Artificial intelligence micro-vibration detection method based on FMCW millimeter wave radar

Abstract

The invention is suitable for the technical field of building micro-vibration detection, and provides an artificial intelligence micro-vibration detection method based on an FMCW millimeter wave radar, and the method comprises the steps: dynamically adjusting the transmitting signal parameters of the FMCW radar through employing a self-adaptive optimization strategy, and collecting a radar intermediate frequency signal; the method comprises the following steps: preprocessing a collected radar intermediate frequency signal, synchronously extracting a time-frequency domain physical feature and a deep learning abstract feature of the radar intermediate frequency signal, performing cross-domain feature fusion, and constructing a fusion feature vector; and inputting the fusion feature vector into a lightweight Transform micro-vibration identification model trained by a domain adaptive framework for processing, and obtaining a micro-vibration detection and classification result. According to the method, closed-loop dynamic optimization and active environment adaptation of emission parameters are realized through reinforcement learning, the signal-to-noise ratio of micro-vibration signals is improved, a high-quality data foundation is laid for subsequent analysis, and the performance and interpretability of the model are considered through fusion of physical features and deep learning features.

Images