math basics
- orthornormal basis
- identity matrix (a special othornormal basis)
- variance (square of standard deviation) and its goal
1. noise: noise should contains low or no information needed to be extracted, it can be quantified by SNR.
2. redundancy: it indicates the correlation between two datasets (or parameters, or vectors), for example, if two datasets are fully correlated, we can simply remove one dataset and use the left one for analysis.
- covariance
for the two matrix A and B (already converted into row vectors), the covariance of them can be described as:
() is the normalization factor.
in general, for a sized matrix X, the covariance is:
The diagonal terms of are the variance of particular measurement types. --&g