Ain Shams University

Faculty of Engineering
Electronics and Electrical Communications Eng. Dept.

ECE431 Optoelectronic lab

4th Year Electrical Communications and Electronics
Lab of Lasers and Optical Communications

Experiment 3
Fiber Attenuation Measurement
 1. Objectives:
● To be familiar with optical interference.
● To understand the use of optical spectrometers in material characterization.
● To Familiarize with the fiber connectors and their use in the optical systems.
●
2. Equipment:
● Fiber coupled NIR FTIR spectrometer
● Multi mode optical fiber (2 m)
● Long multi mode optical fiber (1.6 km)

3. Optical Fiber attenuation

Attenuation of light as it propagates along an optical fiber is an important consideration in the
design of optical communication system, as it plays a major role in determining the maximum
transmission distance between a transmitter and a receiver. The basic attenuation mechanisms in
a fiber are material absorption, scattering and radiative losses.

3.1 Attenuation units:

As light travels along a fiber, its power decreases exponentially with distance. If P(0) is the
optical power in a fiber at the origin (z=0), then the power P(z) at a distance z:
𝑃(𝑧) = 𝑃(0)𝑒 −𝛼𝑝 𝑧 (1)
where 𝛼𝑝 is the fiber attenuation coefficient in Np/km. The common procedure is to express the
attenuation coefficient in units of dB/km. Designating this parameter by 𝛼:
10 𝑃(0)
𝛼(𝑑𝐵/𝑘𝑚) = log [ ] = 4.343 𝛼𝑝 (2)
𝑧 𝑃(𝑧)

3.2 Scattering losses:

Scattering losses in silica fiber arise from compositional fluctuations and structural
inhomogeneities that may occur on the micro scale during fiber fabrication. These effects give rise
to refractive index variations that cause a Rayleigh-type scattering of the light. Rayleigh scattering
is the phenomenon that scatters light from the sun in the atmosphere, thereby giving rise to a blue
sky. The expression for scattering induced attenuation is inversely proportional to 𝜆4, hence is
dominant at small wavelengths.

3.3 Absorption:
Absorption in glass fiber is caused by atomic defects in glass composition, impurity atoms or from
the basic constituent atoms of the fiber material. Typically, the dominant absorption factor in silica
fibers is the presence of impurities which include OH- (water) ions that are dissolved in the glass
during fabrication steps. This type of attenuation is dominant at high wavelength. Typical
attenuation spectrum of standard optical fiber is shown in Figure 1. By reducing the residual OH
content of fibers to below 1 ppb, standard commercially available single mode fibers have nominal
attenuations of 0.4 dB/km at 1310 nm and less than 0.25 dB/km at 1550 nm. Further elimination
of water ions diminishes the absorption around 1440 nm as known as low-water-peak fibers or dry
fiber.

Figure 1: Fiber attenuation spectrum as a function of wavelength.

4. Scanning interferometers
4.1 Theory:
The design of many interferometers used for infrared spectrometry today is based on that of the
two-beam interferometer originally designed by Michelson in 1891. Many other two-beam
interferometers have subsequently been designed that may be more useful than the Michelson
interferometer for certain specific applications. Nevertheless, the theory behind all scanning two-
beam interferometers is similar.
The Michelson interferometer is a device that can divide a beam of radiation into two paths and
then recombine the two beams after a path difference has been introduced. A condition is thereby
created under which interference between the beams can occur. The variation of intensity of the
beam emerging from the interferometer is measured as a function of path difference by a detector.
The simplest form of the Michelson interferometer is shown in Figure 2. It consists of two mutually
perpendicular plane mirrors, one of which can move along an axis that is perpendicular to its
plane.
 Figure 2: Fourier transform infrared spectroscopy theory based on Michelson interferometer.
Bisecting the fixed mirror and the movable mirror is a beamsplitter, where a collimated beam of
radiation from an external source can be partially reflected to the fixed mirror and partially
transmitted to the movable mirror. When the beams return to the beamsplitter, they interfere and
are again partially reflected and partially transmitted. Because of the effect of interference, the
intensity of each beam passing to the detector and returning to the source depends on the difference
in path of the beams in the two arms of the interferometer. The variation in the intensity of the
beams passing to the detector and returning to the source as a function of the path difference
ultimately yields the spectral information in a Fourier transform spectrometer.
To understand the interferometer’s theory of operation, let us consider light source with single
wavelength of radiation (𝜆𝑜 ), hence a wavenumber (𝜐̃𝑜 = 1/𝜆𝑜 ). The field at the detector, as shown
in Figure 2 (a), for an ideal interferometer with 50% splitting ratio is:
𝐸𝑖𝑛 −𝑗2𝜋𝜐̃ 𝐿 𝐸𝑖𝑛 −𝑗2𝜋𝜐̃ 𝐿
𝐸𝑜𝑝 = 𝑒 𝑜 1 + 𝑒 𝑜 2 (3)
2 2
𝐸𝑖𝑛 −𝑗2𝜋𝜐̃ 𝐿 ̃𝑜 (𝐿1 −𝐿2 )
𝐸𝑜𝑝 = 𝑒 𝑜 1 (1 + 𝑒 𝑗2𝜋𝜐 ) (4)
2
𝐸𝑖𝑛 −𝑗2𝜋𝜐̃ 𝐿 𝑗2𝜋𝜐̃𝑜 (𝐿1−𝐿2) −𝑗2𝜋𝜐̃𝑜 (𝐿1−𝐿2) (𝐿1 −𝐿2 )
𝐸𝑜𝑝 = 𝑒 𝑜 1𝑒 2 (𝑒 2 + 𝑒 𝑗2𝜋𝜐̃𝑜 2 ) (5)
2
𝐸𝑖𝑛 −𝑗2𝜋𝜐̃ 𝐿 ̃𝑜
Δ𝐿
̃𝑜
Δ𝐿
𝐸𝑜𝑝 = 𝑒 𝑜 𝑎𝑣 (𝑒 −𝑗2𝜋𝜐 2 +𝑒
𝑗2𝜋𝜐
2 ) (6)
2
Δ𝐿
𝐸𝑜𝑝 = 𝐸𝑖𝑛 𝑒 −𝑗2𝜋𝜐̃𝑜 𝐿𝑎𝑣 cos (2𝜋𝜐̃𝑜 ) (7)
2

The detector output is proportional to the intensity:

 2 Δ𝐿 (8)
𝑆 ′ (Δ𝐿) = 𝐸𝑜𝑝 𝐸𝑜𝑝
∗
= 𝐸𝑖𝑛 cos 2 (2𝜋𝜐̃𝑜 )
2
𝐼(𝜐̃𝑜 )
𝑆 ′ (Δ𝐿) = (1 + cos(2𝜋𝜐̃𝑜 Δ𝐿)) (9)
2
It can be seen that 𝑆 ′ (Δ𝐿) is composed of a constant (dc) component equal to 0.5𝐼(𝜐̃𝑜 ), and a
modulated (ac) component equal to 0.5𝐼(𝜐̃𝑜 ) cos(2𝜋𝜐̃𝑜 Δ𝐿). Only the ac component is important
in spectrometric measurements, and it is this modulated component that is generally referred to as
the interferogram, 𝑺(𝚫𝑳). The interferogram from a monochromatic source measured with an
ideal interferometer is given by the equation:
𝐼(𝜐̃𝑜 ) (10)
𝑆(Δ𝐿) = cos(2𝜋𝜐̃𝑜 Δ𝐿)
2
In practice, the amplitude of the interferogram as observed after detection is proportional not only
to the intensity of the source but also to the beamsplitter efficiency and detector response. All
instrumental characteristics can be accounted and represented in a single parameter 𝐵(𝜐̃𝑜 ) (Figure
3 (a)):
𝑆(Δ𝐿) = 𝐵(𝜐̃𝑜 ) cos(2𝜋𝜐̃𝑜 Δ𝐿) (11)
Mathematically, 𝑆(Δ𝐿) is said to be the cosine Fourier transform of 𝐵(𝜐̃𝑜 ). Hence, the spectrum is
calculated from the interferogram by computing the cosine Fourier transform of 𝑆(Δ𝐿), which
accounts for the name given to this spectrometric technique: Fourier transform spectrometry.
The Fourier transform of equation (10) is an impulse function located at the frequency
(wavelength) of the monochromatic source as shown in Figure 3 (b).
Now, assume a broadband source with a wide range of input frequencies (wavelengths). The
interferogram is given as:
+∞
𝑆(Δ𝐿) = ∫ 𝐵(𝜐̃)cos(2𝜋𝜐̃Δ𝐿)𝑑𝜐̃ (12)
−∞
The interferogram for broadband source has an envelope that limits the light interference to a
limited range of optical path difference Δ𝐿 as shown in Figure 3 (c).
The other half of cosine pair is:
+∞
𝐵(𝜐̃) = ∫ 𝑆(Δ𝐿)cos(2𝜋𝜐̃Δ𝐿) 𝑑Δ𝐿 (13)
−∞

Because 𝑆(Δ𝐿) is an even function, the spectrum could be written as:

+∞
𝐵(𝜐̃) = 2 ∫ 𝑆(Δ𝐿)cos(2𝜋𝜐̃Δ𝐿) 𝑑Δ𝐿 (14)
0

The power spectral density of the source, shown in Figure 3 (d), is proportional to |𝐵(𝜐̃)|2 .
 (a) (b)

(c) (d)
Figure 3: Signal interferogram vs optical path difference (OPD) (𝛥𝐿 ) and corresponding spectrum
of (a, b) monochromatic light source, (c, d) wide band white light source.

4.2 Spectrometer specifications:

• Spectral resolution: Minimum possible wavelength separation that can be resolved. The
resolution of the spectrometer is limited due to the limited distance travelled by the mirror.
This can be modeled by multiplying the interferogram by a rectangular pulse (convolution
with sinc function in spectral domain). Practically, the spectral resolution is ∆𝜐 ∝ 1/𝑂𝑃𝐷𝑚𝑎𝑥 .
For low resolution, the spectrum details cannot be distinguished as shown in Figure 4.

Figure 4: Effect of limited spectral resolution on measured spectrum

• Spectral range: The maximum range in wavelength for which the spectral properties of the
signal can be measured, as shown in Figure 5. Practically, range may be limited by the detector
response, the bandwidth of the reflecting mirror or the beam splitter.
 Figure 5: Spectral range of FTIR spectrometer

• Wavelength accuracy: It is quantified by how much the spectrum can be shifted from the true
value of the spectrum. Figure 6 shows an example of spectrum drift.

Figure 6: Effect of wavelength accuracy on measured spectrum.

• Signal to noise ratio (SNR): It determines the error in the spectrum due to different noise
sources. Noise sources include optical detector noise, electronic circuits thermal noise, ADC
quantization noise, errors in position measurement …
• Scanning time: Time required to perform one scanning cycle.
• Size or dimensions of system components: Light redirecting optical elements, IR light source,
photodetector, sampling compartment ...
• Temperature, power requirements, …etc.

4.3. MEMS based spectrometer:

MEMS acronym, which stands for “Micro-Electro-Mechanical-Systems”, is a miniature system or
subsystem involving one or more micro-machined components and often integrating other
functions on the same chip or in the same package. Implementing Michelson interferometer using
MEMS technology would have several advantages:
• Small size (micro scale).
• Self-alignment of micro-optical and mechanical components.
• Batch processing and low cost.

The transition of Michelson interferometer from bulky components into microsystems is visualized
in
Figure 7. The design is fabricated on silicon-on-insulator substrate using deep etching technology.
The moving mirror is driven by MEMS actuator that translates to certain positions due to applied
electrostatic potential. The MEMS chip is integrated with optical coupling fibers, photodetector
and electronic boards, that drives MEMS, reads detector’s signal and communicates with
computer, in a portable handheld module.

Figure 7: Transformation of bulky FTIR spectrometer system into miniaturized version1,2.

5. Procedure
1) Connect the setup shown in Figure 8 (a) to take reference measurement.
2) Open Neospectra SpetroMost software.
3) In Spectrum tab, Adjust the following parameters:
a. Measurement time: 180 sec.
b. Resolution: 16 nm.
c. Zero padding: 7
4) Press Background to take reference measurement.
5) Connect the setup shown in Figure 8 (b) to take sample measurement.
6) Press Run to take fiber attenuation measurement.
7) Repeat the previous steps for Resolution 8 nm.
8) Save the measured spectra.

1
B. Saadany, H. Omran, M. Medhat, F. Marty, D. Khalil, and T. Bourouina, “MEMS tunable Michelson
interferometer with robust beam splitting architecture,” in 2009 IEEE/LEOS International Conference on
Optical MEMS and Nanophotonics, 2009, pp. 49–50.
2
Y. M. Sabry, D. Khalil, and T. Bourouina, “Monolithic silicon-micromachined free-space optical
interferometers onchip,” Laser Photon. Rev., vol. 9, no. 1, pp. 1–24, Jan. 2015.
 (a)

(b)

Figure 8: Fiber attenuation spectrum measurement setup using MEMS FTIR spectrometer, (a)
reference measurement, (b) sample measurement.

6. Report
Part 1:
a. Plot measured spectra at both resolutions on the same graph.
b. Identify the measured fiber attenuation peak wavelength.
c. Identify the resolution effect on absorption peak and signal to noise ratio of measurement.
Part 2, Answer the following questions:
a. what are the advantages of a MEMs integrated spectrometer over a bench top Michelson
interferometer setup.
b. Write down a method to practically measure the resolution of the interferometer.
c. Mention multiple uses of optical interferometers.

Deliverables and Notes:

• Single PDF including the 2 parts. The measured results are included in part 1.
• Answers of questions of Part 2.
• Cover Page with your names, section and group.
• Show clear figures with labelled axes.