Metro 100G and Beyond Fiber Optic Networks

The explosion of digital video, cloud access and data center traffic is driving the growth of 100 Gb/s fiber optic networks, or 100G, in metro environments. Today, 20G or 40G is sufficient for most metro networks, but "going forward, 100G will be the bandwidth of choice," said Michael Belanger, a product manager at Ciena Corporation.

Randy Nicklas, executive vice president and engineering and chief technology officer for Windstream, said his company is seeing increased demand for bandwidth, especially to residential homes. “We deliver packets to our customers in both directions in increasing scale. That's why 100G is important to us,” Nicklas said.

Speaking about the economics of building 100G networks, Ryan Yu, vice president of business development and general manager of active products at Oplink Communications, pointed out that telecom companies have traditionally been the biggest spenders on infrastructure. But due to their increasing data transmission demands, Internet companies such as Facebook, Netflix and Baidu are catching up fast and are expected to outpace telecom companies in spending in the coming years.

As the industry shifts to higher data speeds, there are many challenges that will need to be overcome. For example, Jinghui Li, president of the optical thin film filter supplier Auxora, noted that below 20G, most client-side transceivers rely only on single wavelengths, but as speeds increased to 40G and now 100G, optics technologies such as WDM become necessary for long-distance data transmission.

The fiber pigtailed WDM modules used in the first generation of 100G transceivers were bulky and expensive. The filter block and PLC-AWG WDM used in the second generation 100G transceivers were smaller and less expensive and will still be used for years to come, Li said. Third-generation 100G transceivers that use silicon photonics WDM modules will soon be on the market, but details about their performance and costs are still unknown. Li predicted that depending on the application, filter block, PLC-AWG, and silicon photonics WDMs could co-exist.

Mike Capuano, vice president of marketing at Infinera, said that as data speeds increase, new network models will need to be created that are capable of tailoring the bandwidth to particular needs. For example, a 400G line could be reserved for a node in the network that only needs 80G. "As the industry moves to 100G to 200G to 400G in the metro, there's a lot of potential to waste a lot of bandwidth," Capuano said. "As an industry, we want to figure out how to get maximum capacity but also get the right amount of granularity to feed the system at the appropriate rate."


All These Terminologies ahout WDM Technology ?

As an unprecedented opportunity to dramatically increase the bandwidth capacity, WDM (Wavelength Division Multiplexing) technology is an ideal solution to get more bandwidth and lower cost in nowaday telecommunications networks. By virtue of fame, WDM becomes a household word now. Yet, most of the time, we only know what is "WDM" but do not really know WDM technology. Actually, there are various of terminologies used in WDM that are always a headache for us. Now, let's see what are they.

1.Basic Terminologies of WDM
2.WDM Transmission System
3.Network Topologies
4.Filtration Technologies Applied to WDM
5.WDM Equipment
6.Ports on WDM Equipment
7.Parameters of WDM Systems
8.New Network Technologies Based on WDM

WDM (Wavelength Division Multiplexing)
A technology that multiplexes a number of optical carrier signals onto a single optical fiber by using different optical wavelengths (i.e., colors) of laser light. It breaks white light passing through fiber optic cable into all the colors of the spectrum, much like light passed through a prism creates a rainbow. Every wavelength carries an individual signal that does not interfere with the other wavelengths.

1.CWDM (Coarse Wavelength Division Multiplexing)
CWDM is a specific WDM technology defined by the ITU (International Telecommunication Union) in ITU-T G.694.2 spectral grids, using the wavelengths from 1270 nm to 1610 nm within a 20nm channel spacing. It is a technology of choice for cost efficiently transporting large amounts of data traffic in telecoms or enterprise networks.

2.DWDM (Dense Wavelength Division Multiplexing)
DWDM is a specific WDM technology also defined by the ITU but in ITU-T G.694.1 spectral grids. The grid is specified as frequency in THz, anchored at 193.1 THz, with a variety of specified channel spacing from 12.5 GHz to 200 GHz, among which 100 GHz is common. In practice, DWDM frequency is usually converted to wavelength. DWDM typically has the capability to transport up to 80 channels (wavelengths) in what is known as the Conventional band (C-band) spectrum, with all 80 channels in the 1550 nm region.

When referring to fiber optic transmission in WDM system, you should know these:

1.Single Fiber Transmission
Single fiber, namely bi-directional communication on one single fiber. This system utilizes two identical sets of wavelengths for both directions over a single fiber. Individual channels residing on the single fiber system may propagate in either direction.

2.Dual Fiber Transmission
Dual fiber, namely comprised of two single fibers, one fiber is used for the transmit direction and the other is used for the receive direction. In dual fiber transmission system, the same wavelength is normally used in both the transmit and receive directions. The second fiber may serve as a backup fiber as in a redundant system, or it may provide an optical path in the opposite direction.

The topology in which WDM systems are used plays a key role in determining the extent to which the WDM network is utilized.
The related terminologies are:

1.Network Topologies
WDM products bring higher efficiency to fiber networks through multiple channel usage of fiber. Networks are identified by their fiber layout or topology. Network topologies such as Mesh, Ring, P2P (Point-to-Point), and P2MP (Point-to-Multipoint) will sometimes use WDM products particularly designed for the network. So, it is important to understand the intended network use when selecting WDM products. Entire networks are often comprised of several kinds of sub-network topologies.

2.Ring Topology
In metropolitan area networks, infrastructures are generally organized over a ring topology. Ring topology is a type of network topology consisting of a closed loop. Fiber ring networks are comprised of a series of fiber spans that terminate at network nodes spread throughout the loop. Each node in the ring will connect to two, and only two, adjacent nodes. Ring networks are often dual fiber systems. Contrast ring topology with an unclosed, end-to-end or point-to-point fiber span.

There are three competing filtration technologies applied to WDM. They are:

1.Arrayed Waveguide Grating (AWG)
AWG, including Athermal AWG (AAWG) and Thermal AWG (TAWG), is commonly used as optical MUX/DeMUX in WDM systems. AAWG have equivalent performance to standard TAWG but require no electrical power, software or temperature.
2.Fiber Bragg Grating (FBG)
FBGs are versatile wavelength filters for multiplexing and demultiplexing WDM signals. They also can compensate for chromatic dispersion that can degrade the quality of the WDM signal in an optical fiber.
3.Thin Film Filter (TFF)
Thin film filters were adopted very early on and have been widely deployed since because they have the unique attributes that meet the stringent requirements of optical communication systems. The main advantage of thin film filters is its ability to achieve high accuracy in processing in small device sizes when compared it to competing technologies.

To build a WDM system, these WDM equipment are required:

1.Mux (Multiplexer)
WDM multiplexer is a device that multiplexes or combines optical signals of different wavelengths (colors) together on one single fiber.

2.DeMux (De-Multiplexer)
In contrast to multiplexer, DeMux is a device that de-multiplexes or splits optical transmission comprised of multiplexed wavelengths onto individual fibers assigned to each wavelength.
Note: In today's market, there are CWDM Mux/DeMux products and DWDM Mux/DeMux products. These products have the Mux and DeMux inside and comes in a package like 1RU 19" rackmont, LGX box and ABS module etc.

3.OADM (Optical Add-Drop Multiplexer)
OADM is a device used in WDM systems for multiplexing and routing different channels of light into or out of a single fiber.

4.FWDM (Filter-Based Wavelength Division Multiplexer)
Filter-based Wavelength Division Multiplexer (FWDM) is a kind of WDM multiplexer based on the Thin Film Filter (TFF) technology. FWDM combines or separates light at different wavelengths in a wide wavelength range and is extensively used in EDFA, Raman amplifiers, and WDM optical networks.

5.Compact WDM
As the name suggests, these are multi-channel WDM products that have relatively small footprints so that they can provide more channels with a device footprint small enough to fit within a FOSC (Fiber Optic Splice Closure), splice tray or splice holder. These products utilize a free-space multiple bounce technology in which light reflects from each filter element directly onto the next filter element instead of being collimated and launched into a fiber as in individual, discrete TFF components. In addition, bend insensitive fiber permits the use of smaller housings for joining (concatenating) individual TFFs into a multi-channel product.

6.Banded Skip Filters
Banded skip filters are used to build BWDM (Band WDM) products. These filters are TFFs that have wide pass bands, which contain multiple channels. For example, DWDM Red/Blue C-band Filter is used to separate or combine Red and Blue band wavelength signals in C-band DWDM systems and high-power amplification systems. It is just like a regular FWDM, with the only difference that the wavelengths are split in Red/Blue filter while bonded in WDM.

Ports on WDM equipment, do you really know their functions?

1.Common Port
The connection point of a WDM product where combined channels appear. For a MUX product, combined channels are transmitted from the common port. For a DEMUX, the combined channels are received at the common port.

2.Express or Upgrade Port
For CWDM products, there will normally be either an upgrade or an express port, but not both. The upgrade or express port on a CWDM Mux or DeMux is used to add, drop, or pass through additional channels which enables the cascading of two CWDM Mux/DeMux modules, doubling the channel capacity on the common fiber link.
For DWDM products, the purpose of an upgrade port is to be able to add, drop, or pass through C-band DWDM channels not already in use, namely only channels that reside in the band 1530 – 1565 nm. If the DWDM product also has an express port, then that port is normally used for additional channels residing outside the C-band, such as most of the CWDM channels.

3.1310nm Port
The 1310nm port is a wide band optic port added to other specific CWDM wavelengths in a module. For example if an 8 channel CWDM is called out it may use wavelengths 1470 nm to 1610 nm and request the 1310nm port. The 1310nm port is used in some legacy networks and sometimes as a return path. If an existing legacy network is using 1310nm port and they have exhausted all fibers and are looking for ways to increase their network capacity they can add in other CWDM wavelengths on to the same fiber while still allowing the use of the 1310nm port. Meanwhile, it can carry LR optics, LX optics etc.

4.1550nm Port
Similar to 1310nm port, allows a legacy 1550nm signal to pass and can carry ER optics, ZR optics, LX optics, ZX optics etc.

5.Monitor Port
This port is used to monitor or test the power signal coming out of a Muxed CWDM or before it gets demuxed from the signal coming through the fiber network usually at a 5% or less power level. Generally, it can be connected with measurement or monitoring equipment, such as power meters or network analyzers. Network administrators will use this to test of monitor if a signal has failed or changed without having to interrupt the existing network.
You should also know these parameters when operating a WDM system:

Wavelength is the distance, measured in the direction of propagation, between two points of the same phase in consecutive cycles of a wave. The wavelength λm of monochromatic light travelling in a optical fiber is expressed:
λm = λ / n = v / f
λ = optical wavelength in a vacuum
n = the refractive index of the dielectric medium
v = phase velocity, given by c / n
c = the speed of light in a vacuum: 2.99792458 X 108 m/s
f = the optical frequency.

Note: In WDM practice, wavelengths such as the wavelength of a communications laser, the wavelength specifications for optical filters, and the wavelengths of optical transmission channels over fiber are all given as λ, the wavelength in nanometers as would occur in a vacuum.

On the basic of WDM, some new network technologies have emerged, such as:

1.Wavelength Division Multiplexing-Passive Optical Network (WDM-PON)
WDM-PON is an innovative concept for access and backhaul networks. It uses WDM over a physical P2MP fiber infrastructure that contains no active components (i.e., PON). WDM-PON allows operators to deliver high bandwidth to multiple endpoints over long distances.

2.Optical Transport Network (OTN)
OTN was designed to provide support for optical networking using wavelength-division multiplexing (WDM) unlike its predecessor SONET/SDH. It is able to provide functionality of transport, multiplexing, switching, management, supervision and survivability of optical channels carrying client signals.

The knowledge is endless, thus, the word you see on this page as only the tip of an iceberg. But once you know all of them, you may better know about WDM systems.


What Is XFP Transceiver?

XFP Transceivers, short for 10 Gigabit small form factor pluggable is a standard of transceivers for high-speed computer network and telecommunication links that use optical fiber. The XFP standard was defined by an industry group in 2002, along with its interface to other electrical components which is called XFI. The 10-Gigabit XFP transceiver module is a hot-swappable I/O device that plugs into 10-Gigabit ports. The XFP transceiver module connects the electrical circuitry of the system with the optical network.

The XFP packaging was smaller than the XENPAK form-factor and X2 which had been published earlier but are slightly larger than the original small form-factor pluggable transceiver (SFP). As the following picture shown, XFP Structure includes the basic six parts:

① Transmit optical bore
② Receive optical bore
③ Bail clasp
④ Dust plug
⑤ Transceiver socket connector

XFI Electrical Interface Specification
The XFI electrical interface specification was a 10 gigabit per second chip-to-chip electrical interface specification defined as part of the XFP multi-source agreement. It was also developed by the XFP MSA group. XFI provides a single lane running at 10.3125 Gbit/s when using a 64B/66B encoding scheme. A serializer/deserializer is often used to convert from a wider interface such as XAUI that has four lanes running at 3.125 Gbit/s using 8B/10B encoding. XFI is sometimes pronounced as "X" "F" "I" and other times as "ziffie".

Types of XFP Transceiver Modules
XFP are available with a variety of transmitter and receiver types, allowing users to select the appropriate transceiver for each link to provide the required optical reach over the available optical fiber type (e.g. multi-mode fiber or single-mode fiber). XFP modules are commonly available in several different categories:
1.SR - 850 nm, for a maximum of 300 m
2.LR - 1310 nm, for distances up to 10 km
3.ER - 1550 nm, for distances up to 40 km
4.ZR - 1550 nm, for distances up to 80 km


What is Fiber Optics and What is Optical Fiber?

Brief Introduction

It is a technology that uses glass (or plastic) threads (fibers) to transmit data. A fiber optic cable consists of a bundle of glass threads, each of which is capable of transmitting messages modulated onto light waves. It is a hair-like glass conduit that carries virtually any type of signal from one point to another at light speed. Unlike copper based signals, fiber signals are not affected by external power sources or surges and there is no need for shielding or grounding.

Fiber optics has several advantages over traditional metal communications lines:

 ^ transmit more info with greater fidelity over longer distances
 ^ added security, more resistant to electrical interference than traditional cooper cabling
 ^ Fiber optic cables are much thinner and lighter than metal wires.
 ^ Data can be transmitted digitally (the natural form for computer data) rather than analogically.

The main disadvantage of fiber optics is that the cables are expensive to install. In addition, they are more fragile than wire and are difficult to splice. While fiber has the disadvantage of requiring more expensive equipment and training, the extra expenses are well worth the monumental gains in speed and efficiency.

Fiber is the world's most difficult cable to terminate. Optical fibers must be finely cut, cleaned, polished, and inspected to ensure performance. Infiberone feature a huge selection of fiber termination kits, cleavers, testing and cleaning supplies.


As optical fiber is more effective than cable, a lot of Internet, telephone systems and cable TV occasions use fibers. Fiber optic cable allows network builders to divide their network into smaller service areas that prevent large numbers of customers from being affected in an outage. The result is better service and customer relations. Fiber optic cable also gives them a fast return path which they use for internet and telephone connections, thereby increasing their revenue potential.

Type-B Polarity MTP Cable User Guide

In today's data centers, 12-fiber pre-terminated MTP cables are frequently used to establish an optical path between switch tiers. Accomplishing this path in a way that matches the transmit signal (Tx) on one switch port to the corresponding receive signal (Rx) on the other switch port is referred to as polarity. Unlike traditional 2‐fiber configurations LC or SC patch cords, with one send and one receive, since 12-fiber MTP cables have 12 fibers in a connector, the polarity management then becomes more complex. According to TIA standards, there are type-A, type-B and type-C three polarity types MTP cables. Different types polarity cables may have different applications. In this tutorial, we will mainly introduce the type-B polarity MTP cables and its applications.

Type-B Polarity MTP Cable Overview

Often referred to as a 40G MTP cable, the type-B cable is the second standard for polarity, and extremely versatile. This cable assembly can be used to plug directly between your 40G QSFP+ optic transceivers, so it is commonly referred to as a QSFP/QSFP+ cable or Direct Attach 40GBASE-SR4. As you can see in the diagram, this cable has a “flipped” polarity and will result in a Pin 1 to Pin 12 relationship. This is extremely useful because 40G optics utilize parallel optics, meaning instead of alternating Tx and Rx in a duplex pattern, the port will look like the following diagram.

Type-B Polarity MTP Cable

Most 40G optics do not require perfect port symmetry: any Tx can go to any Rx and it will sort it out in the end. This means fiber 12 can interface with fiber 1, because one is Tx and the other is Rx.

Type-B Polarity MTP Cable Types

MTP trunk cable and MTP-LC harness cable are two most common types used in today's data center. Besides, according to MTP genders, MTP connector is divided into male and female two types. Therefore, in terms of MTP trunk cable, there are female-female, female-male and male-male three types. For MTP-LC harness cable, female MTP-LC harness cable and male MTP-LC harness cable are available.

Type-B Polarity MTP Cable Applications

Since MTP cables have male and female two types, it always requires 1 male and 1 female to mate. The male guide pins fit into the female holes to ensure precise fiber alignment. Attempting to mate two female connectors will result in a seemingly secure connection, but with extremely high loss, and attempting to mate two male connectors will most likely damage one or both connectors due to the guide pins clashing.

High-Speed Optical Transceiver PCBs Design

In high-speed printed circuit boards design, Engineers usually face two choices: microstrip line and stripline. In most case,for high-speed PCB design under 10Gb/s,microstrip waveguide become the dominant waveguide structure largely because it could simplify designand cost lower.Some inherent advantages make the stripline becoming more and more important.It seems that PCB design engineers needn't  worry too much about the difference between microstrip and stripline since previous design experience makes people believe that it is not a problem.In fact,when faced  higher speed pcb design, we need to make a choice.So, microstrip or stripline?

Figure1 shows the different of structure between microstrip and stripline. A microstrip consists of a conductive strip (copper) and a wider ground plane(copper), separated  by a dielectric layer (Er1).Between two wider ground plane(copper), separated  by two dielectric layer (Er1 and Er2),there is a conductive strip (copper) in stripline. Internal conductor in stripline is commonly called the “hot conductor1” while the other two, always connected at signal  ground, are called “cold” or “ground” conductors. If the dielectric layer Er2 is in stead of free space , stripline will become microstrip .

Figure2 Electric E and Magnetic H field lines for fundamental Quasi-TEM in Microstrip 1and Stripline2,3

High-speed signal in the conductor follows the basic principle of electromagnetic described by the maxwell's equations. In most cases,areas of signal transmission and transmission medium are passive, so the current density and charge density are zero. Electric field distribution is determined and constrained in free space and homogeneous medium. Determined electric field distribution decides the magnetic field distribution.EQ.1 is a simplified form of maxwell's equations in the passive free space, and it expresses the fact that electric and magnetic fields are perpendicular to the direction of propagation.

ay(∂Ex/ ∂z=-μ.∂Hy/ ∂t)

ax(ε. ∂Ex/ ∂t=-∂Hy/ ∂z)                                (EQ.13)

Electromagnetic described by EQ.1 is called TEM Wave. It has only two field components(E and H) aligned with the transverse coordinates: Ez =Hz = 0(EQ.21).

Discontinuity of Normal component and continuous of tangential component of the electric field in different medium boundary led to the distortion. Asymmetric structure of msicrostrip must result in electric field distortion in the boundary of free space and the dielectric layer. Usually, Quasi-TEM mode is used to describe the transmission parameters of microstrip while stripline is true TEM Wave. Of course, using Quasi-TEM mode to describe stripline is a good approximation and the high accuracy completely meet the engineering requirements.

More ahout High-Speed Optical Transceiver PCBs Design: