With the growth of the Internet of Things (IoT) comes a flood of malicious traffic in the IoT, intensifying the challenges of network security. Traditional models operate with independent layers, limiting their effectiveness in addressing these challenges. To address this issue, we propose a cross-layer cooperative Feature Subset-Based Malicious Traffic Detection (FSMMTD) model for detecting malicious traffic. Our approach begins by applying an enhanced random forest method to adaptively filter and retain highly discriminative first-layer features. These processed features are then input into an improved state-space model that integrates the strengths of recurrent neural networks (RNNs) and transformers, enabling superior processing of complex patterns and global information. This integration allows the FSMMTD model to enhance its capability in identifying intricate data relationships and capturing comprehensive contextual insights. The FSMMTD model monitors IoT data flows in real-time, efficiently detecting anomalies and enabling rapid response to potential intrusions. We validate our approach using the publicly available ToN_IoT dataset for IoT traffic analysis. Experimental results demonstrate that our method achieves superior performance with an accuracy of 98.37%, precision of 96.28%, recall of 95.36%, and F1-score of 96.79%. These metrics indicate that the FSMMTD model outperforms existing methods in detecting malicious traffic, showcasing its effectiveness and reliability in enhancing IoT network security.

To address the problem of temporal feature extraction and recognition of high-resolution range profiles (HRRP) of midcourse ballistic targets, a recognition method based on bidirectional temporal convolutional networks with additive fusion (AF-BiTCN) was proposed, which could make full use of the bidirectional temporal information of HRRP of midcourse ballistic targets and further improve the recognition performance. Firstly, the HRRP data was processed into a bidirectional sequence by the bidirectional sliding window algorithm. Then, the BiTCN was constructed to extract bidirectional deep temporal features of HRRP in each layer, and the bidirectional features were fused by an additive strategy. Finally, more robust fusion features were utilized to recognize ballistic targets, and the Adam algorithm was used to optimize the convergence speed and stability of AF-BiTCN. The experimental results show that the proposed HRRP recognition method of midcourse ballistic targets based on AF-BiTCN in this paper has higher accuracy and faster recognition speed compared with six methods such as stack long short-term memory (SLSTM), stack gate recurrent unit (SGRU) and so on, and it achieves an accuracy of 96.60% on the test set. Moreover, the proposed method indicates better robustness on noise datasets.