Final Report Summary - PETAL (Polarization condEnsation for Telecom AppLications)
In nowadays optical networks, the main part of signal processing is performed in the electronic domain, that is to say, after an opto-electronic conversion, rather than in the optical physical layer. As the amount of transmitted information is unrelentingly increasing, the current trend is to move towards transparent networks so as to perform the signal processing in the optical domain and confine opto-electronic conversion to network boundaries. To reach this ambitious target, it now becomes mandatory to master all the parameters that characterize the transmitted light, especially the state-of-polarization. In fact, in many fields of photonics, especially in fiber-based systems, the light state-of-polarization remains one of the most elusive and underexploited parameter that is still challenging to all-optically control. Indeed, despite the progress of manufacturing process of standard optical fibers, the residual birefringence combined with environmental variations and local mechanical stress such as bending, squeezing, vibrations or temperature fluctuations make the polarization of light completely random and unpredictable after a few hundreds of meters of propagation. Consequently and despite the high efficiency of digital signal processing and sophisticated algorithms implemented in current coherent transmission systems, all-optical control of light polarization still remains an open issue. In this context, the purpose of the PETAL project was to explore a novel approach to polarization control issue and to transform this parameter into a fully exploited asset. While current opto-electronic technologies are principally based on complex active-feedback loop control and algorithms, the breakthrough idea of the PETAL project is to explore a new type of nonlinear effect based on the unexpected ability of light to self-organize its own state of polarization in optical fibers as well as to exploit this phenomenon in order to develop new optical functionalities for telecom applications. The principle of operation is based on a counter-propagating four-wave mixing process occurring within a device called Omnipolarizer. It basically consists in a nonlinear Kerr medium, here an optical fiber, in which an initial forward signal nonlinearly interacts through a cross-polarization interaction with its own backward replica generated and amplified at fiber end by means of a reflective element.
The first results of the PETAL project was to provide a theoretical model of this counter-propagating effect in order to fully understand and master the polarization condensation phenomenon. In particular, we have derived some design rules which enable to predict the different dynamics generated by the optical feedback within the Omnipolarizer. Due to the particular boundary conditions imposed by the reflective element inserted at fiber end, we have theoretically and experimentally identified 3 working regimes which depend on the power ratio between both counter-propagating beams. For nearly equal powers, the Omnipolarizer can operate as an ideal polarizer and thus enables to repolarize any input signal; experimental demonstration has been successfully achieved on a 40-Gbit/s telecom signal and represents the first experimental proof-of-principle of the polarization condensation phenomenon. Then for a ratio of powers below unity, the device becomes bistable and can operate as a digital beam splitter. We have exploited this bistability to develop an optical memory as well as an all-optical routing device for optical packets in a 10-Gbit/s experiment. Finally, when a strong power imbalance is applied between the forward and backward signals, we have also observed a chaotic behaviour within the Omnipolarizer which has been exploited to demonstrate an all-optical polarization scrambler as well as a random bit generator. Finally, all these regimes of operation have been exploited in a combine experiment to provide a novel approach of securing or spying 10-Gbit/s optical transmissions.
The first results of the PETAL project was to provide a theoretical model of this counter-propagating effect in order to fully understand and master the polarization condensation phenomenon. In particular, we have derived some design rules which enable to predict the different dynamics generated by the optical feedback within the Omnipolarizer. Due to the particular boundary conditions imposed by the reflective element inserted at fiber end, we have theoretically and experimentally identified 3 working regimes which depend on the power ratio between both counter-propagating beams. For nearly equal powers, the Omnipolarizer can operate as an ideal polarizer and thus enables to repolarize any input signal; experimental demonstration has been successfully achieved on a 40-Gbit/s telecom signal and represents the first experimental proof-of-principle of the polarization condensation phenomenon. Then for a ratio of powers below unity, the device becomes bistable and can operate as a digital beam splitter. We have exploited this bistability to develop an optical memory as well as an all-optical routing device for optical packets in a 10-Gbit/s experiment. Finally, when a strong power imbalance is applied between the forward and backward signals, we have also observed a chaotic behaviour within the Omnipolarizer which has been exploited to demonstrate an all-optical polarization scrambler as well as a random bit generator. Finally, all these regimes of operation have been exploited in a combine experiment to provide a novel approach of securing or spying 10-Gbit/s optical transmissions.