All-fibered Optical Sources for Telecommunication Applications – SO FAST

Most of the telecommunication operators now offer number and number of services at higher and higher bit rates, involving the deployment of all-optical techniques until the user home (FFTH: Fiber To The Home). This strong increase of the global bit rate in the access networks (1 Gbit/s available to the customer in Japan) will increase the total flow on networks. Some forecasts show an increase of the total traffic for transoceanic transmissions up to 100 Tbit/s in 2022, compared with 4 Tbit/s in 2008. Therefore, this trend will necessary lead to a strong demand for low cost tunable optical sources at high repetition rates, in particular for optical time division multiplexing applications, only solution to currently overcome the electronic bandwidth limitations. This growth of the global traffic will also require new all-optical high-speed components, such as saturable absorbers, optical filters, chromatic dispersion compensators, multiplexers, or all-optical regenerators. These new devices will then require optical sources of high repetition rates and/or short pulses in order to validate their performances or to achieve resynchronization operations. The SO FAST project is an experimental development program and aims to design and study low cost all-fibered optical sources at high repetition rates of 10 to 100 GHz and at telecom wavelengths around 1.55 µm. This type of sources is based on the nonlinear compression of a sinusoidal beat-signal through the multiple four wave mixing process taking place in optical fibers. They have the advantage of being tunable in wavelength, pulse width as well as repetition rate and more importantly, without mode-locking process. The generation of high repetition pulse train (20GHz-1THz) through the nonlinear compression of a sinusoidal beating is a technique mastered by ICB This technique provides good quality pulses at high repetition rates around 1550 nm, are Fourier transform limited for a duty cycle (time period on pulse width ratio) ranging from 2 to 16. However, the generation of the initial sinusoidal beat-signal requires the use of two independent laser diodes. Consequently, without an enslavement system of the repetition rate, this configuration currently presents any synchronization or clock recovery. Therefore, the implementation of such a device in industrial or Telecom applications is currently vanished. Another difficulty is also provided by the backscattering Brillouin phenomenon occurring within the compression fiber. Indeed, a significant part of the injected light is backscattered, reducing the effectiveness of the nonlinear compression. The aim of this project is thus to develop all-fibered optical sources at high repetition rates based on this nonlinear compression technique by overcoming these two major difficulties. Our final goal and success criterion will be to provide a commercial prototype at the end of the 24-month SO FAST program.