An Efficient Specific Emitter Identification Method Based on Complex-Valued Neural Networks and Network Compression

TLDR

Specific emitter identification (SEI) uses radio‑frequency fingerprinting to uniquely tag emitters, and recent deep‑learning approaches, especially complex‑valued neural networks, improve accuracy but suffer from high model complexity that hinders deployment in IoT scenarios. This work proposes an efficient SEI approach that combines a complex‑valued neural network with network compression to enhance performance while reducing model size. The method employs a CVNN to directly process complex baseband signals for high identification accuracy, then applies compression techniques to shrink the model to 10–30 % of its original size without sacrificing performance. Simulations show the compressed CVNN achieves nearly 100 % accuracy at high SNR, outperforms existing DL‑based SEI methods in both accuracy and convergence speed, and demonstrates that compression causes negligible performance loss.

Abstract

Specific emitter identification (SEI) is a promising technology to discriminate the individual emitter and enhance the security of various wireless communication systems. SEI is generally based on radio frequency fingerprinting (RFF) originated from the imperfection of emitter's hardware, which is difficult to forge. SEI is generally modeled as a classification task and deep learning (DL), which exhibits powerful classification capability, has been introduced into SEI for better identification performance. In the recent years, a novel DL model, named as complex-valued neural network (CVNN), has been applied into SEI methods for directly processing complex baseband signal and improving identification performance, but it also brings high model complexity and large model size, which is not conducive to the deployment of SEI, especially in Internet-of-things (IoT) scenarios. Thus, we propose an efficient SEI method based on CVNN and network compression, and the former is for performance improvement, while the latter is to reduce model complexity and size with ensuring satisfactory identification performance. Simulation results demonstrated that our proposed CVNN-based SEI method is superior to the existing DL-based methods in both identification performance and convergence speed, and the identification accuracy of CVNN can reach up to nearly 100% at high signal-to-noise ratios (SNRs). In addition, SlimCVNN just has 10% ~ 30% model sizes of the basic CVNN, and its computing complexity has different degrees of decline at different SNRs; there is almost no performance gap between SlimCVNN and CVNN. These results demonstrated the feasibility and potential of CVNN and model compression.