Improved mmWave Signal Mode Decomposition and Time-frequency Domain Conversion Algorithm

In this study, A complete signal processing algorithm of 60 GHz millimeter wave radar echo is designed. The improved ICEEMDAN adaptive modal decomposition algorithm is used to obtain the time domain map of the chest fluctuation, and the improved all-phase FFT (ApFFT) and fast Fourier transform are used to obtain more multi-dimensional images. Through the introduction of adaptive noise, the improvement of Fourier transform window and the experimental adjustment of parameters, the accuracy of the algorithm reached 95.3% in the real experiment, and it can well realize long-term detection and recognition of various lesions such as accelerated heart rate, weakened breathing and so on.