Signal Transmission Stability Optimization and Network Security Protection Mechanism of Power Communication Network Based on ResNeXt

Abstract

In the process of signal transmission in power communication network, the interference characteristics in the transmission environment are complex and changeable, and it is difficult for the existing optimization and security mechanisms to form effective coordination. To this end, this paper adopts a multi-task learning framework based on the ResNeXt model, combined with signal optimization and anomaly detection tasks, to achieve the coordinated optimization of signal transmission and network security protection. Through the group convolution and multi-path feature extraction of ResNeXt, the signal interference characteristics are analyzed, and the transmission parameter optimization strategy is dynamically generated; at the same time, the forged signal and data tampering behavior are detected, and the coordination ability between distributed nodes is improved based on the federated learning mechanism. Experimental results show that the multi-task learning framework built based on ResNeXt in this paper improves channel utilization by 14.2% and reduces the signal bit error rate (BER) by 3% compared with the original basic model; the detection accuracy of forged signals and data tampering attacks reaches 96.8% and 96.4% respectively; the response time is 16.5ms and 16.7ms respectively. The method proposed in this paper not only optimizes the transmission performance but also effectively improves the network’s security protection capability, providing an innovative solution for an efficient and secure power communication network.