Multi-Sensor Fusion-Based Anomaly Traffic Monitoring and Optimization Method for Communication in 6G-Enabled Information Warehouse Systems

In complex 6G-enabled industrial communication networks, the diversity of thresholds for anomaly traffic monitoring arises due to the extensive use of multiple sensors for external information collection. Current methods, which rely on single thresholds or anomaly traffic monitoring, suffer from low monitoring accuracy and poor timeliness. By leveraging the high speed and low latency of the 6G network, this paper introduces a multi-sensor information fusion technique to design an anomaly traffic monitoring method for industrial communication networks. First, the multi-sensor fusion technique is applied to improve the estimation accuracy of local filters by using predicted values of lost observations as compensation. The cross-covariance matrix between any two estimation errors is provided, and this filtering method is used to process communication network data. Then, based on the matrix model's plane and 2D coordinate attributes, traffic features are sorted according to certain rules. Since the feature experience library can identify abnormal traffic in communication networks, an anomaly traffic matching model is established by combining location features to monitor abnormal traffic in the network and determine its source. Finally, experimental results show that the new monitoring method has higher accuracy and ensures good timeliness, enabling rapid feedback on anomaly traffic monitoring.