Rotational Doppler Signal Classification Based on Support Vector Machine

The rotational Doppler effect (RDE) has been playing a vital role in the detection of rotational motion in recent years. However, the results obtained from this detection method are easily influenced by the detection conditions. It means that the characteristic of the frequency signals varies greatly under different detection conditions. How to efficiently and automatically classify and recognize different signal features is of great significance for the accurate extraction of rotating speed information in the future. Based on the well-known support vector machine (SVM) model, we built an SVM learn model to automatically classify and recognize RDE signals under different detection conditions. The results show that the SVM learn model can effectively recognize the category of detection signals with high accuracy. By accurately identifying signal categories, it can provide a great foundation for extracting target speed and other information in the future, and has broad application prospects in engineering practice.