Specific Emitter Identification Based on Multi-Domain Feature Fusion and Integrated Learning

Abstract

Specific Emitter Identification (SEI) is a key research problem in the field of information countermeasures. It is one of the key technologies required to be solved urgently in the target reconnaissance system. It has the ability to distinguish between different individual radiation sources according to the varying individual characteristics of the emitter hardware within the transmitted signals. In response to the lack of scarcity among labeled samples in specific emitter identification, this paper proposes a method combining multi-domain feature fusion and integrated learning (MDFFIL). First, the received signal is preprocessed to obtain segmented time domain signal samples. Then, the signal is converted to time–frequency distribution using wavelet transform. Afterwards, an integrated learning two-stage recognition classification method is designed to extract data features of 1D time domain signals and 2D time–frequency distribution signals using the symmetry network structures of CVResNet and ResNet. Finally, fused features are fed into the complex-valued residual network classifier to obtain the final classification results. We demonstrate through the analysis results of the measured data that the proposed method has a higher accuracy as compared with the classical feature extraction method, and that this can improve the identification of communication radiation sources with fewer labeled samples.

An investigation into the impacts of deep learning‐based re‐sampling on specific emitter identification performance

“…A key motivating factor for DL‐based SEI is eliminating feature‐engineering or domain‐specific, expert knowledge. DL‐based SEI has been well investigated by multiple efforts [6, 10–45]. The work in [10, 11] proposes semi‐supervised SEI approaches based on GANs.…”

Transferring Learned Behaviors between Similar and Different Radios

“…Automatic modulation classification (AMC) and specific emitter identification (SEI) are two of these SI tasks, which involve determining the modulation scheme and the identity of a received signal, respectively [ 4 , 5 ]. RFML techniques for these tasks have shown great success in classifying signals based on unique features that are consistently present in signals of particular classes [ 6 , 7 , 8 , 9 ].…”

RFFsNet-SEI: A multidimensional balanced-RFFs deep neural network framework for specific emitter identification

“…A deep SEI method based on complex-valued neural network was proposed in [16]. A multi-domain feature fusion SEI method via integrated learning (MDFFIL) was addressed in [17].…”