BOUND FOR MINKOWSKI METRIC OR QUADRATIC METRIC APPLIED TO VQ CODEWORDSEARCH

A bound for a Minkowski metric based on L(p) distortion measure is pro posed and evaluated as a means to reduce the computation in vector qua ntisation. This bound provides a better criterion than the absolute er ror inequality (AEI) elimination rule on the Euclidean distortion meas ure. For the Minkowski metric of order n, this bound contributes the e limination criterion from L(1) metric to L(n) metric. This bound can a lso be extended to a quadratic metric which can be applied to the hidd en Markov model with Gaussian mixture probability density function.