利用窗口特征的信号重构对比学习方法OA

With the rapid growth of signal data in industrial internet,the issue of missing labels has become increasingly prominent,making self-supervised learning a critical solution.To address the problems of coarse feature granularity,unstable representation,and weak transferability in existing contrastive learning methods for signal recognition tasks,a window-based signal reconstruction contrastive learning approach was proposed.The method divided feature maps into multiple fixed windows and introduces local similarity constraints to construct a fine-grained contrastive structure.It also incorporated a signal reconstruction module to enhance the stability and semantic consistency of feature representations.Furthermore,a loss function integrating reconstruction error was designed to improve the feature fitting capability to the original signals.In the standard recognition experiment on the ADS-B dataset,the method proposed in this paper achieves a Top 1 accuracy improvement of 28.32％compared to existing comparative methods;it also outperforms various comparative methods on the RML dataset.Furthermore,in cross-domain transfer experiments between pairs of the RML,ADS-B,and CSI datasets,the method proposed in this paper also achieves the best results,which verifies its strong transferability and generalization capability.