Search papers, labs, and topics across Lattice.
This paper introduces DKDNet, a novel architecture for automatic modulation classification (AMC) that leverages dual knowledge from signal prior representations and data-driven features to enhance cross-domain generalization. By analyzing five signal representations, the authors identify in-phase/quadrature (IQ), amplitude-phase (AP), and autocorrelation function (ACF) as the most effective for guiding feature extraction. Experimental results on simulated and public datasets confirm that DKDNet outperforms existing methods by effectively integrating modulation-specific structures and achieving improved classification accuracy across varying domains.
Prior knowledge about signal representations can significantly boost the performance of automatic modulation classification in shifting communication environments.
The dynamics of communication environments induce significant distribution shifts across domains, challenging the generalization of deep learning-based automatic modulation classification (AMC) models. While existing UDA methods alleviate this problem by aligning source and target features, they give limited consideration to modulation-specific structures that remain informative across domain conditions. In this paper, we consider signal prior knowledge, grounded in communication protocols and physical principles, as a potential way to enhance cross-domain representation learning. Given that different priors may vary in modulation discriminability, domain stability, and complementarity, this paper first analyzes five commonly adopted signal representations that instantiate different signal priors. From them, in-phase/quadrature (IQ), amplitude--phase (AP), and autocorrelation function (ACF) are selected as compact prior-guided inputs. Based on that, a dual knowledge and data-driven network (DKDNet) is proposed for cross-domain AMC. The multi-representation feature encoder (MRFE) and dynamic lightweight fusion unit (DLFU) are designed to achieve unified representation learning and adaptive feature fusion, and the resulting fused features are optimized with modulation classification and adversarial domain alignment objectives. Experiments on both simulated and public datasets validate the rationality of the prior selection and demonstrate the superiority of the proposed method.
