ICTON 2002 184 Tu.P.1  The work reported in this paper was supported by the Slovenian Ministry of Education, Science and Sport. 0-7803-7375-8/02/$17.00 2002 IEEE Fiber Nonlinear coefficient measurement scheme based on Four- Wave Mixing method with externally modulated laser source Bostjan Batagelj, Matjaz Vidmar Laboratory of Optical Communications, Faculty of Electrical Engineering, University of Ljubljana, Trzaska 25, 1000 Ljubljana, Slovenia Tel: (386) 14768423, Fax: (386) 14768424, e-mail: bostjan.batagelj@fe.uni-lj.si ABSTRACT In this paper, we present measurement scheme for determining the optical fiber non-linear coefficient γ, using externally modulated continuous wave laser source. The measurement technique is based on Four-Wave Mixing. The measurement source which uses one laser and one optical erbium-doped fiber amplifier only is described precisely. The proposed simple measurement scheme is polarization independent, enables high sensitivity and accuracy, and is applicable to all currently available types of fiber. Keywords: Optical Fiber Communications, Fiber Measurements, Non-linear Coefficient, Four-Wave Mixing. 1. INTRODUCTION As new fibers are being designed for non-linear optics applications, a consistent approach to measurement of optical non-linearity is required for systematic materials engineering and nonlinear devices optimization. Simultaneously, as overall transmitted data estimated on WDM or soliton systems increase, the effects of fiber non-linearity come to play an even more decisive role in the design and performance of modern optical communications links. For this reason it is important to be able to measure the non-linear coefficient of fibers of different compositions simply and accurately. Many schemes have been previously proposed to make measurement of non-linear coefficient (γ) [1]. Four- Wave mixing (FWM) is a reliable technique for determining the fiber γ. The well-known measuring setup [2]-[4] based on FWM is made up of two DFB lasers, where the wavelength spacing is adjusted by temperature tuning. It uses more than one optical erbium-doped fiber amplifier (EDFA) to raise the total optical power. One of the biggest disadvantages of this method is that the polarisation controllers must be adjusted all the measurement time in order to obtain maximum FWM efficiency. The determination of γ employing such FWM method has a limited measurement capability due to the polarization adjustment and insufficient amplification. We have devised and demonstrated a measurement scheme based on FWM method [5]-[6], which uses one externally modulated laser source only. The use of such source eliminates polarization dependence and making measurement scheme simple and highly sensitive, what leads to higher accuracy. The introduced measurement scheme is applicable to all currently available fiber types including Photonic Crystal Fibers. 2. THEORY OF THE FWM METHOD The FWM is a non-linear process induced by Kerr effect in optical fiber. It occurs when two or more wavelengths of light propagate together through an optical fiber. Supposed the equal signal input power, the ratio between the power of new frequencies generated through FWM at the end of the fiber and the signal output power is written as [7] ( 29 ( 29               ⋅ ∆ ⋅ - ⋅ + ∆ + ⋅ ⋅ ⋅ = - - 2  sin 1 4 1  . . ) 0 (  ) ( ) ( 2 2 2 2 2 2 L e e L P L P L P L L eff s s FWM , (1) where P s (0) is the signal input power, P s (L) is the signal output power and L eff is the fiber effective interaction length. The phase-matching condition is given by (29         - ⋅ ⋅ -         ⋅ ∆ + ⋅ ∆ ⋅ ⋅ = ∆ - eff eff L s p p L e P d dD c f D f c α γ λ λ λ λ π α 1 0 ) ( 2  2 2 2 , (2) where ∆f is frequency spacing, D(λ) is the fiber chromatic dispersion, dD/dλ is the dispersion slope, and λ p is the wavelength corresponding to the pump frequency. The second part of equation (2) corresponds to the intensity dependence of the phase matching [8]. In our case, the measured fiber should be relatively short, in order to make measurements in the flat part of FWM characteristic and to achieve phase matching, so that sine term of equation (1) could be neglected. However, a numerical algorithm based on the above described FWM is usually used to relate the input and output fields and to determine the γ.