**3.3.2 Intensity modulation**

Practically, the average optical gain and the slope of the optical gain - bias current curve determine the optimum working state of the SOA as a modulator. The curve can be divided into three parts (Fig.14.). In the first one the amplification just starts and it is not effective, the second one is the almost linear region and after it the slope of the optical gain starts to decrease. The middle of linear region of this curve should be chosen for operation point, because of the low static non-linear distortion effect and the high slope [Udvary2].

Fig.14. shows the measured optical gain and modulation behavior of the SOA-modulator versus bias current. The three regions are well seen in this figure, too. The injection current

Multi-Functional SOAs in Microwave Photonic Systems 99

2/3 72 83 dB Hz SFDR is required [Olshansky]. The determination of SFDR, IP2 and IP3

Modulation frequencies=199, 200MHz, Modulation Power=4dBm, Input Optical

1

*P (P P ) P (P P ) SFDR P (P P ) P*

*SFDR[dB] IP [dBm] P [dBm]*

The experimental work was done on different types of SOAs. The presented results characterize a commercial SOA having 13 dB of small signal gain, 15dBm of saturation power, 100nm of optical bandwidth, 300mA of bias current. All the measuring instruments were checked to have higher dynamic range and better linearity than the value expected from the SOA-modulator. 7% modulation depth was applied, because the modulation indices are usually less than 10 % in typical SCM systems. However, it was also checked for

The nonlinear behavior depends on several parameters [Marozsak]. Fig.16. shows the noise level, IP3 and SFDR versus bias current. The graph can be divided into three parts. First, the SOA is near the transparency, the nonlinearity is high; hence the IP3 and the SFDR improve as it approached the near linear range. In the second part the modulation and nonlinear products don't change significantly but the noise level rises, hence the SFDR decreases. Finally, the intermodulation products start rising and the degradation of the SFDR is faster. The shapes of the curves are similar for the results estimated from the one tone simulations. The reason of some difference (some dB) in the exact values is that for the measurements the SOA device's internal parameters were not available. So we could only use estimated

*IP [dBm] P [dBm] P [dBm]*

1

*IP [dBm] P [dBm] P [dBm]*

2 2 <sup>1</sup> 3 3 2

1

<sup>2</sup> <sup>3</sup> 3

*th*

*noise*

*sec*

1

*in th noise th noise in noise noise*

are presented in Fig.15.

Fig. 15. Determination of SFDR, IP2, IP3.

a wide range of modulation depth (3-30%).

parameters for the simulations.

Power=1mW, without Isolators, Temperature=20°C

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 starts to decrease because the slope of the gain curve falls. The lower curve at modulation behavior figure represents the result without input optical power, i.e. just the amplified spontaneous emission power (ASE) produces the modulated signal and the broadband O/E (optical-toelectronic) converter can detect this poor fluctuation. This effect can be dramatically decreased by a narrow band optical bandpass filter.

Fig. 14. Measured Optical Gain (a) and Modulation behaviour (b) of SOA-modulator versus bias current of the device. Wavelength=1550nm, Temperature=20C, Modulation power=-30dBm, Modulation frequency=400MHz

The modulation bandwidth is limited by the speed at witch the carrier density can be changed, this is usually limited by the spontaneous lifetime of the carriers in the SOA (in the nanosecond range). The lifetime in the presence of a strong, saturating input signal is reduced due to stimulated recombination. The real speed depends on the structure of the device, but in general it is larger than 10GHz speed.
