Tag
E-SpecFormer introduces LiTAN, a Softmax- and LayerNorm-free attention mechanism, enabling efficient end-to-end automatic modulation and covert channel recognition on edge devices with under 10k parameters and FPGA/CPU co-execution speeds of 92μs per frame.
This paper proposes Dualformer, a dual-channel neural network architecture based on transformers, designed for efficient feature extraction from complex-valued signals in blind communication analysis tasks such as automatic modulation recognition, signal scheme recognition, and signal structure parsing. Extensive experiments show consistent performance improvements over existing methods.