Transform and Masking Coders
We indicated very early on that there may be some advantage in terms of the reduction in data rate to taking the Fourier transform of a block of audio data and transmitting the coefficient data. The use of a technique known as the discrete cosine transform is similar in concept and is used in the AC-2 system designed by Dolby Laboratories. This system can produce a high-quality audio signal with 128 kb per channel.
The MUSICAM process also relies on a model of the human ear’s masking processes. The encoder receives a stream of conventionally encoded PCM samples, which are then divided into 32 narrow bands by filtering. The allocation of the auditive significance of the contribution that each band can make to the overall program is then carried out prior to arranging the encoded data in the required format. The principle in this encoder is similar to that planned for the Philips digital compact cassette system.
The exploitation of the masking thresholds in human hearing lies behind many of the proposed methods of achieving bit rate reduction. One significant difference between them and conventional PCM converters is the delay between applying an analogue signal and the delivery of the digital sample sequence. A similar delay is involved when the digital signal is reconstituted. The minimum delay for a MUSICAM encoder is in the region of 9 to 24 ms depending on how it is used. These delays do not matter for a program that is being replayed but they are of concern when the coders are being used to provide live linking program material in a broadcast application.
A second, potentially more damaging, problem with these perceptual encoders is that there has been insufficient work carried out on the way in which concatenations of coders will affect the quality of the sound passing through. Although this problem affects the broadcaster more, the domestic user of such signals may also be affected. Be sure that perceptual coding techniques remove data from the original, as these data cannot be restored. Thus a listener who wishes to maintain the highest quality of audio reproduction may find that the use of his preamplifier’s control or room equalizer provides sufficient change to an encoded signal that the original assumptions concerning masking powers of the audio signal may no longer be valid. Thus the reconstituted analogue signal may well be accompanied by unwelcome noise.