01-02-2010, 12:06 AM
[attachment=1615]
Introducing DWDM
The following discussion provides some background on why dense wavelength division multiplexing (DWDM) is an important innovation in optical networks and what benefits it can provide. We begin with a high-level view of the segments of the global network and the economic forces driving the revolution in fiber optic networks. We then examine the differences between traditional time-division multiplexing (TDM) and wavelength division multiplexing (WDM). Finally, we explore the advantages of this new technology.
Global Network Hierarchy
It is the nature of modern communications networks to be in a state of ongoing evolution. Factors such as new applications, changing patterns of usage, and redistribution of content make the definition of networks a work in progress. Nevertheless, we can broadly define the larger entities that make up the global network based on variables such as transport technology, distance, applications, and so on.
One way of describing the metropolitan area network (MAN) would be to say that it is neither the long-haul nor the access parts of the network, but the area that lies between those two (see Figure 1-1).
Figure 1-1: Global Network Hierarchy
Long-Haul Networks
Long-haul networks are at the core of the global network. Dominated by a small group of large transnational and global carriers, long-haul networks connect the MANs. Their application is transport, so their primary concern is capacity. In many cases these networks, which have traditionally been based on Synchronous Optical Network (SONET) or Synchronous Digital Hierarchy (SDH) technology, are experiencing fiber exhaust result Of high bandwidth demand
Figure 1-1:
Access Networks
At the other end of the spectrum are the access networks. These networks are the closest to the end users, at the edge of the MAN. They are characterized by diverse protocols and infrastructures, and they span a broad spectrum of rates. Customers range from residential Internet users to large corporations and institutions. The predominance of IP traffic, with its inherently bursty, asymmetric, and unpredictable nature, presents many challenges, especially with new real-time applications. At the same time, these networks are required to continue to support legacy traffic and protocols, such as IBMâ„¢s Enterprise System Connection (ESCON).
Metropolitan Area Networks
Between these two large and different networking domains lie the MANs. These networks channel traffic within the metropolitan domain (among businesses, offices, and metropolitan areas) and between large long-haul points of presence (POPs). The MANs have many of the same characteristics as the access networks, such as diverse networking protocols and channel speeds. Like access networks, MANs have been traditionally SONET/SDH based, using point-to-point or ring topologies with add/drop multiplexers (ADMs).
The MAN lies at a critical juncture. On the one hand, it must meet the needs created by the dynamics of the ever-increasing bandwidth available in long-haul transport networks. On the other hand, it must address the growing connectivity requirements and access technologies that are resulting in demand for high-speed, customized data services.
Metropolitan and Long-Haul Networks Compared
There is a natural tendency to regard the MAN as simply a scaled-down version of the long-haul network. It is true that networks serving the metropolitan area encompass shorter distances than in the long-haul transport networks. Upon closer examination, however, these differences are superficial compared to other factors. Network shape is more stable in long-haul, while topologies change frequently in the MAN. Many more types of services and traffic types must be supported in MANs, from traditional voice and leased line services to new applications, including data storage, distributed applications, and video. The long-haul, by contrast, is about big pipes.
Another important way in which metropolitan networks today differ from trunk-oriented long haul networks is that they encompass a collection of low bit-rate asynchronous and synchronous transmission equipment, short loops, small cross-sections, and a variety of users with varying bandwidth demands. These fundamental differences between the two types of networks have powerful implications for the requirements in the metropolitan domain. Protocol and speed transparency, scalability, and dynamic provisioning are at least as important as capacity, which rules in the long-haul market.
An Alternative View
The preceding breakdown of the global network represents a somewhat simplified view. In reality, the lines between the domains are not always so clear-cut. Long-haul and metropolitan networks are sometimes not clearly delineated; the same holds true for the access and metropolitan domains.
Furthermore, other views of the global network exist. One, for example, defines the access network as part of, rather than separate from, the MAN, while also including enterprise connectivity in the metropolitan domain. In this view, the metropolitan market breaks down as follows:
¢ Coreâ€These are essentially scaled-down long-haul systems. They are considered the core of the MAN, because they interconnect carrier POPs and do not directly interface with end users.
¢ Metropolitan accessâ€this is the segment between carrier POPs and access facilities, which could be equipment at customer premises or at an aggregation point.
¢ Enterpriseâ€This is the part of the network dedicated to serving the needs of enterprises. Using owned or leased fiber (or leased fiber capacity), connectivity is provided between geographically disparate enterprise sites and for new applications, such as storage area networks (SANs).
Economic Forces
As we enter the twenty-first century, it goes without saying that information services have permeated society and commerce. Information, while still a tool, has become a commodity in itself. Yet the universal acceptance and ubiquitous adoption of information technology systems has strained the backbones on which they were built. High demandâ€coupled with high usage rates, a deregulated telecommunications environment, and high availability requirementsâ€is rapidly depleting the capacities of fibers that, when installed 10 years ago, were expected to suffice for the foreseeable future.
Bandwidth Demand
The explosion in demand for network bandwidth is largely due to the growth in data traffic, specifically Internet Protocol (IP). Leading service providers report bandwidths doubling on their backbones about every six to nine months. This is largely in response to the 300 percent growth per year in Internet traffic, while traditional voice traffic grows at a compound annual rate of only about 13 percent (see Figure 1-2).
Figure 1-2: Data Traffic Overtakes Voice Traffic
At the same time that network traffic volume is increasing, the nature of the traffic itself is becoming more complex. Traffic carried on a backbone can originate as circuit based (TDM voice and fax), packet based (IP), or cell based (ATM and Frame Relay). In addition, there is an increasing proportion of delay sensitive data, such as voice over IP and streaming video.
In response to this explosive growth in bandwidth demand, along with the emergence of IP as the common foundation for all services, long-haul service providers are moving away from TDM based systems, which were optimized for voice but now prove to be costly and inefficient. Meanwhile, metropolitan networks are also experiencing the impact of growing congestion, as well as rapidly changing requirements that call for simpler and faster provisioning than is possible with older equipment and technologies. Of key importance in the metropolitan area is the growth in storage area networks (SANs), discussed in the Storage Area Networks section.
Competition and Reliability
While the demand for bandwidth is driven largely by new data applications, Internet usage, and the growth in wireless communications, two additional factors come into play: competition and network availability.
The telecommunication sector, long a beneficiary of government regulation, is now a highly competitive industry. Competition was first introduced into the U.S. long-distance market in 1984, and the 1996 Telecommunications Reform Act is now resulting in an increasingly broad array of new operators. These new carriers are striving to meet the new demand for additional services and capacity.
There are two main effects on the industry from competition:
¢ Enhanced services are created by newcomers trying to compete with incumbents. In the metropolitan market, for example, there are broadband wireless and DSL services to homes and small and medium-sized business, high-speed private line and VPN services to corporations, and transparent LAN services to enterprise network customers.
¢ New carriers coming onto the scene create new infrastructure so that they do not have to lease from existing operators. Using this strategy, they have more control over provisioning and reliability.
As telecommunications and data services have become more critical to business operations, service providers have been required to ensure that their networks are fault tolerant. To meet these requirements, providers have had to build backup routes, often using simple 1:1 redundancy in ring or point-to-point configurations. Achieving the required level of reliability, however, means reserving dedicated capacity for failover. This can double the need for bandwidth on an already strained infrastructure (see Figure 1-3).
Figure 1-3: Reserving Bandwidth Reduces Overall Capacity
Options for Increasing Carrier Bandwidth
Faced with the challenge of dramatically increasing capacity while constraining costs, carriers have two options: Install new fiber or increase the effective bandwidth of existing fiber.
Laying new fiber is the traditional means used by carriers to expand their networks. Deploying new fiber, however, is a costly proposition. It is estimated at about $70,000 per mile, most of which is the cost of permits and construction rather than the fiber itself. Laying new fiber may make sense only when it is desirable to expand the embedded base.
Increasing the effective capacity of existing fiber can be accomplished in two ways:
¢ Increase the bit rate of existing systems.
¢ Increase the number of wavelengths on a fiber.
Increase the Bit Rate
Using TDM, data is now routinely transmitted at 2.5 Gbps (OC-48) and, increasingly, at 10 Gbps (OC-192); recent advances have resulted in speeds of 40 Gbps (OC-768). The electronic circuitry that makes this possible, however, is complex and costly, both to purchase and to maintain. In addition, there are significant technical issues that may restrict the applicability of this approach. Transmission at OC-192 over single-mode (SM) fiber, for example, is 16 times more affected by chromatic dispersion than the next lower aggregate speed, OC-48. The greater transmission power required by the higher bit rates also introduces nonlinear effects that can affect waveform quality. Finally, polarization mode dispersion, another effect that limits the distance a light pulse can travel without degradation, is also an issue..
Increase the Number of Wavelengths
In this approach, many wavelengths are combined onto a single fiber. Using wavelength division multiplexing (WDM) technology several wavelengths, or light colors, can simultaneously multiplex signals of 2.5 to 40 Gbps each over a strand of fiber. Without having to lay new fiber, the effective capacity of existing fiber plant can routinely be increased by a factor of 16 or 32. Systems with 128 and 160 wavelengths are in operation today, with higher density on the horizon. The specific limits of this technology are not yet known.
Time-Division Multiplexing
Time-division multiplexing (TDM) was invented as a way of maximizing the amount of voice traffic that could be carried over a medium. In the telephone network before multiplexing was invented, each telephone call required its own physical link. This proved to be an expensive and unscalable solution. Using multiplexing, more than one telephone call could be put on a single link.
TDM can be explained by an analogy to highway traffic. To transport all the traffic from four tributaries to another city, you can send all the traffic on one lane, providing the feeding tributaries are fairly serviced and the traffic is synchronized. So, if each of the four feeds puts a car onto the trunk highway every four seconds, then the trunk highway would get a car at the rate of one each second. As long as the speed of all the cars is synchronized, there would be no collision. At the destination the cars can be taken off the highway and fed to the local tributaries by the same synchronous mechanism, in reverse.
This is the principle used in synchronous TDM when sending bits over a link. TDM increases the capacity of the transmission page link by slicing time into smaller intervals so that the bits from multiple input sources can be carried on the link, effectively increasing the number of bits transmitted per second (see Figure 1-4).
Figure 1-4: TDM Concept
With TDM, input sources are serviced in round-robin fashion. Though fair, this method results in inefficiency, because each time slot is reserved even when there is no data to send. This problem is mitigated by the statistical multiplexing used in Asynchronous Transfer Mode (ATM). Although ATM offers better bandwidth utilization, there are practical limits to the speed that can be achieved due to the electronics required for segmentation and reassembly (SAR) of ATM cells that carry packet data.
SONET and TDM
The telecommunications industry adopted the Synchronous Optical Network (SONET) or Synchronous Digital Hierarchy (SDH) standard for optical transport of TDM data. SONET, used in North America, and SDH, used elsewhere, are two closely related standards that specify interface parameters, rates, framing formats, multiplexing methods, and management for synchronous TDM over fiber.
SONET/SDH takes n bit streams, multiplexes them, and optically modulates the signal, sending it out using a light emitting device over fiber with a bit rate equal to (incoming bit rate) x n. Thus traffic arriving at the SONET multiplexer from four places at 2.5 Gbps will go out as a single stream at 4 x 2.5 Gbps, or 10 Gbps. This principle is illustrated in Figure 1-5, which shows an increase in the bit rate by a factor of four in time slot T.
Figure 1-5: SONET TDM
The original unit used in multiplexing telephone calls is 64 kbps, which represents one phone call. Twenty-four (in North America) or thirty-two (outside North America) of these units are multiplexed using TDM into a higher bit-rate signal with an aggregate speed of 1.544 Mbps or 2.048 Mbps for transmission over T1 or E1 lines, respectively. The hierarchy for multiplexing telephone calls is shown in Table 1-1.
Table 1-1: Telco Multiplexing Hierarchy Signal Bit Rate Voice Slots
DS0 64 kbps 1 DS0
DS1 1.544 Mbps 24 DS0s
DS2 6.312 Mbps 96 DS0s
DS3 44.736 Mbps 28 DS1s
These are the basic building blocks used by SONET/SDH to multiplex into a standard hierarchy of speeds, from STS-1 at 51.85 Mbps to STS-192/STM-64 at 10 Gbps. Table 1-2 shows the relationship between the Telco signal rates and the most commonly used levels of the SONET/SDH hierarchy (OC-768 is not yet common).
Table 1-2: SONET/SDH Multiplexing Hierarchy Optical Carrier SONET/SDH Signal Bit Rate Capacity
OC-1 STS-1 51.84 Mbps 28 DS1s or 1 DS3
OC-3 STS-3/STM-1 155.52 Mbps 84 DS1s or 3 DS3s
OC-12 STS-12/STM-4 622.08 Mbps 336 DS1s or 12 DS3s
OC-48 STS-48/STM-16 2488.32 Mbps 1344 DS1s or 48 DS3s
OC-192 STS-192/STM-64 9953.28 Mbps 5379 DS1s or 192 DS3s
Figure 1-6 depicts this multiplexing and aggregation hierarchy. Using a standard called virtual tributaries for mapping lower-speed channels into the STS-1 payload, the 28 DS1 signals can be mapped into the STS-1 payload, or they can be multiplexed to DS3 with an M13 multiplexer and fit directly into the STS-1. Note also that ATM and Layer 3 traffic, using packet over SONET (POS), can feed into the SONET terminal from switches equipped with SONET interfaces.
Figure 1-6: TDM and SONET Aggregation
SONET/SDH does have some drawbacks. As with any TDM, the notions of priority or congestion do not exist in SONET or SDH. Also, the multiplexing hierarchy is a rigid one. When more capacity is needed, a leap to the next multiple must be made, likely resulting in an outlay for more capacity than is initially needed. For example, the next incremental step from 10 Gbps (STS-192) TDM is 40 Gbps (STS-768). Also, since the hierarchy is optimized for voice traffic, there are inherent inefficiencies when carrying data traffic with SONET frames. Some of these inefficiencies are shown in Table 1-3. DWDM, by contrast, can transport any protocol, including SONET, without special encapsulation.
Table 1-3: Ethernet in SONET Inefficiencies Ethernet SONET/SDH Signal Bit Rate Wasted Bandwidth
10BASE-T (10 Mbps) STS-1 51.8540 Mbps 80.709%
100BASE-T (100 Mbps) STS-3/STM-1 155.520 Mbps 35.699%
1000BASE-T (1000 Mbps) STS-48/STM-16 2488.32 Mbps 59.812%
Table 1-3.
To summarize, the demand placed on the transport infrastructure by bandwidth-hungry applications and the explosive growth of the Internet has exceeded the limits of traditional TDM. Fiber, which once promised seemingly unlimited bandwidth, is being exhausted, and the expense, complexity, and scalability limitations of the SONET infrastructure are becoming increasingly problematic.
Wavelength Division Multiplexing
WDM increases the carrying capacity of the physical medium (fiber) using a completely different method from TDM. WDM assigns incoming optical signals to specific frequencies of light (wavelengths, or lambdas) within a certain frequency band. This multiplexing closely resembles the way radio stations broadcast on different wavelengths without interfering with each other (see Figure 1-7). Because each channel is transmitted at a different frequency, we can select from them using a tuner. Another way to think about WDM is that each channel is a different color of light; several channels then make up a rainbow.
Figure 1-7: Increasing Capacity with WDM
________________________________________
Note The term wavelength is used instead of the term frequency to avoid confusion with other uses of frequency. Wavelength is often used interchangeably with lambda and channel.
________________________________________
In a WDM system, each of the wavelengths is launched into the fiber, and the signals are demultiplexed at the receiving end. Like TDM, the resulting capacity is an aggregate of the input signals, but WDM carries each input signal independently of the others. This means that each channel has its own dedicated bandwidth; all signals arrive at the same time, rather than being broken up and carried in time slots.
The difference between WDM and dense wavelength division multiplexing (DWDM) is fundamentally one of only degree. DWDM spaces the wavelengths more closely than does WDM, and therefore has a greater overall capacity.The limits of this spacing are not precisely known, and have probably not been reached, though systems are available in mid-year 2000 with a capacity of 128 lambdas on one fiber. DWDM has a number of other notable features, which are discussed in greater detail in the following chapters. These include the ability to amplify all the wavelengths at once without first converting them to electrical signals, and the ability to carry signals of different speeds and types simultaneously and transparently over the fiber (protocol and bit rate independence).
________________________________________
Note WDM and DWDM use single-mode fiber to carry multiple lightwaves of differing frequencies. This should not be confused with transmission over multimode fiber, in which light is launched into the fiber at different angles, resulting in different modes of light. A single wavelength is used in multimode transmission.
________________________________________
TDM and WDM Compared
SONET TDM takes synchronous and asynchronous signals and multiplexes them to a single higher bit rate for transmission at a single wavelength over fiber. Source signals may have to be converted from electrical to optical, or from optical to electrical and back to optical before being multiplexed. WDM takes multiple optical signals, maps them to individual wavelengths, and multiplexes the wavelengths over a single fiber. Another fundamental difference between the two technologies is that WDM can carry multiple protocols without a common signal format, while SONET cannot. Some of the key differences between TDM and WDM are graphically illustrated in Figure 1-8.
Figure 1-8: TDM and WDM Interfaces
Additional Drivers in Metropolitan Area Networks
Bandwidth, the chief driver in the long-haul market, is also a big driver in metropolitan area, access, and large enterprise networks (see Figure 1-9). In these types of networks additional applications driving demand for bandwidth include storage area networks (SANs), which make possible the serverless office, consolidation of data centers, and real-time transaction processing backup.
Figure 1-9: High-Speed Enterprise WAN Bandwidth Migration
There is also rapidly increasing demand on access networks, which function primarily to connect end users over low-speed connections, such as dial-up lines, DSL, cable, and wireless, to a local POP. These connections are typically aggregated and carried over a SONET ring, which at some point attaches to a local POP that serves as an Internet gateway for long hauls. Now, the growing demand for high-speed services is prompting service providers to transform the POP into a dynamic service-delivery center. As a result, it is increasingly likely that a customer now obtains many high-speed services directly from the POP, without ever using the core segment of the Internet.
Value of DWDM in the Metropolitan Area
DWDM is the clear winner in the backbone. It was first deployed on long-haul routes in a time of fiber scarcity. Then the equipment savings made it the solution of choice for new long-haul routes, even when ample fiber was available. While DWDM can relieve fiber exhaust in the metropolitan area, its value in this market extends beyond this single advantage. Alternatives for capacity enhancement exist, such as pulling new cable and SONET overlays, but DWDM can do more. What delivers additional value in the metropolitan market is DWDMâ„¢s fast and flexible provisioning of protocol- and bit rate-transparent, data-centric, protected services, along with the ability to offer new and higher-speed services at less cost.
The need to provision services of varying types in a rapid and efficient manner in response to the changing demands of customers is a distinguishing characteristic of the metropolitan networks. With SONET, which is the foundation of the vast majority of existing MANs, service provisioning is a lengthy and complex process. Network planning and analysis, ADM provisioning, Digital Crossconnect System (DCS) reconfiguration, path and circuit verification, and service creation can take several weeks. By contrast, with DWDM equipment in place provisioning new service can be as simple as turning on another lightwave in an existing fiber pair.
Potential providers of DWDM-based services in metropolitan areas, where abundant fiber plant already exists or is being built, include incumbent local exchange carriers (ILECs), competitive local exchange carriers (CLECs), inter-exchange carriers (IXCs), Internet service providers (ISPs), cable companies, private network operators, and utility companies. Such carriers can often offer new services for less cost than older ones. Much of the cost savings is due to reducing unnecessary layers of equipment, which also lowers operational costs and simplifies the network architecture.Carriers, can create revenue today by providing protocol-transparent, high-speed LAN and SAN services to large organizations, as well as a mixture of lower-speed services (Token Ring, FDDI, Ethernet) to smaller organizations. In implementing an optical network, they are ensuring that they can play in the competitive field of the future.
Requirements in the Metropolitan Area
The requirements in the metropolitan market may differ in some respects from those in the long-haul network market; yet metropolitan networks are still just a geographically distinguished segment of the global network. What happens in the core must be supported right to the edge. IP, for example, is the dominant traffic type, so interworking with this layer is a requirement, while not ignoring other traffic (TDM). Network management is now of primary concern, and protection schemes that ensure high availability are a given.
Key requirements for DWDM systems in the MAN include the following:
¢ Multiprotocol support
¢ Scalability
¢ Reliability and availability
¢ Openness (interfaces, network management, standard fiber types, electromagnetic compatibility)
¢ Ease of installation and management
¢ Size and power consumption
¢ Cost effectiveness
Why DWDM
From both technical and economic perspectives, the ability to provide potentially unlimited transmission capacity is the most obvious advantage of DWDM technology. The current investment in fiber plant can not only be preserved, but optimized by a factor of at least 32. As demands change, more capacity can be added, either by simple equipment upgrades or by increasing the number of lambdas on the fiber, without expensive upgrades. Capacity can be obtained for the cost of the equipment, and existing fiber plant investment is retained.
Bandwidth aside, DWDMâ„¢s most compelling technical advantages can be summarized as follows:
¢ Transparencyâ€because DWDM is physical layer architecture, it can transparently support both TDM and data formats such as ATM, Gigabit Ethernet, ESCON, and Fibre Channel with open interfaces over a common physical layer.
¢ Scalabilityâ€DWDM can leverage the abundance of dark fiber in many metropolitan area and enterprise networks to quickly meet demand for capacity on point-to-point links and on spans of existing SONET/SDH rings.
¢ Dynamic provisioningâ€Fast, simple, and dynamic provisioning of network connections give providers the ability to provide high-bandwidth services in days rather than months.
In the following sections we discuss some additional advantages, including migration from SONET and reliability.
SONET with DWDM
By using DWDM as a transport for TDM, existing SONET equipment investments can be preserved. Often new implementations can eliminate layers of equipment. For example, SONET multiplexing equipment can be avoided altogether by interfacing directly to DWDM equipment from ATM and packet switches, where OC-48 interfaces are common (see Figure 1-10). Additionally, upgrades do not have to conform to specific bit rate interfaces, as with SONET, where aggregation of tributaries is locked into specific values.
Figure 1-10: Direct SONET Interfaces from Switch to DWDM
Optical signals become attenuated as they travel through fiber and must be periodically regenerated in core networks. In SONET/SDH optical networks prior to the introduction of DWDM, each separate fiber carrying a single optical signal, typically at 2.5 Gbps, required a separate electrical regenerator every 60 to 100 km (37 to 62 mi). As additional fibers were turned up in a core network, the total cost of regenerators could become very large, because not only the cost of the regenerators themselves, but also the facilities to house and power them, had to be considered. The need to add regenerators also increased the time required to light new fibers
The upper part of Figure 1-11 shows the infrastructure required to transmit at 10 Gbps (4 x OC-48 SR interfaces) across a span of 360 km (223 mi) using SONET equipment; the lower part of the figure shows the infrastructure required for the same capacity using DWDM. While optical amplifiers could be used in the SONET case to extend the distance of spans before having to boost signal power, there would still need to be an amplifier for each fiber. Because with DWDM all four signals can be transported on a single fiber pair (versus four), fewer pieces of equipment are required. Eliminating the expense of regenerators (RPTR) required for each fiber results in considerable savings.
Figure 1-11: DWDM Eliminates Regenerators
A single optical amplifier can reamplify all the channels on a DWDM fiber without demultiplexing and processing them individually, with a cost approaching that of a single regenerator. The optical amplifier merely amplifies the signals; it does not reshape, retime or retransmit them as a regenerator does, so the signals may still need to be regenerated periodically. But depending on system design, signals can now be transmitted anywhere from 600 to thousands of kilometers without regeneration.
In addition to dramatically reducing the cost of regenerators, DWDM systems greatly simplify the expansion of network capacity. The only requirement is to install additional or higher bit-rate interfaces in the DWDM systems at either end of the fiber. In some cases it will only be necessary to increase the number of lambdas on the fiber by deploying existing interfaces, as shown in the upper half of Figure 1-12. The existing optical amplifiers amplify the new channel without additional regenerators. In the case of adding higher bit-rate interfaces, as shown in the lower half of Figure 1-12, fiber type can become a consideration. See the Optical Fibers section for an overview of types of optical fibers and their uses.
Figure 1-12: Upgrading with DWDM
Although amplifiers are of great benefit in long-haul transport, they are often unnecessary in metropolitan networks. Where distances between network elements are relatively short, signal strength and integrity can be adequate without amplification. But with MANs expanding in deeper into long-haul reaches, amplifiers will become useful.
Enhancing Performance and Reliability
Todayâ„¢s metropolitan and enterprise networks support many mission-critical applications that require high availability, such as billing and accounting on mainframes or client-server installations in data centers. Continuous backups or reliable decentralized data processing and storage are essential. These applications, along with disaster recovery and parallel processing, have high requirements for performance and reliability. As enterprises out source data services and inter-LAN connectivity, the burden of service falls on the service provider rather than on the enterprise.
With DWDM, the transport network is theoretically unconstrained by the speed of available electronics. There is no need for optical-electrical-optical (OEO) conversion when using optical amplifiers, rather than regenerators, on the physical link. Although not yet prevalent, direct optical interfaces to DWDM equipment can also eliminate the need for an OEO function.
While optical amplifiers are a major factor in the ability to extend the effective range of DWDM, other factors also comes into play. For example, DWDM is subject to dispersion and nonlinear effects. These effects are further discussed in the Optical Fibers section.
Many components, such as the optical add/drop multiplexer (OADM), are passive and therefore continue to work, even if there is a power cut. In addition, these components tend to have a very high mean time between failures (MTBF). Protection schemes implemented on DWDM equipment and in the network designs are at least as robust as those built into SONET. All these factors contribute to better performance and lower maintenance in the optical network.
Network Management Capability
One of the primary advantages offered by SONET technology is the capability of the data communication channel (DCC). Used for operations functions, DCCs ship such things as alarms, administration data, signal control information, and maintenance messages. When SONET is transported over DWDM, DCCs continue to perform these functions between SONET network elements. In addition, a DWDM system can have its own management channel for the optical layer. For out-of-band management, an additional wavelength (for example, a 33rd wavelength in a 32-wavelength system) is used as the optical supervisory channel (OSC). For inband management, a small amount of bandwidth (for example, 8 kHz) is reserved for management on a per-channel basis.
Additional Benefits
The shift in the makeup of traffic from voice to data has important implications for the design and operation of carrier networks. The introduction of cell-switching technologies such as ATM and Frame Relay demonstrates the limitations of the narrow-band, circuit-switched network design, but the limits of these technologies are being reached. Data is no longer an add-on to the voice-centric network, but is central. There are fundamentally different requirements of a data-centric network; two of these are the aggregation model and the open versus proprietary interfaces.
Aggregation in a voice-centric network consists of multiplexing numerous times onto transmission facilities and at many points in the network. Aggregation in a data-centric network, by contrast, tends to happen at the edge. With OC-48 (and higher) interfaces readily available on cell and packet switches, it becomes possible to eliminate costly SONET multiplexing and digital cross-connect equipment. OC-48 connections can interface directly to DWDM equipment.
Finally, service providers and enterprises can respond more quickly to changing demands by allocating bandwidth on demand. The ability to provision services rapidly by providing wavelength on demand creates new revenue opportunities such as wavelength leasing (an alternative to leasing of physical links or bit rate-limited tunnels), disaster recovery, and optical VPNs
Wavelength-division multiplexing
A bundle of optical fibre. Theoretically, using advanced techniques such as WDM, the modest number of fibers seen here could have sufficient bandwidth to easily carry the sum of all types of current data transmission needs for the entire planet. (~100 terabits per second per fiber [1])
In fibre optic telecommunications, wavelength-division multiplexing (WDM) is a technology which multiplexes multiple optical carrier signals on a single optical fibre by using different wavelengths (colours) of laser light to carry different signals. This allows for a multiplication in capacity, in addition to making it possible to perform bidirectional communications over one strand of fibre.
The term wavelength-division multiplexing is commonly applied to an optical carrier (which is typically described by its wavelength), whereas frequency-division multiplexing typically applies to a radio carrier (which is more often described by frequency). However, since wavelength and frequency are inversely proportional, and since radio and light are both forms of electromagnetic radiation, the two terms are closely analogous.
WDM systems
A WDM system uses a multiplexer at the transmitter to join the signals together, and a demultiplexer at the receiver to split them apart. With the right type of fibre you can have a device that does both at once, and can function as an optical add-drop multiplexer. The optical filtering devices used in the modems are usually etalons, stable solid-state single-frequency Fabry-Perot interferometers.
The concept was first published in 1970, and by 1978 WDM systems were being realized in the laboratory. The first WDM systems only combined two signals. Modern systems can handle up to 160 signals and can thus expand a basic 10 Gbit/s fiber system to a theoretical total capacity of over 1.6 Tbit/s over a single fibre pair.
WDM systems are popular with telecommunications companies because they allow them to expand the capacity of the network without laying more fibre. By using WDM and optical amplifiers, they can accommodate several generations of technology development in their optical infrastructure without having to overhaul the backbone network. Capacity of a given page link can be expanded by simply upgrading the multiplexers and demultiplexers at each end.
This is often done by using optical-to-electrical-to-optical translation at the very edge of the transport network, thus permitting interoperation with existing equipment with optical interfaces.
Most WDM systems operate on single mode fibre optical cables, which have a core diameter of 9 µm. Certain forms of WDM can also be used in multi-mode fibre cables (also known as premises cables) which have core diameters of 50 or 62.5 µm.
Early WDM systems were expensive and complicated to run. However, recent standardization and better understanding of the dynamics of WDM systems have made WDM much cheaper to deploy.
Optical receivers, in contrast to laser sources, tend to be wideband devices. Therefore the demultiplexer must provide the wavelength selectivity of the receiver in the WDM system.
Coarse WDM
WDM systems are divided into two market segments, dense and coarse WDM. Systems with more than 8 active wavelengths per fibre are generally considered Dense WDM (DWDM) systems, while those with fewer than eight active wavelengths are classed as coarse WDM (CWDM).
CWDM and DWDM technology are based on the same concept of using multiple wavelengths of light on a single fibre, but the two technologies differ in the spacing of the wavelengths, number of channels, and the ability to amplify signals in the optical space.
Recently the ITU has standardized a 20 nanometre channel spacing grid for use with CWDM, using the wavelengths between 1310 nm and 1610 nm. Many CWDM wavelengths below 1470 nm are considered "unusable" on older G.652 spec fibres, due to the increased attenuation in the 1310-1470 nm bands. Newer fibres which conform to the G.652.C and G.652.D standards, such as Corning SMF-28e and Samsung Wide pass nearly eliminate the "water peak" attenuation peak and allow for full operation of all twenty ITU CWDM channels in metropolitan networks. For more information on G.652.C and .D compliant fibres please see the links at the bottom of the article:
The Ethernet LX-4 physical layer standard is an example of a CWDM system in which four wavelengths near 1310 nm, each carrying a 3.125 gigabit-per-second data stream, are used to carry 10 gigabits per second of aggregate data.
CWDM is also being used in cable television networks, where different wavelengths are used for the downstream and upstream signals. In these systems, the wavelengths used are often widely separated, for example the downstream signal might be at 1310 nm while the upstream signal is at 1550 nm.
An interesting and relatively recent development relating Coarse WDM is the creation of Small Form Factor Pluggable (SFP) transceivers utilizing standardized CWDM wavelengths. SFP Optics allow for something very close to a seamless upgrade in even legacy systems that support SFP interfaces. Thus, a legacy Ethernet switch can be easily "converted" into a multiwavelength switch simply by judicious choice of transceiver wavelengths, combined with an inexpensive passive optical multiplexing device. This is in contrast to Dense WDM systems which, though optically amplifiable and far more efficient (in terms of bandwidth) are orders of magnitude more expensive. Thus it would seem that CWDM would be poised to find a solid market share in metropolitan systems as well as high-end enterprise.
Dense WDM
The introduction of the ITU-T G.694.1 frequency grid in 2002 has made it easier to integrate WDM with older but more standard SONET systems. WDM wavelengths are positioned in a grid having exactly 100 GHz (about 0.8nm) spacing in optical frequency, with a reference frequency fixed at 193.10 THz (1552.52nm). The main grid is placed inside the optical fiber amplifier bandwidth, but can be extended to wider bandwidths. Today's DWDM systems use 50 GHz or even 25 GHz channel spacing for up to 160 channel operation.
DWDM systems are significantly more expensive than CWDM because the laser transmitters need to be significantly more stable than those needed for CWDM. Precision temperature control of laser transmitter is required in DWDM systems to prevent "drift" off a very narrow centre wavelength. In addition, DWDM tends to be used at a higher level in the communications hierarchy, for example on the Internet backbone and is therefore associated with higher modulation rates, thus creating a smaller market for DWDM devices with very high performance levels, and corresponding high prices. In other words, they are needed only in small numbers and it is therefore not possible to amortize their development cost amongst a large number of transmitters.
Recent innovations in DWDM transport systems include pluggable and software-tunable transceiver modules capable of operating on 40 or 80 channels. This dramatically reduces the need for discrete spare pluggable modules, when a handful of pluggable devices can handle the full range of wavelengths.
Dense Wavelength-division Multiplexing
Dense wavelength-division multiplexing (DWDM) revolutionized data transmission technology by increasing the capacity signal of embedded fiber. This increase means that the incoming optical signals are assigned to specific wavelengths within a designated frequency band, then multiplexed onto one fiber. This process allows for multiple video, audio, and data channels to be transmitted over one fiber while maintaining system performance and enhancing transport systems. This technology responds to the growing need for efficient and capable data transmission by working with different formats, such as SONET/SDH, while increasing bandwidth.
The fiber optic amplifier component of the DWDM system provides a cost efficient method of taking in and amplifying optical signals without converting them into electrical signals. In addition, DWDM amplifies a broad range of wavelengths in the 1550 nm region. For example, with a DWDM system multiplexing 16 wavelengths on a single optical fiber, carriers can decrease the number of amplifiers by a factor of 16 at each regenerator site. Using fewer regenerators in long-distance networks results in fewer interruptions and enhanced efficiency.
DWDM System Considerations
Important components for a DWDM systems are transmitters, receivers, fiber amplifiers, DWDM multiplexers, and DWDM demultiplexer. These components, along with conforming to ITU channel standards, allow a DWDM system to interface with other equipment and to implement optical solutions throughout the network.
Figure 1- DWDM System Application
Multiplexers and Demultiplexers
The recent explosion of DWDM technology forced the fiber optic manufacturers to develop DWDM multiplexers and demultiplexers that can handle closely spaced optical wavelengths. These designs require narrow passbands, usually 0.4 nm wide, steep roll-off to reject adjacent channels, and stable operation over increased temperature. Recently, multiplexers have gained versatility, moving beyond the wideband wavelengths and into densely packed wavelengths that can be integrated into a multiple high frequency, 192 to 200 THz, transmission system. This type of system can maintain up to 16 channels, acting as a 16 fiber channel cable with each frequency channel operating to serve a STM-16/OC-48 carrier.
Demultiplexers need to eliminate crosstalk and channel interference. Couplers and dichroic filter, both passive devices, are the most favorable demultiplexers today. The first DWDM coupler design is based on fiber Bragg grating (FBG) filters illustrated in Figure 2. Bragg gratings are comprised of a length of optical fiber with the index of the core permanently modified periodically usually when exposed to an ultraviolet interference pattern. As a result, the fiber grating behaves as a wavelength dependent reflector and lends itself to precise wavelength separation.
Figure 2 - Bragg Grating
The second design is based on cascaded dichroic filters much like those used in the WDM system shown below in Figure 3. In a DWDM coupler, a second dichroic filter would be placed where the fiber 2 is located, and additional dichronic filters would be cascaded until all wavelengths have been combined or separated. At moderate cost, the dichronic filter method assures stability and excellent isolation between channels.
Figure 3 - Dichronic Filter
Fiber Amplifiers for DWDM
Because DWDM systems handle information optically rather than electrically, it is imperative that long-haul applications do not suffer the effects of dispersion and attenuation. Erbium-doped fiber amplifiers (EDFAs) counteract these problems. EDFAs are silica based optical fibers that are doped with erbium. This rare earth element has the appropriate energy levels in its atomic structure for amplifying light at 1550 nm. A 980 nm pump laser is used to inject energy into the doped fiber. When a weak signal at 1310 nm or 1550 nm enters the fiber, the light stimulates the rare earth atoms to release their stored energy as additional 1310 nm or 1550 nm light. This process continues as the signal passes down the fiber, continually growing stronger. Figure 4 illustrates an erbium-doped fiber.
Figure 4 - Erbium-doped Optical Fiber
The photons amplify the incoming signal optically, boosting the wavelength, and avoiding almost all of the active components. The output power of the EDFA is large, and thus, fewer amplifiers may be needed in any given system design. The amplification process is independent of the data rate. Because of this benefit, upgrading a system means only changing the launch/receive terminals.
As demands for wider bandwidth grow there is a call for more efficient and reliable optical amplifiers. The usable bandwidth of an EDFA is only about 30 nm (1530 nm-1560 nm), but the minimum attenuation is in the range of 1500 nm to 1600 nm. The dual-band fiber amplifier (DBFA) solves the usable bandwidth problem. It is broken down into two sub-band amplifiers. The DBFA is similar to the EDFA, but its bandwidth ranges from about 1528 nm to 1610 nm. The first range is similar to that of the EDFA and the second is known as extended band fiber amplifier (EBFA). Some features of the EBFA include flat gain, slow saturation, and low noise. The EBFA can achieve a flat gain over a range of 35 nm which is comparable to the EDFAs. EBFAs have the advantage of reaching a slower saturation keeping the output constant even though the input increases.
Channel Spacing
DWDM channel spacing governs system performance; 50 GHz and 100 GHz outline the standards of ITU channel spacing. Currently, 100 GHz is the most commonly used and reliable channel spacing. This spacing allows for several channel schemes without imposing limitations on available fiber amplifiers. However, channel spacing depends on the systemâ„¢s components.
Channel spacing is the minimum frequency separation between two multiplexed signals. An inverse proportion of frequency versus wavelength of operation calls for different wavelengths to be introduced at each signal. The optical amplifiers bandwidth and receivers ability to identify two close wavelength, sets the channel spacing. Figure 5 illustrates the typical DWDM specifications.
Figure 5 - Typical Optical Characteristics for DWDM Channels
Signal Direction
DWDM involved sending a large number of closely spaced optical signals over a single fiber. Standards developed by the ITU (International Telecommunications Union) define the exact optical wavelength used for DWDM applications. The center of the DWDM band lies at 193.1 THz with standard channel spacing of 200 GHz and 100 GHz. The closest "standard" spacing (100 GHz) allows transmission of 45 channels on one fiber. A 45 channel system spaced at 100 GHz would cover a optical span of 35 nm and require a costly wide bandwidth, gain-flattened EDFA.
As system designers looked to pack more than the 45 channels at 100 GHz spacing, they started to use closer spaced optical channels. The channel spacing, in GHz, relates to the optical wavelength as follows: A spacing of 200 GHz corresponds to about 1.6 nm, 100 GHz corresponds to about 0.8 nm, and 50 GHz corresponds to about 0.4 nm channels spacing. Most commonly 50 GHz follows 100 GHz, although attempts at 75 GHz and 37.5 GHz show up in literature. While there is nothing magical about any of these numbers, it seems likely that 50 GHz will be the next logical step below 100 GHz. Using a channel spacing of 50 GHz (0.4 nm) allows 45 channels to occupy only 17.5 nm of optical bandwidth. This greatly simplifies the requirement for optical amplifiers in the system. Fiber increases in channels per fiber would likely lead to the use of 25 GHz spacing.
Designing the optical demultiplexer to separate the signals at the receive end defines the greatest challenge in closely spaced optical channels. Because of subtle color differences in each of the optical channels, high performance DWDM optical demultiplexers must have three characteristics. First, it must be very stable over time and temperature. Second, it needs to have a relatively flat passband or region of frequencies. Third, it must reject adjacent optical channels so that they do not interfere. Several basic types of designs can be used in optical demultiplexers to separate the optical channels. Many of these designs have an increasingly difficult time separating the optical channels as the spacing becomes very close. Some, however, such as fiber Bragg gratings actually appear better suited for closer channel spacing. The need for close optic channel spacing is a trade-off between the performance required of the optical amplifiers used in the system and the number of channels to be transmitted per fiber. Figure 6 illustrates the transmission spectra of 0.4 nm spacing DWDM FBGs.
Figure 6 - 0.4 nm Channel Spacing DWDM Fiber Bragg Grating
Red and Blue Bands
The ITU approved DWDM band extends from 1528.77 nm to 1563.86 nm, and divides into the red band and the blue band. The red band encompasses the longer wavelengths of 1546.12 nm and higher. The blue band wavelengths fall below 1546.12 nm. This division has a practical value because useful gain region of the lowest cast EDFAs corresponds to the red band wavelengths. Thus, if a system only requires a limited number of DWDM wavelengths using the red band wavelength yields the lowest overall system cost.
1. The Challenges of Today's Telecommunications Network
To understand the importance of DWDM and optical networking, these capabilities must be discussed in the context of the challenges faced by the telecommunications industry, and, in particular, service providers. Most U.S. networks were built using estimates that calculated bandwidth use by employing concentration ratios derived from classical engineering formulas such as Poisson and Reeling. Consequently, forecasts of the amount of bandwidth capacity needed for networks were calculated on the presumption that a given individual would only use network bandwidth six minutes of each hour. These formulas did not factor in the amount of traffic generated by Internet access (300 percent growth per year), faxes, multiple phone lines, modems, teleconferencing, and data and video transmission. Had these factors been included, a far different estimate would have emerged. In fact, today many people use the bandwidth equivalent of 180 minutes or more each hour.
Therefore, an enormous amount of bandwidth capacity is required to provide the services demanded by consumers. For perspective, in 1997, a long-distance carrier made major strides when it increased its bandwidth capacity to 1.2 Gbps (billions of bits per second) over one fiber pair. At the transmission speed of one Gbps, one thousand books can be transmitted per second. However today, if one million families decide they want to see video on Web sites and sample the new emerging video applications, then network transmission rates of terabits (trillions of bits per second [Tbps]) are required. With a transmission rate of one Tbps, it is possible to transmit 20 million simultaneous 2-way phone calls or transmit the text from 300 years“worth of daily newspapers per second.
No one could have predicted the network growth necessary to meet the demand. For example, one study estimated that from 1994 to 1998 the demand on the U.S. interexchange carriers'(IXCs) network would increase sevenfold, and for the U.S. local exchange carriers' (LECs) network, the demand would increase fourfold. In actuality, one company indicated that its network growth was 32 times that of the previous year, while another company's rate of growth in 1997 alone was the same size as its entire network in 1991. Yet another has said that the size of its network doubled every six months in that four-year period.
In addition to this explosion in consumer demand for bandwidth, many service providers are coping with fiber exhaust in their networks. An industry survey indicated that in 1995, the amount of embedded fiber already in use in the average network was between 70 percent and 80 percent. Today, many carriers are nearing one hundred“percent capacity utilization across significant portions of their networks. Another problem for carriers is the challenge of deploying and integrating diverse technologies in one physical infrastructure. Customer demands and competitive pressures mandate that carriers offer diverse services economically and deploy them over the embedded network. DWDM provides service providers an answer to that demand (see Figure 1).
Figure 1. Optical Transport to Optical Networking: Evolution of the Phototonics Layer
Use of DWDM allows providers to offer services such as e-mail, video, and multimedia carried as Internet protocol (IP) data over asynchronous transfer mode (ATM) and voice carried over SONET/SDH. Despite the fact that these formatsâ€IP, ATM, and SONET/SDHâ€provide unique bandwidth management capabilities, all three can be transported over the optical layer using DWDM. This unifying capability allows the service provider the flexibility to respond to customer demands over one network.
A platform that is able to unify and interface with these technologies and position the carrier with the ability to integrate current and next-generation technologies is critical for a carrier's success.
Capacity Expansion Potential
By beginning with DWDM, service providers can establish a grow-as-you-go infrastructure, which allows them to add current and next-generation TDM systems for virtually endless capacity expansion (see Figure 4). DWDM also gives service providers the flexibility to expand capacity in any portion of their networksâ€an advantage no other technology can offer. Carriers can address specific problem areas that are congested because of high capacity demands. This is especially helpful where multiple rings intersect between two nodes, resulting in fiber exhaust.
Figure 4. Capacity Expansion Evolution: A Strategy for the Long Term
Service providers searching for new and creative ways to generate revenue while fully meeting the varying needs of their customers can benefit from a DWDM infrastructure as well. By partitioning and maintaining different dedicated wavelengths for different customers, for example, service providers can lease individual wavelengthsâ€as opposed to an entire fiberâ€to their high-use business customers.
Compared with repeater-based applications, a DWDM infrastructure also increases the distances between network elementsâ€a huge benefit for long-distance service providers looking to reduce their initial network investments significantly. The fiber-optic amplifier component of the DWDM system enables a service provider to save costs by taking in and amplifying optical signals without converting them to electrical signals. Furthermore, DWDM allows service providers to do it on a broad range of wavelengths in the 1.55µm region. For example, with a DWDM system multiplexing up to 16 wavelengths on a single fiber, carriers can decrease the number of amplifiers by a factor of 16 at each regenerator site. Using fewer regenerators in long-distance networks results in fewer interruptions and improved efficiency.
Resolving the Capacity Crisis
1. Faced with the multifaceted challenges of increased service needs, fiber exhaust, and layered bandwidth management, service providers need options to provide an economical solution. One way to alleviate fiber exhaust is to lay more fiber, and, for those networks where the cost of laying new fiber is minimal, this will prove the most economical solution. However, laying new fiber will not necessarily enable the service provider to provide new services or utilize the bandwidth management capability of a unifying optical layer.
A second choice is to increase the bit rate using time division multiplexing (TDM), where TDM increases the capacity of a fiber by slicing time into smaller intervals so that more bits (data) can be transmitted per second (see Figure 2). Traditionally, this has been the industry method of choice (DS“1, DS“2, DS“3, etc.). However, when service providers use this approach exclusively, they must make the leap to the higher bit rate in one jump, having purchased more capacity than they initially need. Based on the SONET hierarchy, the next incremental step from 10 Gbps TDM is 40 Gbpsâ€a quantum leap that many believe will not be possible for TDM technology in the near future. This method has also been used with transport networks that are based on either the synchronous optical network (SONET) standard for North America or the synchronous digital network (SDH) standard for international networks.
Figure 2. Increased Network Capacityâ€TDM
The telecommunications industry adopted the SONET or SDH standard to provide a standard synchronous optical hierarchy with sufficient flexibility to accommodate current and future digital signals. SONET or SDH accomplishes this by defining standard rates and formats and optical interfaces. For example, multiple electrical and optical signals are brought into a SONET terminal where they are terminated and multiplexed electrically before becoming part of the payload of an STS“1, the building block frame structure of the SONET hierarchy. The STS“1 payloads are then multiplexed to be sent out on the single fiber at a single rate: OC“3 to OC“12 to OC“48 and eventually to OC“192. SDH has a similar structure with STM“n building block resulting in signal rates of STS“1 through STM“64.
SONET and SDH, two closely related standards, provided the foundation to transform the transport networks as we know them today. They govern interface parameters; rates, formats, and multiplexing methods; and operations, administration, maintenance, and provisioning (OAM&P) for high-speed transmission of bits of information in flashing laser-light streams. A synchronous mode of transmission means that the laser signals flowing through a fiber-optic system have been synchronized to an external clock. The resulting benefit is that data streams transmitting voice, data, and images through the fiber system flow in a steady, regulated manner so that each stream of light can readily be identified and easily extracted for delivery or routing.
Capacity Expansion and Flexibility: DWDM
The third choice for service providers is dense wavelength division multiplexing (DWDM), which increases the capacity of embedded fiber by first assigning incoming optical signals to specific frequencies (wavelength, lambda) within a designated frequency band and then multiplexing the resulting signals out onto one fiber. Because incoming signals are never terminated in the optical layer, the interface can be bit-rate and format independent, allowing the service provider to integrate DWDM technology easily with existing equipment in the network while gaining access to the untapped capacity in the embedded fiber.
DWDM combines multiple optical signals so that they can be amplified as a group and transported over a single fiber to increase capacity (see Figure 3). Each signal carried can be at a different rate (OC“3/12/24, etc.) and in a different format (SONET, ATM, data, etc.) For example, a DWDM network with a mix of SONET signals operating at OC“48 (2.5 Gbps) and OC“192 (10 Gbps) over a DWDM infrastructure can achieve capacities of over 40 Gbps. A system with DWDM can achieve all this gracefully while maintaining the same degree of system performance, reliability, and robustness as current transport systemsâ€or even surpassing it. Future DWDM terminals will carry up to 80 wavelengths of OC“48, a total of 200 Gbps, or up to 40 wavelengths of OC“192, a total of 400 Gbpsâ€which is enough capacity to transmit 90,000 volumes of an encyclopedia in one second.
Figure 3. Increased Network Capacityâ€WDM
The technology that allows this high-speed, high-volume transmission is in the optical amplifier. Optical amplifiers operate in a specific band of the frequency spectrum and are optimized for operation with existing fiber, making it possible to boost lightwave signals and thereby extend their reach without converting them back to electrical form. Demonstrations have been made of ultra wideband optical-fiber amplifiers that can boost lightwave signals carrying over 100 channels (or wavelengths) of light. A network using such an amplifier could easily handle a terabit of information. At that rate, it would be possible to transmit all the world's TV channels at once or about half a million movies at the same time.
Consider a highway analogy where one fiber can be thought of as a multilane highway. Traditional TDM systems use a single lane of this highway and increase capacity by moving faster on this single lane. In optical networking, utilizing DWDM is analogous to accessing the unused lanes on the highway (increasing the number of wavelengths on the embedded fiber base) to gain access to an incredible amount of untapped capacity in the fiber. An additional benefit of optical networking is that the highway is blind to the type of traffic that travels on it. Consequently, the vehicles on the highway can carry ATM packets, SONET, and IP.
DWDM Incremental Growth
A DWDM infrastructure is designed to provide a graceful network evolution for service providers who seek to address their customers' ever-increasing capacity demands. Because a DWDM infrastructure can deliver the necessary capacity expansion, laying a foundation based on this technology is viewed as the best place to start. By taking incremental growth steps with DWDM, it is possible for service providers to reduce their initial costs significantly while deploying the network infrastructure that will serve them in the long run.
Some industry analysts have hailed DWDM as a perfect fit for networks that are trying to meet demands for more bandwidth. However, these experts have noted the conditions for this fit: a DWDM system simply must be scalable. Despite the fact that a system of OC“48 interfacing with 8 or 16 channels per fiber might seem like overkill now, such measures are necessary for the system to be efficient even two years from now.
Because OC“48 terminal technology and the related operations support systems (OSSs) match up with DWDM systems today, it is possible for service providers to begin evolving the capacity of the TDM systems already connected to their network. Mature OC“192 systems can be added later to the established DWDM infrastructure to expand capacity to 40 Gbps and beyond.
The Optical Layer as the Unifying Layer
Aside from the enormous capacity gained through optical networking, the optical layer provides the only means for carriers to integrate the diverse technologies of their existing networks into one physical infrastructure. DWDM systems are bit-rate and format independent and can accept any combination of interface rates (e.g., synchronous, asynchronous, OC“3, “12, “48, or “192) on the same fiber at the same time. If a carrier operates both ATM and SONET networks, the ATM signal does not have to be multiplexed up to the SONET rate to be carried on the DWDM network. Because the optical layer carries signals without any additional multiplexing, carriers can quickly introduce ATM or IP without deploying an overlay network. An importan
