Maximum Mutual Information Vector Quantization of Log-Likelihood Ratios for Memory Efficient HARQ Implementations.
Matteo DanieliSøren ForchhammerJakob Dahl AndersenLars P. B. ChristensenSøren Skovgaard ChristensenPublished in: DCC (2010)
Keyphrases
- vector quantization
- log likelihood
- memory efficient
- mutual information
- information theoretic
- image compression
- vector quantizer
- similarity measure
- image registration
- feature selection
- maximum likelihood
- density estimation
- em algorithm
- exponential family
- distortion measure
- scoring function
- fractal image coding
- finite state vector quantization
- codebook design
- reduced complexity
- feature extraction