Using K-L transform to compress data is straightforward, since we can use a d dimensional vector to represent a D dimensional one without lose of much information. -Using KL transform to compress data is straightforward, since we can use ad dimensional vector to represent a D dimensional one without lose of much information.