A BiLSTM-CNN based Multitask Learning Approach for Fiber Fault Diagnosis

TL;DR

This paper introduces a multitask learning model combining BiLSTM and CNN architectures to improve fiber fault diagnosis by simultaneously detecting, locating, and characterizing faults, surpassing traditional methods.

Contribution

It presents a novel BiLSTM-CNN based multitask learning framework specifically designed for fiber fault diagnosis, demonstrating superior performance over conventional techniques.

Findings

Outperforms traditional fiber fault diagnosis methods

Accurately detects and characterizes fiber faults

Efficient multitask learning approach

Abstract

A novel multitask learning approach based on stacked bidirectional long short-term memory (BiLSTM) networks and convolutional neural networks (CNN) for detecting, locating, characterizing, and identifying fiber faults is proposed. It outperforms conventionally employed techniques.