Wave Division Multiplexing WDM Computer Science Essay

Published: Last Edited:

This essay has been submitted by a student. This is not an example of the work written by our professional essay writers.

The sharp increases of internet-related applications and growth of all types of network traffic demand a high speed and bandwidth networks. Fibre optical networks provide a promising future. Theoretically, fibre can handle millions of signals in a fibre, but only small part of the potential was used commercially, in which several optical signals at different wavelengths were multiplexed together onto a single fibre cable. WDM system did not achieve much commercial progresses earlier because the time division multiplexing turned out to be less cost to increase the speed for the same capacity [1]. WDM enables network operators to efficiently exploit fibre's full potential by dividing fibre paths into four, eight, sixteen, or more distinct channels [2].

High-speed applications such as mass data transfer, videoconferencing, and real-time operations, and increasing internet traffic are the driving force for the bandwidth requirements. Adding more fibre is costly affair, and the new fibre needed more equipments and maintenance. The carrier also did not want to upgrade electronics in their architectures, because the existing facilities would be replaced. WDM is the best solution to the problem in the aspect of cost and technologies. To obtain 16 channels capacity through electronic upgrades 144 devices would add to a 310-mile fibre route, while only six new devices would be added by using WDM technology (Figure 1)[2].

[figure 1]

Before 1997, four and eight-channel WDM system were well served, and 16-channel WDM system just began to be deployed by largest IXC's, while the smaller networks and local exchange carriers had not yet used WDM [2]. The 32- and 400-channel were deployed in 1998 by Sprint and AT&T [2]. The comparison of 8-channel WDM and 32-channel WDM is shown in Figure 2. Higher channel WDM saves a lot of devices.

WDM technology has been in the market around for more than twenty years. Only after the erbium- doped fibre amplifier (EDFA) practically provided an efficient, low noise, and broadband gain in the 1500nm low-loss fibre band, the point-to-point WDM became realistic to increase the capacity of long-distance transmission and solve the bandwidth problem in MAN [3]. A number of technologies and filtering and fibre grating techniques have helped enhance the value of WDM [2]. WDM based optical network is playing a key role in this generation of the Internet.

2. Architectures and Technologies in WDM

2.1 Basic concept

WDM is the basic technology of optical networking. There are two kinds of WDM, simple (sparse) WDM and dense WDM. WDM is constructed by using 1310 nm as one wavelength and 1550 nm as the other or 850 and 1310 nm. Dense WDM refers to the closing spacing of channels. Its mean that a series of WDM channels spaced at 3.6 nm apart or the wavelength spacing is 1 nm per channel or less [4]. Figure 3 and 4 shows the difference between them. The simple and dense WDM's can be used together. Figure 5 shows the spectrum of a fibre being used for both at the same time. There is a single channel in the 1300 nm band and 4-channel WDM in the 1550 nm region [4].

[fig 3] 

Figure 3, A simple (Sparse) WDM [4].

[fig 4] 

Figure 4, Dense WDM [4].

[fig 5, click to enlarge] 

Figure 5, The spectrum of dense and sparse WDM on a same fibre [4].

Mostly 1300 and 1550 nm bands are used. The losses (attenuation) are different at the different bands (figure 6). At the range of 200 nm centred at 1300, the attenuation is less than 0.5 dB/km while at 1550 nm about 0.2 dB/km. The peak loss is in the 1400 region because of the hydroxyl ion impurities in the fiber. The number of amplifiers and repeaters can be reduced significantly due to the low loss of signal. The bandwidth can reach as much as 50 THz in the two regions, although the usable bandwidth is limited by the properties of fiber. In addition, the bit error rates (BERs) is less than 10-11 in fiber optic system [1]. Fiber transmission is not interfered by electromagnetic environment.

[fig 6, click to enlarge] 

Figure6. The low loss regions of an optical fibre [1].

The increasing demand for bandwidth, led to the introduction of wavelength division multiplexing (WDM) as one technique to increase capacity in the optical fibre networks. In past 100GHz was set as the standard channel separation by the ITU, but this spacing is due to capacity constraints, it has been already being reduced to 50GHz to accommodate extra data-traffic. Even 25GHz channel separations are currently being discussed and this become the trend of the future on the moderate bit - rate of 10Gbit/s, as an alternative to faster bit-rates whilst faster electronics is maturing. This trend therefore calls for filters that can perform filtering duties that on top of a solid functionality, also act as passive ultra-selective filters that can maintain channel integrity at any cost.

The technology of Fibre Bragg gratings has come a long way since the initial demonstrations in 1978 by Hill et al. at the CRC in Canada. Bragg gratings and in particular apodised Bragg gratings has previously been shown to exhibit near ideal characteristics for compact and high filling factor values on grid spacing's as small as 25GHz [5]. However, it has also been discussed how these filters, despite their near ideal spectral performance, suffer from non-linear phase attributes in the stop-band, that could limit their use in high bit-rate systems (10Gbit/s and above) [6]. Linear-phase filters therefore have been proposed as a solution to this problem, but some previous demonstrations have suffered from low rejection values [6].

Bragg grating square-filters for 25GHz and 50GHz channel separations that exhibit grid filling-factor values of 75% and constant reflectivity's in excess of 99.9% (>30dB transmission loss) over the full drop window. The linear phase and square spectral performance of the gratings are obtained by imposing a slowly varying envelope function (superstructure) (Fig.1) on the rapidly varying refractive index modulation forming the Bragg grating. This envelope function is generated using apodisation and is imposed during writing using a grating manufacturing technique where full control of all vital grating parameters is available. A recently developed inverse scattering algorithm is used for the design of the envelope function [7]. The manufactured filters are tested at 10Gbit/s NRZ and show that <10<sup>-11</sup> bit-error-rate (BER) is obtained for constant received power throughout the useful band (Fig.8a). Additionally, we will discuss the importance of these filters for dispersion-free filtering at high bit-rates, by comparing their performance with "traditionally" apodised Bragg grating filters. It will be demonstrated how this new family of gratings out-performs the standard apodised Bragg gratings for their linear-phase characteristics and that they allow for tuning and drift of the transmitter over the full bandwidth of the grating without being affected by dispersion at any point in the stop band.

C:\Users\user\AppData\Local\Microsoft\Windows\Temporary Internet Files\Content.Word\New Picture.png

C:\Users\user\AppData\Local\Microsoft\Windows\Temporary Internet Files\Content.Word\New Picture (2).png

C:\Users\user\AppData\Local\Microsoft\Windows\Temporary Internet Files\Content.Word\New Picture.png


Single mode

Single mode


Single mode







Energy Loss

Configuration Designed Fibre:

Data Rate: 100Mb/s Bit Error rate: 1x10-9 Distance between OLTE: 500km Operating wavelength: 1.55um

Specifications of Transmitter:

Laser: The laser used in this optical fibre is Distributed Feedback Laser (DFB).

Distributed Feedback: A Laser system in which Feedback is used to make certain Modes in the Resonator oscillate more strongly than others. In Semiconductor lasers, a periodic corrugation in the active Layer replaces the cleaved end mirrors, and the grating spacing is chosen to distribute the feedback in both directions, creating a condition that can approach single mode oscillation. Distributed feedback laser is abbreviated DFB.

Risetime = Falltime= 2ns

Typical optical output power =-3dBm

Laser spectral width=0.3 nm.

Specifications of Optical fibre:

Single-Mode Fibre: This is a glass fibre with single stand and with of diameter of 8.3 to 10 microns that has one mode of transmission. Single Mode Fibre has narrow diameter, through which only one mode will propagate typically 1310 or 1550nm. Carries higher bandwidth than multimode fibre, but requires a light source with a narrow spectral width. Single mode fibre emits higher transmission rate and up to 50 times more distance than multimode, but it also costs more. Single mode fibre has a much smaller core than compare to multimode fibre. The small core and single light wave virtually eliminate any distortion that could result from overlapping light pulses, providing the least signal attenuation and the highest transmission speeds of any fibre cable type. Single mode optical fibre is an optical fibre in which only the lowest order bound mode can propagate at the wavelength of interest typically 1300 to 1320nm.[8]

Attenuation: 0.4dB/km Dispersion=20ps/nm.km

Optical Receiver 1:

p-i-n Photodiode (PIN): PIN is a semiconductor without internal gain. In order to allow operation at longer wavelengths where the light penetrates more deeply into the semiconductor material a wider depletion region is necessary. To achieve this n type material is doped so lightly that it can be considered intrinsic, and to make a low resistance contact a highly doped n type (n+) layer is added. This creates a p-i-n (or PIN) structure, as may be seen in figure below where the entire absorption takes place in the depletion region.

Risetime= Falltime=1ns Dark current= 0A Responsivitty = 0.8 A/W

Optical Receiver 2:

Avalanche Photodiode (APD): APD is a semiconductor with internal gain. The internal gain mechanism in an APD is to increases the signal current to the amplifier and so it improves the Signal Noise Ratio (SNR) because of this load resistance and amplifier noise will not affected i.e. the thermal noise and amplifier noise figure will not be affected. However, the dark current and quantum noise are increased by multiplication process and this may become one the limiting factor. This is because the random gain mechanism introduces excess noise into the receiver in terms of increased shot noise above the level that would result from amplifying only the primary shot noise.

Risetime= Falltime=1ns Dark current = 0A Responsivitty= 1A/W Noise Factor= Gx and x=1

Maximum Length between repeaters arising from the loss limit:



Pi =Mean input (transmitted) optical power launched into a fibre.

Po=Mean output (received) optical power from a fibre.

αfc=Fibre cable loss in decibels per kilometre.

αj= Fibre joint loss in decibels per kilometre.

αcr = Connector loss at transmitter and receiver in decibels.

Np =Number of photons per bit (coherent transmission).

Ma=Safety margin in an optical power budget.


The maximum length with repeaters based on given data cannot more than 72.32 km.

Dispersion Limit of Designed fibre

Dispersion Limit: The pulse spreading in an optical fiber. As a pulse of light propagates through a fiber, elements such as numerical aperture, core diameter, refractive index profile, wavelength, and laser linewidth cause the pulse to broaden. This poses a limitation on the overall bandwidth of the fiber.

In given data, For one kilometre there is a dispersion of 20 ps/nm.km then for 500 kilometre is calculated as below shown:

Dispersion Limit= 1000ps

Maximum Length=0.0019km

Dispersion Limit is a one of the Limiting factor in optical fibre.

When the PIN receiver is considered the distance between the repeaters and numbers of repeaters are calculated below:







Single mode

Single mode

Single mode


PIN Receiver

Energy Loss of 2dBm

The PIN receiver is taken its Sensitivities is assumed as -27 dBm.

To meet the requirement repeaters are placed in the optical fibre.

The repeaters in the optical fibre just repeat the signal in the optical fibre.

As per given data, there is a energy is loss at the transmitter of -3dBm and also at of 2dBm is energy loss at every 1 km there is a loss of 0.4 dBm and at every 3 km there is a loss of 0.1dBM

Therefore, first repeater should be placed at distance of 50.78 km and second one should be placed at the of 55.78km because there won't be connector energy loss in between and remaining repeater are placed same distance.

Therefore, 8 repeaters are required to place in the optical fibre to work system in the proper way.

And Bit error rate at each section is 1x10-9 because

Bit Error Rate or Bit Error Ratio (BER): The number of received bits that have been altered due to noise, interface and distortion, divided by the total number of transferred bits during a studied time interval. BER is a unit less performance measure, often expressed as a percentage number.







Single mode

Single mode

Single mode


APD Receiver

Energy Loss