A One-Class Classification Method for Human Gait Authentication Using Micro-Doppler Signatures

In this letter, a radar-based gait authentication method is proposed. We focus on the overfitting problem target category caused by limited training data in models and propose one-class classification model to alleviate problem. The effectiveness of such verified establishing dataset, which composed micro-Doppler spectrograms derived from nine human subjects. experimental results demonstrate that, under condition data, performances an degrade because misclassification non-target samples easily occurs. proposed effectively avoids risk,performing other existing \textcolor{red}{methods} metric Equal Error Rate.