Improved Q2 shooting algorithm using answer ranges denition to design doped optical ber laser Maryam Karimi, Amir Hossein Farahbod Q1 La Q3 ser and Optics Research School, AEOI, P.O. Box 14399-51113, Tehran, Iran article info Article history: Received 14 September 2013 Received in revised form 1 March 2014 Accepted 4 March 2014 Keywords: Optical ber Laser Shooting method Optimum length abstract Shooting method is a simple and powerful method to solve boundary value problems. By using shooting method and appropriate denition of comparative function with a proper guess for the initial value of pump and signal the computation process has short time and quick converge to correct answers. In this paper by denition of the answer ranges and proper comparative function, the requirement of proper initial guesses for forward signal and backward pump to solve rate equation is resolved. By putting initial and end of answer ranges as two set guesses, the signal and pump initial values were achieved fast and easily using the second method. In other words using answer ranges, the shooting algorithm was equipped with a process of response screening. This computational procedure, without requirement of experimental lookout can be used for ber lasers and ampliers. We also show that the former optimum length denition in strongly pumped power lasers cannot be used in lasers with conventional pump power and must be replaced with suitable one. Forward, backward and bi-directional pump scheme have been investigated using the proposed calculation procedure. An effect of background loss on optimum ber laser length is also veried and the effects of input and output mirrors reectivity, dopant concentration and ber length in laser output power is investigated. & 2014 Published by Elsevier B.V. 1. Introduction Doped optical bers have a potential to be used as a laser components and have many applications such as communications [1], industrial processing [2], sensors [3], medical applications [4,5] and so on. Erbium doped bers (EDFs) are a suitable source for L and C band communication systems [1]. These ampliers have high gain, wide optical bandwidth, high output saturation, low noise gure, polarization independent gain, no crosstalk and low insertion loss [6]. The Nd and Yb doped ber lasers (NDFL, YDFL) were developed for use in high power and double clad ber laser in industry [7]. Fiber lasers have many advantages over bulk glass lasers, such as low cost, broad gain bandwidth, high efciency, good beam quality and small quantum defects [8]. Modeling (simulation and analytical results) is important which investigates static and dynamic characteristics of lasers and ampliers for parameters of practical systems optimization [9]. The dynamic analysis of energy level populations in laser media are often modeled using a system of rate equations [10]. Several authors have studied high power ber lasers (HPFLs) theoretically and got many signicant achievements [5,7,11]. In strongly pumped ber lasers, the population density of the upper lasing level N 2 is assumed to be much smaller than dopant concentration N 0 , so the analytical simulations can lead to the right results in such highly pumped ber lasers [710,12,13]. But, analytical expressions had a large relative error comparing with the exact numerical simulation for conventional pump power and high reectivity of the output mirror [11]. For the forward pumped ampliers, the rate equations can simply have a numerical solu- tion with RungeKutta or other methods [14,15]. These kinds of problems have an initial value for pump and signal, and classied as initial value problems (IVPs). The conventional ber lasers and bi-directional pump on doped ampliers are classied as two point boundary value problems (BVPs) [16,17], which are much more difcult to be solved than the IVPs. There are several numerical algorithms for solving BVPs, such as shooting method (S-method), relaxation method (R-method) and nite difference method (FD-method) [17], or use of genetic algorithm (GA) [18]. Using the FD method is a time consuming technique [19]. Both S and R-methods reduce a BVP to the initial value problem. The accuracy of the R-method has high accuracy for small values of N 0 , and L, where N 0 and L are dopant concentration and ber laser length respectively [20]. However, the S-method is commonly used for solving ber laser problems such as spatial counter propagation and bi-directional multi pump Raman ber laser. The S-method has faster computational speed than GA [18]. In the shooting method, the process starts with 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 Contents lists available at ScienceDirect journal homepage: www.elsevier.com/locate/optcom Optics Communications http://dx.doi.org/10.1016/j.optcom.2014.03.013 0030-4018/& 2014 Published by Elsevier B.V. E-mail address: mykarimi@aeoi.org.ir (M. Karimi). Please cite this article as: M. Karimi, A.H. Farahbod, Optics Communications (2014), http://dx.doi.org/10.1016/j.optcom.2014.03.013i Optics Communications (∎∎∎∎) ∎∎∎∎∎∎