. In this more general case, an optical crossconnect (OCX) can be used to route signals (see Figure 2). True scalable crossconnects require large space-switch matrices, and, if nonblocking, wavelength conversion, which is difficult to achieve without optical-to-electrical conversion.
A recent demonstration by the Norwegian operator Telenor highlights the attractions of a switchless mesh architecture based on tunable lasers (see Figure 3). A reconfigurable all-optical network was designed using passive optical-wavelength routing elements at the node. By altering the wavelength of the laser, a specific path is chosen across the network that will deliver the signal directly to its end destination without any processing, switching, or reconfiguration at the intermediate node. The tunable laser and the |
