the CW light from the two SOAs interferes destructively.
According to the XOR operation in Fig. 1(c), no pulse is
observed at the output port I of the interferometer .
However, the phase
will vary in accordance to the
input pattern, as depicted Fig. 1(b) .
B. Experimental Setup
Fig. 2 illustrates the experimental setup of the proposed
optical phase modulator. It consists of an external cavity laser
peaking at 1549.32 nm, followed by a polarization controller
and a Mach-Zehnder external modulator. The NRZ data signal
generated by a serial BERT (Agilent N4901B) is then
amplified by an EDFA (IPG-EAD-500-C3-W) and split into
two equal parts using a 3 dB coupler. Both signals are
synchronized using optical delay lines. Polarization
controllers are included at ports A and D of a MZI-SOA (CIP
40G-2R2-ORP), in order to optimize the destructive output
signal at port I.
The control signal, a CW light beam lasing at 1546.12 nm,
is launched into port B of the MZI-SOA in a co-propagating
direction with the data signals.
Finally, the control signal is recovered at port I, using a filter
with a 40 GHz bandwidth (X-tract Net Test). We use two
measurement instruments to analyze the output signal: an
oscilloscope (Agilent 86100A), connected through a PIN
photodiode (HP-11982A), and an optical complex spectrum
analyzer (APEX AP 2441A) to gather phase and power
information of the output signal, for time domain
III. EXPERIMENTAL RESULTS AND DISCUSSION
In order to validate the feasibility of MZI-SOA based phase
modulators, experiments at 2.5 and 10 Gb/s were carried out
with the same experimental setup.
An average extinction ratio (ER) of 11.3 dB and a mean
power of 2.5 dBm were measured for data signals launched
into ports A and D of the MZI-SOA.
The bias current (Isoa) of both SOAs were varied
simultaneously, from 150 to 300 mA for 2.5 Gb/s and from
150 to 400 mA for 10 Gb/s. For each bias current, the mean
power of the control signal (P
) was increased from -6 to
The voltage applied to the phase shifters was adjusted in
order to maximize the destructive interference at output port I.
A. Phase modulator experiments at 2.5 Gb/s
Due to limitations imposed by the OCSA, the length’s
sequence at 2.5 Gb/s was restricted to 4 bits .
Fig. 3(c) shows the bit pattern launched at the
interferometric ports (A and D) of the MZI-SOA. Fig. 3(d)
illustrate the output signal at port I for bias current at 250 mA
and control signal at -4 dBm. The phase shift related to
different logic levels is well defined and the output power
signal is inverted when compared with the input data signals.
In Fig. 3(a), phase span is plotted as a function of P
several bias currents. The results show that they increase as
the bias current is raised. Mean values vary between 35º and
It can be observed in Fig. 3(b) that the mean output power
is also proportional to the increase of Isoa and P
SOAs gain is not saturated.
B. Phase modulator experiments at 10 Gb/s
The proposed optical phase modulator was also evaluated
at 10 Gb/s. The tests were performed using data sequences
with 16 bits .
Fig. 2. Experimental setup of the optical phase modulator. DFB: distributed-feedback laser; PC: polarization controller; MZM: Mach-Zehnder modulator;
EDFA: Erbium Doped Fiber Amplifier; ODL: optical delay line; CW: continuous wave laser; VOA: variable optical attenuator; SOA: semiconductor optical
amplifier; PS1, 2: phase shifters; OCSA: optical complex spectrum analyzer; PG: pattern generator; O/E: PIN photodiode. Solid lines represent fiber-optic paths
and dashed lines indicate electronic connections.