Abstract

The propagation near the zero-dispersion wavelength was proposed as a method to reduce the detrimental two-soliton interaction.¹ This effect is attributed to the decay of the two-soliton bound state into two fundamental solitons with their corresponding amplitudes due to third-order dispersion (TOD).² Those solitons propagate with different velocities leading to an increasing separation. In an earlier paper we have shown that an alternatively introduced phase difference π between pulses in soliton trains leads to the conservation of the pulse shapes, a walk-off between pulses and a decreasing of their velocity.³ Very recently, phase modulation was proposed to reduce two-soliton interaction in the presence of gain and gain dispersion, but without TOD.⁴

© 1994 IEEE