With the advent of IoT and the seamless advent of the ubiquitous network using the 5G technologies, the ability to provide a continuous service is the requirement of every Internet Service Provider (ISP). The IoT presents a network of humongous network size, troubleshooting and maintaining this network is a challenge, in light of this an automated system to do network diagnostics and predictive self-healing is the need of the current era. The proposed system introduces an automated network diagnostics and self-healing technique for 5G environment using predictive analysis. The performance parameters of the device or network are considered to collect the data and analyze the possible anomalies. When the performance parameters are deviated from the normal ranges, the problems occurred in the network are diagnosed in a productive way and the predictive analysis is done. The time series analysis helps to predict the performance of the network in various time intervals. The proposed technique has been implemented in the live network environment provided by one of the leading ISP and the performance analysis has proven that the predictive analysis and network diagnostics improves the network performance with self-healing in 5G networks.

随着物联网的出现以及使用5G技术无缝普及到无处不在的网络，提供连续服务的能力是每个Internet服务提供商（ISP）的要求。物联网提出了一个庞大的网络规模，对网络进行故障排除和维护是一个挑战，鉴于这种自动化系统需要进行网络诊断和预测性自我修复是当前时代的需求。拟议的系统采用预测分析为5G环境引入了自动网络诊断和自我修复技术。考虑设备或网络的性能参数以收集数据并分析可能的异常情况。当性能参数偏离正常范围时，对网络中发生的问题进行有效诊断，并进行预测分析。时间序列分析有助于预测各种时间间隔内的网络性能。这项提议的技术已在领先的ISP之一提供的实时网络环境中实施，性能分析已证明，预测分析和网络诊断可通过5G网络中的自我修复功能提高网络性能。