Algorithm 1 Wavelet packet decomposition of bearing signal

Input:  bearing signal sequence $f\left(t\right),t=1,2,\cdots ,m\lambda$, window width $\lambda$, wavelet packet scale l, wavelet packet function type Output:  Time-frequency feature matrix $A_{m\times s}$ 1:Perform sliding window processing on the bearing signal sequence $f\left(t\right),t=1,2,\cdots ,m\lambda$; 2:The $f\left(t\right),t=1,2,\cdots m\lambda$ was divided into m sequence, and each sequence fragment was $\lambda$; 3:for$j=1:m$do 4:  According to the type of wavelet packet function, the wavelet packet coefficient $c_{l,\nu }^n$ of the j-th sample is obtained by Mallat decomposition algorithm formula of wavelet packet; 5:  $c_{l,\nu }^n$ is arranged according to the corresponding order under the l-th scale to form the j-th row of $A_{m\times s}$; 6:end for