首页> 外文期刊>Journal of Lightwave Technology >Computing Group Velocities and Group-Velocity Dispersions of Optical Fibers Through Automatic Differentiation of Explicit Forms of Propagation Constants
【24h】

Computing Group Velocities and Group-Velocity Dispersions of Optical Fibers Through Automatic Differentiation of Explicit Forms of Propagation Constants

机译:通过自动分化显式形式的传播常数计算光纤的群速度和分组速度分散

获取原文
获取原文并翻译 | 示例
           

摘要

We present a method to compute differential coefficients of the first and second orders, for propagation constants of nondegenerate and degenerate modes, propagating in an optical fiber to calculate dispersion characteristics of group velocities (GV) and group-velocity dispersions (GVD). This method automatically differentiates explicit forms of eigenvalues extracted from determinants of a generalized eigenvalue problem. This problem is converted from a nonlinear eigenvalue problem derived from the Trefftz method (e.g., transfer matrix and multipole methods) by the Sakurai-Sugiura method. We compute the differential coefficients of the second order of degenerate modes without computing eigenvectors and their first order differential coefficients in the generalized eigenvalue problems. Therefore, in parametric optimization, we can compute the differential coefficients of all propagating modes without remodeling the waveguide cross-section, as analyzing the whole cross-section avoids the symmetric conditions derived from the numerical model. Computed results with the proposed method for step-index and holey fibers validate the method and its effectiveness. In particular, our results show that our method computes the GV and GVD more accurately than computation using numerical differentiation.
机译:我们提出了一种计算第一和第二订单的差分系数的方法,用于非应于和退化模式的传播常数,在光纤中传播以计算群速度(GV)和组速度分散体(GVD)的色散特性。该方法自动区分从广义特征值问题的决定因素提取的明确形式的特征值。通过樱桃-Sugiura方法从来自Trefftz方法(例如,转移矩阵和多极方法)导出的非线性特征值问题来转换该问题。我们计算退化模式的二阶的差分系数,而不计算特征向量及其在广义特征值问题中的第一阶差分系数。因此,在参数优化中,我们可以计算所有传播模式的差分系数而不重构波导横截面,因为分析整个横截面避免了从数值模型导出的对称条件。计算结果采用阶梯索引和多孔纤维的提出方法验证了该方法及其有效性。特别是,我们的结果表明,我们的方法比使用数值分化的计算更准确地计算GV和GVD。

著录项

相似文献

  • 外文文献
  • 中文文献
  • 专利
获取原文

客服邮箱:kefu@zhangqiaokeyan.com

京公网安备:11010802029741号 ICP备案号:京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有
  • 客服微信

  • 服务号