C. Dunn, "Aspects of Scalable Audio Coding", presented at the 122nd Convention of the Audio Engineering Society, Vienna, paper 7081 (2007 May).
ABSTRACT - Banded weight data is transmitted as side information within coded audio bitstreams in order to achieve psychoacoustically-appropriate shaping of quantisation noise. Methods of reducing the information overhead corresponding to weight data are discussed in the context of scalable bitplane coding. Two approaches to coding band weights are compared in terms of coding efficiency and error resilience. In the first, weights are coded as a block of data at the beginning of each frame, using a predictor and Golomb coding of weight prediction residuals to achieve high coding efficiency. This approach is compared to coding weights for bands as they become significant, with weight data distributed across each coded bitstream frame.
C. Dunn, "Scalable Bitplane Runlength Coding", presented at the 120th Convention of the Audio Engineering Society, Paris, paper 6749 (2006 May).
ABSTRACT - Low-complexity audio compression offering fine-grain bitrate scalability can be realised with bitplane runlength coding. Adaptive Golomb codes are computationally simple runlength codes that allow bitplane runlength coding to achieve notable coding efficiency. For multi-block audio frames, coefficient interleaving prior to bitplane runlength coding results in a substantial increase in coding efficiency. It is shown that bitplane runlength coding is more compact than the best known SPIHT arrangement for audio coding, and achieves coding efficiency that is competitive with fixed-rate quantisation.
C. Dunn, "Efficient Audio Coding with Fine-Grain Scalability", presented at the 111th Convention of the Audio Engineering Society, New York, paper 5492 (2001 Nov).
ABSTRACT - A comparison of audio coder quantisation schemes that offer fine-grain bitrate scalability is made with reference to fixed-rate quantisation. Coding efficiency is assessed in terms of the number of bits allocated to significant transform coefficients, and the average number of significant coefficients coded. A new method of arranging the transform hierarchy for SPIHT zero tree algorithms is shown to result in significantly improved performance relative to previously reported SPIHT implementations. Results for a new quantisation algorithm are presented which suggest low-complexity fine-grain scalable coding is possible with no coding efficiency penalty relative to fixed-rate coding.