최소 왜곡을 갖는 Embedded 양자화에 관한 연구 = A study on Embedded Quantization for Minimum Distortion

다국어 초록 (Multilingual Abstract)

The Max quantizer is a well-known algorithm for accomplishing efficient approximate representation of analog samples from a signal source by using values when the probability density function of the source is known. In some application, it is desired to have embedded quantization that allows successively better approximation to the input sample when additional information bits become available. We know that the Max quantizer lacks this feature. However, we will show that the Max quantizers for nonuniformly distributed sources are embedded quantization. And then, we proposed two different methods to achieve embedded quantization for minimum distortion. The first is method which is considered to be the reconstruction level of the transmitted bits. The second is method using centroids according to the transmitted bits to minimize the MSE with the transmitted bits only.