Microwave and optical technology Letter

15. K.L. Wong, W.Y. Chen, and T.W. Kang, On-board printed

coupled-fed loop antenna in close proximity to the surrounding

ground plane for penta-band WWAN mobile phone, IEEE Trans

Antennas Propag 59 (2011), 751–757.

16. K.L. Wong and S.C. Chen, Printed single-strip monopole using a

chip inductor for penta-band WWAN operation in the mobile

phone, IEEE Trans Antennas Propag 58 (2010), 1011–1014.

17. SPEAG SEMCAD, Schmid & Partner Engineering AG, Available

at: http://www.semcad.com.

VC 2012 Wiley Periodicals, Inc.

ESTIMATION OF COUPLINGPARAMETERS FOR AUTO-MOTORIZEDFABRICATION OF FUSED FIBERCOUPLER

Dedi Irawan,1 Saktioto,1,2 Jalil Ali,2 Mohamed Fadhali,3

and Erwin 1

1 Faculty of Science, APSI, Universiti Teknologi Malaysia, Johore,Malaysia 81310; Corresponding author: [email protected] Faculty of Math and Sciences, Department of Physics, Universityof Riau, Pekanbaru, Indonesia3 Faculty of Science, Department of Physics, Ibb University, Ibb,Yemen

Received 27 October 2011

ABSTRACT: A directional fiber coupler with exertion loss 0.03 dB has

been successfully fabricated using fusion technique with typical couplingratio 1–90%. The coupling region of two twisted single mode fiber isheated by injecting hydrogen gas at 2.5 bar. During fusion process, both

two sides of fibers are pulled by stages that are automatically motorizedin range of 800–4800 lm, and stopped when the desired coupling ratio

is reached. The parameters of automated mechanical motion of pullingstages and coupling parameters at fusion region have been calculatedby using kinetic model. The effect of heating and elongation reduces the

diameter of cross section tapered region with a diameter 6.35micrometer scale. As the fabrication of fiber couplers described bydegree of fusion, which is function of heating and pulling length, it can

be seen clearly that the coupling coefficient between the fibers increasesexponentially with increasing the degree of fusion. However, by knowing

coupling power and mechanical motion parameters, the fabrication ofdirectional fiber coupler can be optimized. VC 2012 Wiley Periodicals,

Inc. Microwave Opt Technol Lett 54:1932–1935, 2012; View this article

online at wileyonlinelibrary.com. DOI 10.1002/mop.26937

Key words: directional fiber coupler; pulling length; degree of fusion;coupling ratio; coupling coefficient

1. INTRODUCTION

The use of optical devices on network communication system has

been widely expanded. It provides an optical circuit that can carry

out data in terms of audio, video, data processing, etc. with big

capacity, low loss, and faster. Optical directional fiber coupler is

a passive device and main component of optical networking sys-

tem. It is used to split, to combine, or to switch optical signal. As

the networking system is built by using optical components, a

directional fiber coupler with various coupling ratio is always

needed. The fiber coupler can be fabricated by using twist-etching

techniques, polishing technique, and fusion technique [1, 2].

Fusion and elongation is an easier technique that has been

used to fabricate fiber coupler. In 2005, Fused bi-conical tapered

(FBT) coupler was fabricated by this technique. The coupling

region was heated by CO2 laser, and resulting good FBT cou-

plers with 3-dB splitting ratio [3]. Because of CO2 laser beam is

a high cost technology, fusion technique by injecting H2 gas to

the torch flame was purposed for hearting the coupling region

[4]. During fusion, fibers are elongated in micrometer scale until

the coupling ratio is reached. Since that fabrication results high

coupling loss, the study of the automated fiber coupler fabrica-

tion system based on fusion technique becomes necessary to

determine coupling parameters [5–9].

In this article, the coupling parameters are estimated to optimize

auto control parameters for fabrication of directional fiber coupler

so that good performance of directional fiber coupler reached.

2. THEORETICAL CONSIDERATION

The propagation of an optical signal in the coupled waveguide

medium was determined clearly from the Maxwell equation

using coupling-mode theory method [5]. The modeling and

experiment of power parameters SMF coupler were also studied

[6]. It showed the coupling ratio as the function of coupling

coefficient between the fibers. As the refractive indices of

coupled fibers are constant, and geometrical fibers are also iden-

tical as shown in Figure 1, the amplitude of power exchanges

between two fibers given by Eq. (1).

PaðzÞPbðzÞ

� �¼ cosrz� jd

r sinrz �jj2r sinrz

� jj1r sinrz cosrz� jd

r sinrz

" #Pað0ÞPbð0Þ

� �(1)

where Pa and Pb are power amplitudes in fibers 1 and 2, respec-

tively. The coupling coefficient is denoted by k, and L is cou-

pling length. As the coupling region is tapered due to fusion and

elongation, and by defining the degree of fusion f is a factor that

describes how close two fibers joined or f ¼ x/y, a simple rela-

tionship between degree of fusion separation between fiber’s

core d can be written as follows:

Figure 1 Illustration of geometrical fused fiber coupler, (a) tapered

directional fiber coupler and (b) cross section of coupling region

1932 MICROWAVE AND OPTICAL TECHNOLOGY LETTERS / Vol. 54, No. 8, August 2012 DOI 10.1002/mop

Figure 3 Power output of fundamental fused fiber coupler. [Color

figure can be viewed in the online issue, which is available at

wileyonlinelibrary.com]

d ¼ 2bf1� f ð2�ffiffiffi2

pÞg (2)

with 1 � f � 2 or 0 � f � 1. When f ¼ 2 or f ¼ 0, this means

two core fiber completely does not joined, increasing that value

means two fiber begins coupled, and separation between them

closer and closer until f ¼ 1 that means two fiber completely

join and assumed as a circular.

To calculate the coupling parameters at the output ports, a

kinetic model of continuity equation given by Eq. (3) is deter-

mined to the coupling ratio [4].

@j@t

þrðmjÞ ¼ S (3)

@j@ðCRÞ ¼

@j0@ðCRÞ þ f ðk;PÞ þ f ðTÞ (4)

If the coupling ratio is defined as a relative power at output

fiber 2 Pb to the total input power at the input ports, Eq. (4) can

be written as follows:

CR ¼ j2

j2 þ d2sin2 ðj2 þ d2Þ1=2z

h i� 100% (5)

An empirical coupling coefficient, k can be calculated from the

experimental work of normalized frequency,

j ¼ p2

� � ffiffiffid

p

a

!e½�ðAþ B�d þ C�d2Þ� (6)

where

A ¼ 5:2789� ð3:663VÞ þ ð0:3841V2ÞB ¼ �0:7769� ð1:2252VÞ þ ð0:0152V2ÞC ¼ �0:0175� ð0:0064VÞ þ ð0:0009V2Þ@ ¼ ðn21 � n22Þ=n21�d ¼ 2bf1� f ð2� ffiffiffi

2p Þg

a

Figure 2 Experimental set up to fabricate SMF coupler. [Color figure

can be viewed in the online issue, which is available at

wileyonlinelibrary.com]

3. EXPERIMENTAL SET UP

An experimental set up shown in Figure 2 is based on fusion and

elongation techniques. It is carried out by placing two or more op-

tical fibers on the stages. Corning fiber (SMF-28eVR ), with diame-

ter of core and cladding 125 and 8.2 micrometer, respectively, are

connected to the laser source (1310 nm) and displayed to the

photo detector. The two fibers are twisted and held by a vacuum

system in both stages. All components are recorded and by a data

acquisition card installed to the computer system. The initial step

is to set parameters such as coupling ratio, maximum pulling

length, x–y–z position of torch flame, and flowing of H2 gas.

After setting the parameters set up, the system has to be hom-

ing process or initial process. One milliwatt laser launched to the

one of input ports is detected by photo detector and kept for cali-

bration. At the same time, fusion and pulling process are started.

During torch flame heating the coupling region, the fibers are

elongated by pulling stages with suitable pulling speed. Heating

and pulling process will be automatically stopped when the preset

coupling ratio is reached. The final step is the coupling region

saved from any vibration by packaging it. All mechanical motions

of this fabrication system are motorized in micrometers scale.

4. RESULTS AND DISCUSSION

Two single mode fibers have been successfully coupled with

splitting ratio shown by Figure 3. The coupling region was fused

and pulled until 800–1350�C and 4800 micrometer, respectively.

During the coupling region heated by injecting H2 gas in 2.5 bar,

the pulling stages are motorized to pull the coupling region.

Although elongation and fusion process change the geometrical

fibers, at the same time, there are small changes in gradient of re-

fractive index. This is due to two fibers joined, with air as third

refractive index, and then the coupling ratio is slowly increases as

a function of pulling length. Figure 4 shows the coupling ratio at

the output ports of fiber coupler for two typical wavelengths

(1310 and 1550 nm). It is gradually increases until the coupling

length about 2.5 mm, then sharply increases to 90%. It is possible

to reach the typical of coupling ratio 99% by increasing the pull-

ing length and fusion period. It can also be seen that change of

the coupling ratio is slightly periodically, and become sharply

change with higher pulling length. This means refractive index of

coupled fiber become higher until the fiber geometry change at

the maximum starching and fusion process.

As the degree of fusion describes how strong two or more

fibers joined after fusion, it is clear that the increment of the

coupling ratio reduces the diameter of coupling region. Figure 5

DOI 10.1002/mop MICROWAVE AND OPTICAL TECHNOLOGY LETTERS / Vol. 54, No. 8, August 2012 1933

shows the geometrical coupling region after fusion for coupling

ratio 90%. Before two single-mode fibers joined, it has the di-

ameter of coupling region 250 lm.

During fusion, it will gradually decrease as a function of

fusion and pulling length. Experimentally, by heating and pull-

ing the coupling region, the coupling ratio 50% was reached

with coupling diameter (x) decreased to be 125.64 lm, and the

each fiber’s diameter (y) is 82 lm. This means the degree of

fusion f50% � 1.513. Figure 4(b) shows that to reach coupling

ratio 90% reduces the diameter of coupling region until 35.59

lm, and the calculation of degree of fusion is f90% � 92.431.

However, heating and elongation process is a function of

decreasing tapered coupling region diameter. By assuming width

of high temperature or hot zone is Dz, the tapered-cross section

diameter D (z) at any z position is given as follows:

DðzÞ ¼ 2r0expðz=DzÞ (7)

where r0 is radius of twisted coupler before fusion. Figure 6

shows the change of tapered cross-section diameter at any cou-

pling length or z. It can be seen that the diameter changes is a

parabolic function, which follows the profile of torch flame. The

minimum diameter of coupling region or cut-off diameter is

about 6.35 micrometer, and it cannot achieve to zero, this means

that the heating and elongation process be able to couple two

single-mode fibers until it is minimum at coupling ratio near to

100%. For example, to get experimentally the coupling ratio

90%, the coupling region diameter was reduced and it becomes

�35.59 lm. As the flow rate of H2 gas is controlled to obtain

stable torch flame with a stable temperature distribution in the

coupling region area, the diameter of coupling region will fol-

low the theoretical profile, and it will result a uniform propaga-

tion constant at fundamental single-mode tapered directional

fiber coupler.

The fabrication of SMF coupler based on auto-motorized sys-

tem depicts how the coupled region formed by fusion process. It

is clear that change in geometrical fiber affects the refractive

index of fiber’s core and cladding, so that the light can be

transferred from fiber 1 to other fibers. As the power transfer

between the fibers is described by the coupling ratio at output

ports, the coupling coefficient is an important parameter that

must be controlled during fusion process. The coupling coeffi-

cient depends on separation between the fiber’s axes. Certainly,

fusion and elongation reduce it according to the decreasing fiber

diameter at the coupling region. As the fibers’ axis becomes

closer, the coupling coefficient between them will be increased

exponentially by increasing the degree of fusion.

By determining the degree of fusion as how closer two or

more fibers joined by heating and elongation, and function of

cross-section tapered diameter, the coupling parameter can be

optimized. As the coupling length can be controlled by adjusting

the pulling length, the coupling parameter can be shown in three

dimensions. Figure 7 clearly describes that to get high coupling

ratio need longer coupling length, which can be reached by

increasing the heating and elongation process.

Figure 4 Ratio of power exchange due to increase the pulling length

Figure 5 Diameter of coupling region Fused SMF coupler with typi-

cal CR 90%. [Color figure can be viewed in the online issue, which is

available at wileyonlinelibrary.com]

Figure 6 Diameter profile of tapered cross section in coupling region

fiber coupler

Figure 7 Coupling parameter of SMF coupler. [Color figure can be

viewed in the online issue, which is available at wileyonlinelibrary.com]

1934 MICROWAVE AND OPTICAL TECHNOLOGY LETTERS / Vol. 54, No. 8, August 2012 DOI 10.1002/mop

5. CONCLUSION

The parameter of auto-motorized fiber coupler fabrication sys-

tem based on fusion and elongation technique has been esti-

mated. It is found that the power output described by coupling

ratio significantly depends on the degree of fusion and coupling

length. Increasing the degree of fusion will increase the coupling

coefficient between the fibers. During heating process, the pro-

file of coupling-region diameter is a parabolic function with

minimum diameter �6.35 micrometer. However, a directional

fiber coupler can be fabricated with high accuracy coupling ratio

by controlling the degree of fusion and elongation process.

ACKNOWLEDGMENTS

Wewould like to thank the Institute of Advanced Photonic Science,

Faculty of Science, Universiti Teknologi Malaysia (UTM) Spon-

sored by GUP Tier 1 No.QJ13000, 712601h02, and Physics Dept.

University of Riau, Indonesia for generous support in this research.

REFERENCES

1. A. Hosseini, D.N. Kwong, Y. Zhang, Y. Liu, and R.T. Chen, The

optimum design for 1xN multimode interference coupler based

beam splitters, OSA/IEEE Integrated Photonics Research, Silicon

and Nano Photonics (IPR), paper JTuB9, 2010.

2. H.S. Daniel, D.R. Moore, and V.J. Tekippe, Broad band of MXN

fiber and method for making, United States Patent, Patent num-

ber:5,355,426, Millersville, 1994.

3. Saktioto, J. Ali, and M. Fadhali, Theoretical and empirical compar-

ison of coupling coefficient and refractive index estimation for

coupled waveguide fiber, J Appl Sci Eng Technol2008.

4. Saktioto, J. Ali and M. Fadhali, Simplified coupling power model

for fiber fusion, J Opto-Electron Rev 17 (2009), 193–199.

5. A.W. Snyder, Coupled-mode theory for optical fibers, J Opt Soc

Am 62 (1972), 1267–1277.

6. A.W. Snyder and J.D. Love, Optical waveguide theory, Institute of

Advanced Studies Australian National University, Canberra, Aus-

tralia, 1983.

7. D.W. Stove and M. Corke, Fiber optic power combiner, United

States Patent, Patent number.5,121,452, Medfield,1992.

8. Y. Wang, D. Xue, and X. Lu, Power transfer characteristics among

N parallel single-mode optical fibers, Science direct, Zhejiang Uni-

versit, Hangzhou, Zhejiang, China, 2007.

9. DediIrawan, Saktioto, and J. Ali, Linear and triangle order of NX3

optical directional couplers: Variationcoupling coefficient, SPIE

Proc 7781 (2010), 77810J.

VC 2012 Wiley Periodicals, Inc.

A 1–12-GHz VARIABLE-GAINLOW-NOISE AMPLIFIER MMIC USING0.25-lm SiGe BiCMOS TECHNOLOGY

Woojin Chang, Sang-Heung Lee, Jae-Kyoung Mun, andEunsoo NamRF Convergence Component Research Team, Department ofPhotonic/Wireless Convergence Components Research,Convergence Components and Materials Research Laboratory,Electronics and Telecommunications Research Institute (ETRI), 161Gajeong-Dong, Yuseong-Gu, Daejeon, Korea; Correspondingauthor: wjchang @etri.re.kr

Received 27 October 2011

ABSTRACT: This article introduces an 1–12-GHz differential two-stage variable-gain low-noise amplifier (VGLNA) using 0.25-lm SiGe:CBiCMOS commercial process technology for ultra-wideband system. The

results of the fabricated monolithic microwave integrated circuit

amplifier show 18-dB gain with a 3-dB frequency band of 1.3–11.9 GHz

and noise figure of less than 5 dB under the bias condition of 2.5-Vsupply voltage and 55-mW total dc power consumption. The gain-control range is from �17 dB to þ18 dB. The chip size of the

manufactured VGLNA is 1.1 � 0.9 mm2 including all testing pads forRF and dc probes. VC 2012 Wiley Periodicals, Inc. Microwave Opt

Technol Lett 54:1935–1937, 2012; View this article online at

wileyonlinelibrary.com. DOI 10.1002/mop.26936

Key words: variable-gain low-noise amplifier; silicon-germanium;BiCMOS; ultra-wideband; microwave monolithic integrated circuit

1. INTRODUCTION

The wideband operating requirements of the ultra-wideband

(UWB) RF front-end makes the design very difficult. Low-noise

amplifiers (LNAs) for UWB systems should provide high gain

and low noise figure over a wideband. Variable-gain low-noise

amplifiers (VGLNAs) were realized to maximize the overall sys-

tem dynamic range. As it is difficult to get gain over a wide

bandwidth for various gain levels, distributed amplifiers can be

used to make a wideband operation. For a distributed amplifier,

the small signal gain can be controlled by the bias voltage and

current of the devices. However, a conventional distributed am-

plifier has a low gain control range (GCR) and a high power

consumption. A variable-gain distributed amplifier with 7-GHz

bandwidth was reported in Ref. 6, but the GCR was limited

from �10 to þ8 dB. And another variable-gain distributed am-

plifier with 10 GHz bandwidth was reported in Ref. 7, but it had

the total power consumption of 40 mW although it was not a

differential amplifier but a single-ended amplifier.

In this article, we propose a wideband, VGLNA, which is

composed of current mirror and differential two-stage amplifier.

To achieve completely differential operation of the amplifier,

the same current from the current mirror flows identically

through each transistor of the differential amplifier. And the

resistors in bias circuits of the amplifier are used instead of

inductors in the bias circuits for wideband frequency operation.

2. 1–12-GHz VGLNA DESIGN

A 1–12-GHz VGLNA was designed and fabricated by using

IHP commercial 0.25-lm SiGe:C BiCMOS technology. The

VGLNA for UWB systems is shown in Figure 1. The VGLNA

is composed of a current mirror and a two-stage differential

Figure 1 Schematic of 1–12-GHz VGLNA

DOI 10.1002/mop MICROWAVE AND OPTICAL TECHNOLOGY LETTERS / Vol. 54, No. 8, August 2012 1935