Spectrum vector quantization

This paper proposes a new spectrum vector quantization algorithm (SVQ). SVQ conducts vector quantization in spectrum space. It is characterized by some novels. The first is the informed initialization of prototypes, which is achieved by a modified support vector clustering procedure. The second is the SVD-based spectrum analysis. This technique employs singular vector decomposition to derive all datapsilas spectrum information from a subset. Thirdly, the updating of prototypes is treated in two fashions. That is different from traditional VQ, where all prototypes are adjusted in a same manner. Experiments are conducted on real datasets to check the performance of initialization strategy, the SVD-based spectrum analysis and SVQ. Empirical evidence shows the fine performance of proposed strategies over the state of the art.