Algorithm 1.

Principal component analysis (PCA)

	Input:    Sample set $\text{S}(t)={(\text{s}(t_1),\text{ s}(t_2),\cdots ,\text{s}(t_n))}^T$ Low dimensional dimension n, Process:
1:	Centralize all samples:
	$$S(t_i)\leftarrow S(t_i)-\frac{1}{n}{\displaystyle {\sum }_{i=1}^{n}S(t_i)}$$
2:	Calculate the covariance matrix of sample: SST
3:	Solving the correlation coefficient matrix
	$$R={\left(r_{ij}\right)}_{n\times n}(r_{ij}=r_{ji},r_{ii}=1)$$
4:	Solving the eigenvalues of the correlation coefficient matrix:
	$${\lambda }_1\ge {\lambda }_2\cdots \ge {\lambda }_n\ge 0$$
5:	Determine the number of principal components: m
	$${\displaystyle \sum _{i=1}^m{\lambda }_i/\sum _{i=1}^p{\lambda }_i\ge \alpha ,\alpha =80\%}$$
6:	Calculate the corresponding eigenvector:
	$$x_1=\left(\begin{array}{c}{x}_{11}\\ x_{21}\\ ⋮\\ x_{p1}\\ \end{array}\right),x_2=\left(\begin{array}{c}{x}_{12}\\ x_{22}\\ ⋮\\ x_{p2}\end{array}\right),\dots ,x_m=\left(\begin{array}{c}{x}_{1m}\\ x_{2m}\\ ⋮\\ x_{pm}\\ \end{array}\right)$$
7:	Calculate principal components:
	$$Z_i=x_{1i}{S}_1+x_{2i}{S}_2+\cdots +x_{pi}{S}_p$$
	$$i=(1,2,\cdots ,\text{m})$$