Interview with Akito Chiba

The first four-Mach-Zehnder type optical modulator to be monolithically integrated on LiNbO3 (LN) for optical 16-level quadrature amplitude modulation (QAM) has been demonstrated by collaborating researchers in Japan. Dr Akito Chiba, a researcher at the National Institute of Information and Communications Technology, tells us more about their achievement.

Dr Akito Chiba, a researcher at the National Institute of Information and Communications Technology, tells us more about their achievement.

Why is advanced optical modulation of such current interest?

Advanced optical modulation formats enable us to enhance the data transportation rate without an increase in modulation speed, by superposing data on several degrees of freedom of a lightwave, such as amplitude and phase, in a multilevel signal. In current optical communication, the main technical limitation in modulation speed is the bandwidth of components, such as a modulator and a detector.  Wavelength-division multiplexing is employed for the transmission of large amounts of data; however, the wavelength range is limited by the gain bandwidth of the optical fibre amplifiers, the optical loss and the dispersion of optical fibre, so a modulation format with a high data density in wavelength is useful to achieve a huge capacity of optical signal transmission.  Some of the advanced modulation formats, such as QAM, have already been employed in wireless communication.

Why is the work in your Electronics Letters paper an important step?

In this Letter we monolithically integrate four structures for electro-optical modulation on an LN substrate and show that the optical device can generate a 16-level optical QAM signal with a symbol rate of more than 10 Gbaud.  Previous devices were composed of two kinds of materials: LN and SiO2, so a complicated fabrication procedure was required.  By using a sophisticated configuration of optical waveguides and electrodes, we achieved the cost-effective fabrication of a compact monolithic optical modulator with a broad modulation bandwidth.  

What other highly integrated LiNbO3 optical devices could be fabricated with this technique?

One example is a single quad Mach-Zehnder in-phase and quadrature modulator, where two pairs of cascaded Mach-Zehnder structures are embedded in an LN substrate.  By using this optical device, any parasitic optical phase drift at data transition can be highly suppressed so that 80-Gb/s optical minimum-shift keying (MSK) is successfully generated.  This bit rate is the highest demonstrated for the optical MSK format and such high bit-rate optical signal generation would be difficult using a combination of several optical components.  
Another possibility is a polarisation-insensitive optical switch integrated on LN by adapting a polarisation diversity configuration. This has already been fabricated using a hybrid-integration technique.  Polarisation dependence is usually a weak point of LN, but if it became possible to fabricate a rapid-response optical switch with low polarisation monolithically, dependence can be obtained.  

What’s next for your group and your collaborators?

The next step is to show the feasibility of the device for generation of an optical 16-level QAM signal.  For example, an increase of the Baud rate of an optical 16-level QAM signal is one challenge.  Based on an evaluation result of the electro-optical bandwidth, the device reported in the Letter has the potential to generate the signal with a much higher rate than the 10 GBaud we reported.  It will be interesting to combine the technology with WDM to achieve huge-capacity signal transmission.  To develop the device, further dense monolithic integration techniques would be desired for increasing the number of optical elements, which could lead to the realisation of a more complicated modulation format, such as optical 64 QAM.  

Where do you see this field heading and what are the future challenges?

In future optical communication systems spectrally-efficient optical modulation is indispensable for the enhancement of transmission capacity.  Most previous demonstrations are based on conventional optical components with the help of a complicated electronic circuit for D/A conversion whose load becomes so much for high-speed operation. On the other hand, optical D/A conversion based on an integrated waveguide device has no speed limitation so that it would be useful to boost operation speed without increasing load on the electronics driving the optical devices. Integrated optical modulation devices open the door for optical communication being close to the Shannon limit. These experimental investigations have just started.  In addition to an increase of the modulation format level, a combination of the modulation format with the orthogonal frequency-division multiplexing or polarisation multiplexing would be challenging functions of a single optical device. 

The Letter presenting the results on which this interview is based can be found on the IET Digital Library