Digital audio signal coding method through allocation of quantization bits to sub-band samples split from the audio signal

A digital audio signal coding system comprising:

• splitting means for splitting a digital audio signal into a plurality of sub-band signals each being sampled and each having a specified frequency;
• and
• quantization bit number control means for controlling a number of quantization bits allocated to each sample of each of the sub-band signals in order to code the samples,
• wherein said quantization bit number control means further comprises:
• level calculating means for obtaining the level for each sub-band signal;
• logarithm calculating means for obtaining a logarithm of the level of each sub-band calculated by said level calculating means;
• and
• index calculating means for obtaining an index by multiplying the logarithm value of the level of each sub-band signal by a weight which is dependent on the level and by a weight which is dependent on the frequency specified to each sub-band,
• whereby the number of quantization bits allocated to each sample of the sub-band is determined using said index obtained by said index calculating means and using a prescribed coding bit rate permitted to the entire band for coding the sub-band signals.

A digital audio signal coding method for reducing the number of quantization bits allocated to each sample of each of a plurality of sub-band signals in order to code the samples, said method comprising the steps of:

• splitting a digital audio signal into said plurality of sub-band signals each being sampled and each having a specified frequency;
• calculating a level for each of said plurality of sub-band signals;
• calculating a logarithm value of the level of each sub-band signal;
• calculating an index by multiplying the logarithm value of the level by a weight which is dependent on the level and by a weight which is dependent on the specified frequency allocated to each sub-band;
• and
• determining the number of quantization bits to be used for each sample according to the calculated index and a prescribed coding bit rate permitted to the entire band for coding the sub-band signals, wherein the number of quantization bits allocated to each sub-band is controlled according to a ratio of an index of each sub-band to the sum of indexes of the plurality of sub-bands to which quantization bits are to be allocated.

A digital audio signal coding method for reducing the number of quantization bits allocated to each sample of each of a plurality of sub-band signals in order to code the samples, said method comprising the steps of:

• splitting a digital audio signal into said plurality of sub-band signals each being sampled and each having a specified frequency;
• calculating a level according to a peak value for each of said plurality of sub-band signals;
• calculating a logarithm value of the level value of each sub-band signal;
• calculating an index by multiplying the logarithm value of the level by a weight which is dependent on the level and by a weight which is dependent on the specified frequency allocated to each sub-band;
• and
• determining the number of quantization bits to be used for each sample according to the calculated index and a prescribed coding bit rate permitted to the entire band for coding the sub-band signals.