We experimentally demonstrate a fiber-optic programmable optical pulse shaper based on time-domain binary phase-only linear filtering, which is capable of switching picosecond pulse shapes at unprecedented sub-GHz rates by simply updating the binary signal driving an electro-optic phasemodulator (EO-PM). The required binary phase-filtering functions are computed using a genetic algorithm (GA). One limitation of the binary phase-filtering approach is the inherent symmetry of the output temporal shapes. To generate fully arbitrary time-domain intensity profiles (including asymmetric temporal waveforms) we must employ a multi-level phasefiltering function. However, the size of the solution-space and complexity of the computation multiplies to manifolds as the number of levels in the phase-filtering function increases. Here we numerically demonstrate a simple strategy, by combining the Gerchberg-Saxton algorithm (GSA) and GA, for the fast computation of multi-level phase-filtering functions. The performance of this approach is numerically proven by generating different asymmetric pulse shapes of practical interest, assuming experimentally feasible design parameters.
Fiber-Based Programmable Picosecond Optical Pulse Shaper
MALACARNE, Antonio;FRESI, Francesco;
2010-01-01
Abstract
We experimentally demonstrate a fiber-optic programmable optical pulse shaper based on time-domain binary phase-only linear filtering, which is capable of switching picosecond pulse shapes at unprecedented sub-GHz rates by simply updating the binary signal driving an electro-optic phasemodulator (EO-PM). The required binary phase-filtering functions are computed using a genetic algorithm (GA). One limitation of the binary phase-filtering approach is the inherent symmetry of the output temporal shapes. To generate fully arbitrary time-domain intensity profiles (including asymmetric temporal waveforms) we must employ a multi-level phasefiltering function. However, the size of the solution-space and complexity of the computation multiplies to manifolds as the number of levels in the phase-filtering function increases. Here we numerically demonstrate a simple strategy, by combining the Gerchberg-Saxton algorithm (GSA) and GA, for the fast computation of multi-level phase-filtering functions. The performance of this approach is numerically proven by generating different asymmetric pulse shapes of practical interest, assuming experimentally feasible design parameters.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.