A novel signal representation in SEI: Manifold.

• We propose a novel SEI framework with a separate representation module to balance the trade-off between comprehensiveness and efficiency. Such a representation module can simultaneously describe comprehensive system information and extract distinguishable fingerprint features. Hence, the real system can store representations instead of raw data. Intuitively, this process benefits the database updates and reduces storage space requirements. Also, the complexity of designing fingerprint features is reduced, especially when new items are added to the database. • Within the proposed framework, we introduce manifolds as signal representations for the SEI task. To the best of our knowledge, representing emitters with manifolds has not been proposed yet. On the one hand, manifolds highlight the emitter's nonlinear dynamic characteristics brought by hardware imperfections. On the other hand, manifolds conclude the emitter system states and their evolution rules, implying comprehensive system information. To provide theoretical support, we prove the one-to-one correspondence between manifolds and emitter individuals, which is concluded as two lemmas. Besides, experiments with visualization further confirm the argument. • For describing manifold uniqueness, we design multi-level manifold features, containing intrinsic dimensions, topological features, conformal features, and Riemannian metric features. Those features, describing manifolds from multiple views, can act as fingerprint features. Considering the power amplifier nonlinearity and the local oscillator spurious radiation as an example, we analyze the divergence in manifold features for different hardware parameters. The corresponding experiments verify the effectiveness of manifold features. • We propose a novel manifold-based SEI method by selecting manifolds and their multi-level features as representation and fingerprint features, respectively. First, we utilize PSR and manifold learning methods to reconstruct signal manifolds. Then, multi-layer manifold features are extracted and input into the Adaboost classifier. Extensive experiments illustrate that the proposed method achieves high accuracy, efficiency, and adaptability. [Display omitted] Specific emitter identification (SEI), as an important problem in situational awareness, identifies emitters via unique characteristics. However, current SEI methods mostly suffer from appropriately setting the trade-off between comprehensiveness and efficiency when extracting fingerprint features. To address the issue, this paper provides a novel SEI framework with a separate representation module. Within the novel framework, manifolds are proposed to be signal representations and multi-level manifold features are extracted as fingerprint features. We first build the SEI model from the nonlinear dynamic perspective, where the SEI process identifies the nonlinear systems via a measurement sequence. Then, we demonstrate that manifolds can represent emitters equivalently and prove the one-to-one correspondence between manifolds and emitter individuals. Hence, manifolds can highlight unique nonlinear dynamic characteristics and simultaneously describe comprehensive system working processes. The coordinate delayed technique and manifold learning methods are employed to reconstruct the phase space and manifold, respectively. For accomplishing the identification task, multi-level manifold features, comprising intrinsic dimension, topological features, conformal features, and Riemannian metric features, are extracted from the reconstructed manifolds and input to an ensemble learning scheme, named Adaboost. Extensive simulation and real-world experiments agree with our analytical conclusions and confirm the proposed method's efficiency. The results also demonstrate that the proposed method achieves a high recognition accuracy, outstanding adaptability, and strong robustness. [ABSTRACT FROM AUTHOR]