Branching Function by Semiconductor Optical Amplifier (SOA) in Subcarrier Multiplexed (SCM) Optical Systems Eszter Udvary, Tibor Berceli Budapest University of Technology and Economics Department of Broadband Infocommunication Systems Tel.: (36)-1-463-2634, Fax: (36)-1-463-3289, e-mail:
[email protected] H-1111 Budapest, Goldmann György tér 3, Hungary
Abstract - The paper describes the theoretical and experimental studies of the modulation and detection functionality and the optimal working state of the multifunctional SOA. The subcarrier multiplexed system with SOA branching stages is studied and presented. I.
INTRODUCTION
The proper goal for long distance optical links is the increase in the transmission capacity. However, there are many applications in which cases other points are more important. Two complementary technologies, the optical fibre and wireless communications drive explosive growth in communications. In a combined optical and wireless communication system the mobility is provided by the wireless system part and the flexibility is ensured by the optical system because optical fibre have massive bandwidth. Some new approaches are planned to improve the performance of the combined system. The two technologies start to converge into fibre-radio systems, in which broadband services are delivered over fibre to millimeterwave picocells. The wide band of the optical highway makes possible the transmission of the radio channels without changing their format, the radio carrier frequencies are considered to be subcarriers in the optical transmission. This combined technique is used or will be used in several systems (mobile communications, local area and subscriber access networks, television programme distribution, Internet access, etc.). The fibre radio technology has a number of significant advantages compared to either copper cable, radio or baseband optical systems. Subcarrier multiplexing allows the radio frequency carriers to be directly transported over the optical fibre without need for frequency conversion or multiplexing/demultiplexing functions. The complex processing equipment of the radio distribution point can be located in a central exchange. Hence the field installation and maintenance procedures are simplified and physically small pole mounted distribution points can be used. Moreover the subcarrier type optical transmission offers a high flexibility because the number and the frequencies of the subcarriers can easily be modified according to the traffic needs. [1]
In an optically supported millimetre wave cellular radio system the noise figure or output signal to noise ratio and intermodulation free dynamic range are the most important parameters. Ideally it would be most beneficial to improve both of them. Great dynamic range and high linearity are necessary to have good system performance and avoid channel crosstalk. These parameters can be determined by the optical devices and depend on frequency and level of optical reflection. SOA traditionally can be used in any system that is loss limited to compensate for the optical losses. Local area network and the fibre radio system, where the main losses are from optical power splitters, branching and taps are also loss limited and can benefit from optical amplifiers. The SOA produces wide band optical noise, it can be suppresed by the application of a narrow band optical filter. It will else decrease the signal to noise ratio, but it depends on the position of the device [2]. Same time the RF response of the optical link without SOA has deep notches. These notches are shifted to higher frequencies and the microwave or millimeterwave carrier suppression effect is reduced, because of the negative chirp of the SOA. [3]. Additionallty the nonlinearity of external optical modulator can be dimished by saturated SOA. Summerizing, the well adjusted SOA can improve the transmission performance of analog optical link. [4] II.
BRANCHING FUNCTION
The SOA may be well used as a multifunctional device for the branching function in SCM systems [5]. In this approach the SOA operates as a modulator to add a new channel, as a detector to drop the needed channel and as an in-line amplifier to amplify the other channels, simultaneously. It realizes a compact, small size and cost-effective radio repeater for signal distribution. These modes of operation have not been dealt with extensively, although it seems important to understand whether SOA’s can be used as efficient high-speed modulators and detectors. A simplified experimental set-up (Fig.1) was tested and the device gives favorable properties according to the experimental studies.
2nd Stage
1st Stage Laser Popt Source
SOA m
Ibias
Ibias
1st channel
3rd ch.
2nd ch. Branching 1st ch. filter Low Noise Amplifier
Detector
SOA
SOA
Power Amplifier
Low Noise Amplifier
1st & 2nd& 3rd channel
1st or 2nd
Power Amplifier
Fig.1. Simplified experimental setup block diagram In these multi-channel analogue optical fiber links high signal to noise ratio and at the same time low nonlinear distortion are required, because of the cascability. The separation of add and drop channels can be achieved by an electrical branching filter or an electrical circulator. In the first case the realisation of a reconfigurable add/drop multiplexer is difficult, in the second case an electrical filter is needed also. In the drop branch the signal is amplified by a high power RF amplifier, than it is radiated via the antenna. In the add branch a low noise pre-amplifier provides the suitable SNR for the SOA modulator input. Additionally, it is difficult to give individual optimisation of bias current for the SOA, we have to make a compromise, because different bias points are optimal for amplifier, modulator and detector functions.
15
14
10
12
5
10
0
8
-5
6
-10
Temperature=20C Wavelength=1550nm
-15 -20
4
Optical Gain
Optical Gain [dB]
III. ADD FUNCTION The add function means RF-to-optical conversion. For that function the SOA is used as an external modulator. The electrical bias current is modulated, the material gain is modulated, consequently in case of CW input the intensity of the output power will be modulated [6]. The optical gain-bias current curve can be divided into three parts (Fig.2). In the first one the amplification just starts, the second one is the near linear region and after it the slope of the optical gain starts decrease. The most linear region of the optical gain (or output optical power) versus bias current graph is the most suitable. The middle of this region should be chosen for operation point, because of the low static non-linear distortion effect and the high slope.
2 0
0
100
200
300
400
Bias Current [mA]
Fig.2. Optical gain depends on the bias current
The operation can be derived from the rate equations, when the time dependent current contains an invariant current part and a modulated part. Considering smallsignal modulation the number of carriers and photons are also time dependent and the shape of these parameters are similar to the shape of the modulation. Another approach based on the slope of the measured optical gain curve (md) can be derived. The current and the optical gain of the device are: I (t ) = I 0 + ∆I mod ⋅ cos(ωt ) (1) G (t ) = G0 + ∆G ⋅ cos(ωt )
where I0 is the constant (dc) current, ∆I is the current modulation amplitude, G0 is the constant optical gain of SOA, ∆G is the modulation. Hence the optical signal at the output of the SOA-modulator takes the form Pout = G0 ⋅ Pin ⋅ (1 + m ⋅ cos(ωt )) (2) where the modulation index (m) is 2 ⋅ Pmod ∆G m d ⋅ ∆I mod md m= = = ⋅ (3) G0 G0 G0 Z where Pmod is the modulation electrical power, Z is the microwave impedance of SOA. The intensity modulated output signal of the SOA-modulator is transmitted via optical fiber and the information is detected at the end of the optical link by an opticalelectrical converter. The detected electrical power (Pdet) P2 Pdet = η 2 ⋅ in2 ⋅ m d2 ⋅ Pmod (4) a η is the detection efficiency, a is the optical loss between the SOA and the detector. So the modulation depth proportional to the slope of gain curve and the electrical modulation power, but it is in inverse relation to the average optical gain. However the detected electrical power increases when the modulation power (direct relation), the slope of the gain curve and the input optical power of the SOAmodulator (quadratic relation) increase. Same conclusions can be observed from the experiments. Naturally the modulation depth realized by the SOAmodulator can be computed from the measured detected electrical power with the knowledge of the modulation power and the input average optical power of the detector.
Detected Electrical Power [dBm]
opt is the average optical power at the input of where Pdet the detector. During our work the detected electrical power was measured with different parameters, Fig.3 describes the typical experimental results. The upper curve shows the detected power as a function of the bias current of the SOA-modulator. The three regions are well seen in this figure. The injection current is not enough for the expected work and the detected power is low in the first part. The power is near constant in the second, linear part and after it the detected product start decrease, because the slope of the gain curve falls. The lower curve represents the result without input optical power, i.e. just the amplified spontaneous emission power (ASE) is modulated and the broadband O/E converter can detect this poor fluctuation.
-40
working state in the saturation region. The results represent that in the first part of the graph the gain of the device increases, hence the IP3 and the SFDR improve. In the second part the optical gain doesn’t change significantly but the noise level and intermodulation products slope, hence the IP3 and the SFDR decrease. The noise effect and the nonlinear distortion products are more significant in case of strong optical reflection level, i.e. without optical isolators. The device ensures efficient SFDR for the general optical networks (>90dB) in this nonlinear operation region, too. 25
100
0
95
-25
90
-50
85 Modulation:0dBm Wavelength=1550nm 199&200MHz Optical Input:1mW Isolators
-75
-100
-50
-125
Pin=850uW Pin=0
-60 -70
-100 0
100
200 300 400 Bias Current of SOA [mA]
75 70 65
500
50
100 150 200 Bias Current [mA]
250
300
Fig.4. Nonlinear behavior of SOA modulator
Frequency=400MHz Temperature=25C Modulation Power=-30dBm Optical Bandwidth=400nm
-90
80
-150 0
-80
IP2 IP3 noise floor SFDR
2/3
(5)
SFDR [dB Hz ]
2⋅ P 2 ⋅ a 2 ⋅ P det = opt det 2 2 2 2 Pin ⋅ G0 ⋅ η det ⋅ Z Pdet ⋅ η det ⋅Z
IP2, IP3 [dBm] Noise [dBm/Hz]
m=
600
Fig.3. Modulation with SOA The detected electrical power is high in case of SOA modulator, because of the optical gain of the SOA in contrary to the optical insertion loss of other external modulators. The SOA has rapid response time, hence the electrical circuits determine the modulation bandwidth. In the linear operation range the SOA as a modulator provides lower nonlinear distortion, than the other external modulators. Cascadability is critically important in optical SCM networks where several electrical subcarriers are transmitted on the same optical signal. Degradation of the transmission system will occur due to the crosstalk between the subcarriers (nonlinearity) and noise expandation (ASE). In the intermodulation experiments the SOA modulator was biased and modulated by the sum of two microwave signals. The output noise and signal levels were measured for the fundamental (P1), the second (P2) and the third (P3) order harmonic mixing products. It depends on many things: bias, temperature, laser structure and the optical reflection [7]. In linear regime the SOA modulator shows low, not measurable nonlinearity because the noise generated by the SOA modulator will dominate in the system. The intermodulation products overcome the noise floor in case of extraordinary high modulation indices. Fig.4 shows the noise level, IP2, IP3 and SFDR versus SOA
Summarizing, the SOA modulator requires low modulation power, the detected electrical power is high enough because of the optical gain of the SOA in contrary to the optical insertion loss of other external modulators. The SOA has rapid response time, hence the electrical circuits determine the modulation bandwidth. In the linear operation range it provides lower nonlinear distortion than the other external modulators. The SOA has relevant optical noise and the inherent design trade-off between efficiency and linearity demands more advanced amplifier-modulator working state planning. IV.
DROP/DETECTION FUNCTION
The drop function means optical-to-RF conversion, in this situation the SOA operates as a detector [8]. Two different mechanisms induce detection in the SOA. Operated at an injection current corresponding to an electron density below transparency, the device works as a photodetector and the detection signal arises from absorption of the injected light and the creation of electron-hole pairs. However an injection current above transparency, that is, in the usable amplifying regime, the injected optical signal will cause stimulated transitions, which will reduce the carrier density in the gain medium. Due to these two different mechanisms of interaction, it is clear that the detected electrical signal will change polarity at transparency. Fig.5 depicts the electrical current as a function of the input optical power of SOA-detector. If the input signal is intensity modulated, the fluctuation in the optical intensity due to modulation will induce fluctuation in the injection current.
V.
Change of Current [mA]
7 6 5 4
Reference Bias Current=150mA Wavelength=1550nm Temperature=20C
3 2 0
100
200
300
400
500
600
700
Input Optical Power [mW]
Fig.5. Current change versus average optical power An equivalent circuit can be applied for modelling of SOA detector operation mode, which can be derived from the rate equations with time dependent input optical power. We shall consider sinusoidal intensity modulated input optical signal P P Pin = 0 + m ⋅ 0 ⋅ cos(ωt ) (6) ain ain Hence the detected electrical current has cosinusoidal component I det SOA = I DC + ∆I det SOA ⋅ cos(ωt ) (7) where P0 is the constant optical power, ain is the optical loss before the SOA-detector, m is the modulation depth, IDC is the constant detected current, ∆IdetSOA is the modulation information detected by the SOA. The SOA-detector responsivity (R) can be computed from the detected electrical power (PdetSOA)
2 ⋅ a in2 ⋅ Pdet SOA (8) 2 m 2 ⋅ PDC ⋅Z The detection responsivity of the SOA detector is smaller than that of a conventional PIN photodiode, but SOA provides multifunctions simultaneously. It means an in-line detector, which amplifies the incoming signal and parallel detects the information. When the bias current increases, i.e. the population inversion and the gain are higher the detected power and the responsivity of the device also increase (Fig.6). The diagram follows the nature of the optical gain curve. PDC ⇒R= ain
0,05
-80
0,04
-90
0,03
-100
0,02
-110
Wavelength=1550nm
Transparency
0,01
Input Optical Power=40uW
-120
Modulation depth=20% Temperature=20C
0
-130 0
50
100
150
Bias Current [mA]
Fig.6. Detection with SOA
200
Detected Power [dBm]
Responsivity [A/W]
∆I det SOA = R ⋅ m ⋅
CONCLUSION
This paper presents the applications of Semiconductor Optical Amplifiers (SOAs) in subcarrier multiplexed (SCM) fiber radio systems with theoretical and experimental approach. In these multichannel analogue optical fibre links high signal to noise ratio and at the same time low non-linear distortion are required. The paper presents a concept with semiconductor optical amplifier branching stages and the optimal operation point is studied. The SOA gives favourable properties as a multifunctional device according to the experimental studies, hence it offers a fine solution for the add/drop functions problem. REFERENCES
[1] A.J.Seeds: “Broadband Fibre-Radio Acces Networks”, MWP’98, October 12-14 1998, Princeton, New Jersey [2] Tamás Marozsák, Eszter Udvary, Tibor Berceli: “Transmission Characteristics of All Semiconductor Fiber OpticLinks Carrying Microwave Channels”, 30th European Microwave Conference, Vol. 2. pp.52-55, Paris, France, 3rd-5th October 2000 [3] J.Herrera, F.Ramos, J.Marti: ”Semiconductor Optical Amplifier Chirp Model for Dispersive and Nonlinear Transmissions Throught Single-mode Fibers in Microwave Optical Link”, MWP2002, pp.201-204, 2002 [4] Sang-Yun Lee, Bon-Jo Koo, Hyun-Do Jung, and Sang-Kook Han: „Reduction Of Chromatic Dispersion Effects And Linearization Of DualDrive Mach-Zehnder Modulator By Using Semiconductor Optical Amplifier In Analog Optical Links”, ECOC2002, Coppenhagen, Denmark, 2002 [5] Hiroyo Ogawa, K. Horikawa, H. Kamitsuna. O. Kobayashi, Y. Imaizumi, I. Ogawa: “Application of Semiconductor Optical Amplifiers to Microwave Signal Processing”, IEEE MTT-S Digest, pp1177-1180, 1995 [6] Jesper Mork, Antonio Mecozzi, Gadi Eisentein: “The modulation response of a Semiconductor Laser Amplifier”, IEEE Journal on Selected Topics in Quantum Electronics, vol.5, No.3, May/June 1999, pp. 851-860 [7] Tamás Marozsák, Eszter Udvary, Attila Kovács, Tibor Berceli: “Effect of Optical Reflection on Nonlinear Characteristics of Direct Modulated Lasers”, paper submitted to MWP 2003, 10-12 September 2003, Budapest, Hungary [8] M.Gustavsson, A.Karlsson, L.Thylen: “Traveling wave semiconductor laser amplifier detectors” J. of Lightwave Technology, vol.8., pp.610-617,1990
