Root Cause Analysis of Communication Network Based on Deep Fuzzy Neural Network

In the realm of communication networks, root cause analysis plays a vital role in maintaining efficient and reliable operation. However, existing root cause analysis methods face limitations and drawbacks, including their inability to handle complex data and disturbances, as well as inaccuracies in identifying root causes. To this end, the paper presents the Deep Fuzzy Neural Network approach as an innovative solution. Integrating the strengths of deep learning and fuzzy logic inference, where the deep learning technique utilizes the parallel computing fusion of convolutional neural network and long short-term memory to extract the spatial-temporal features from sophisticated fault data of communication network. By leveraging this parallel computing fusion module, the proposed framework effectively addresses the flaws of traditional root cause analysis methods. Furthermore, the incorporation of fuzzy logic enables the proposed model to manage disturbances such as uncertainty and noise inherent in the data, ensuring robust performance. Experimental results also demonstrate our proposed deep fuzzy neural network approach is an effective method for network root cause analysis in overcoming limitations inherent in existing methods and providing superior accuracy and resilience.