Scholarlink

<p>Space division multiplexing with multicore or multimode fiber is one potential future-generation multiplexing technique for optical fiber communications. A spatially separated channel transmits multiple independent signals at the same wavelength using space division multiplexing (SDM). An artificial neural network model for performance prediction was trained using the synthetic dataset generated by simulations. The capacity of ANN to categorize SDM system performance based on input parameters has been demonstrated by its 95% validation accuracy for one dataset, and 85% validation accuracy for another. The efficiency of the model was further validated by metrics including precision-recall curves and Receiver Operating Characteristic Area under the Curve (ROC AUC) scores. The paper emphasizes how ANNs may optimize system design and performance evaluation, as well as SDM’s ability to handle ultra-large transmission systems.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Artificial neural networks as predictive tools for space division multiplexing system performance

  • 期刊论文

  • 发表日期：

    2025年2月20日

查看全文
  • Hardeep Kaur,
  • Ramandeep Kaur,
  • Rajandeep Singh,
  • Gurpreet Kaur

摘要

Space division multiplexing with multicore or multimode fiber is one potential future-generation multiplexing technique for optical fiber communications. A spatially separated channel transmits multiple independent signals at the same wavelength using space division multiplexing (SDM). An artificial neural network model for performance prediction was trained using the synthetic dataset generated by simulations. The capacity of ANN to categorize SDM system performance based on input parameters has been demonstrated by its 95% validation accuracy for one dataset, and 85% validation accuracy for another. The efficiency of the model was further validated by metrics including precision-recall curves and Receiver Operating Characteristic Area under the Curve (ROC AUC) scores. The paper emphasizes how ANNs may optimize system design and performance evaluation, as well as SDM’s ability to handle ultra-large transmission systems.