BCRNet-SNN: Body Channel Response-Aware Spiking Neural Network for User Recognition

• Published in: IEEE transactions on industrial informatics Vol. 21; no. 8; pp. 6017 - 6027
• Main Authors: Kang, Taewook, Shin, Chanwoo, Lee, Jongseok, Lee, Jae-Jin, Sim, Donggyu, Kim, Seong-Eun
• Format: Journal Article
• Language: English
• Published: Piscataway IEEE 01.08.2025; The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
• Subjects: Accuracy; Artificial intelligence; Artificial neural networks; Biometrics; Body pulse response; Chirp signals; convolutional neural network (CNN); Convolutional neural networks; Couplings; Datasets; Feature extraction; Frequency modulation; Industrial applications; knowledge distillation (KD); Neural networks; Parameters; Performance degradation; Performance evaluation; Reliability; spiking neural network (SNN); Spiking neural networks; Training; user recognition; Vectors
• ISSN: 1551-3203; 1941-0050
• DOI: 10.1109/TII.2025.3558309

Spiking neural networks (SNNs) have been recently highlighted as an attractive approach for implementing artificial intelligence models in resource-constrained edge devices for various industrial applications. These SNNs leverage low-power and wide dynamic range processing through biologically inspired event-driven operations in a massively parallel manner. In this respect, we propose BCRNet-SNN, an SNN model designed to utilize an electric body channel response (BCR) as a biometric feature for user recognition, where each element in the BCR dataset for 15 subjects is a 1-D vector comprising 380 feature points from the measured envelope by applying chirp signals to the body. The network parameters for implementing the posttrained BCRNet-SNN are inherited from the proposed convolutional neural network (CNN) that extensively extracts BCR-based biometric features (BCRNet), followed by applying knowledge distillation-based network lightening process on BCRNet. The performance evaluation results compared to ResNet18 and ResNet6 for the BCR dataset show that BCRNet achieves a greater than 2<inline-formula><tex-math notation="LaTeX">\%</tex-math></inline-formula> and 1.4<inline-formula><tex-math notation="LaTeX">\%</tex-math></inline-formula> improvement, respectively, in the average classification accuracy, while significantly reducing the number of network parameters to less than 1<inline-formula><tex-math notation="LaTeX">\%</tex-math></inline-formula>. The student network distilled from the teacher BCRNet (BCRNet-S) requires only 5.1<inline-formula><tex-math notation="LaTeX">\%</tex-math></inline-formula> network parameters than BCRNet, which can be eligible for implementation in BCRNet-SNN. The proposed structure of BCRNet-SNN to effectively accommodate the CNN-to-SNN-converted parameters from BCRNet-S can achieve up to a maximum accuracy of 98.11<inline-formula><tex-math notation="LaTeX">\%</tex-math></inline-formula>, without observable performance degradation compared to BCRNet.