Variable rate coder

Claim 2

A coder according to claim 1, further comprising: a plurality of decoding means (201, 202, 20m) being provided corresponding to said plurality of coding means (101, 102, 10m) a plurality of error obtaining means (61, 62, 63, 65, 66, 67) being provided corresponding to said plurality of decoding means (201, 202, 20m), and each obtaining an error which is generated through a corresponding coding means, based on said input signal and an output of a corresponding decoding means; and a plurality of S/N obtaining means (21,22, 23) being provided corresponding to said plurality of decoding means (201, 202, 20m), and each obtaining a signal-to-noise ratio, based on the power of said input signal and the power of an output of a corresponding error obtaining means; wherein said predetermined standard comprises a first requirement that a signal-to-noise ratio in the construction through a coding means and a corresponding decoding means is not to be smaller than a predetermined threshold, and a second requirement that an output having a smallest number of output bits is to be selected among the outputs of the coding means satisfying said first requirement.

Claim 3

A coder according to claim 1, wherein: an analog to digital converter is provided in an input stage, for converting said analog input signal to a digital signal; and said bit rate calculating means (360) is arranged for determining a plurality of sucecessive numbers as said numbers of output bits of said plurality of coding means (101, 102, 10m), additionally based on an ADC signal-to-noise ratio determined by an input amplitude range and resolution in said analog to digital converter.

Claim 4

A coder according to claim 3, wherein said plurality of successive numbers are determined as outputs of 2k+1 coding means (101, 102, 10m) having number of output bits i-k to i+k with a center value of the number of output bits i, where the number i is determined by the equations, i=(S/N-PG+10.8-20log₁₀AR)/6 when S/N ≦ SNI, and i=(SNI-PG+10.8-20log₁₀AR)/6 when S/N ≧ SNI, S/N is said required value of the signal-to-noise ratio of the coder, SNI is said ADC signal-to-noise ratio in said analog to digital converter, PG is said prediction gain in each coding means, and AR is said input amplitude range expressed by a root mean square of the input signal level.

Claim 5

A coder according to claim 3 or 4, further comprising: a plurality of decoding means (201, 202, 20m) being provided corresponding to said plurality of coding means (101, 102, 10m); a plurality of error obtaining means (61, 62, 63, 65, 66, 67) being provided corresponding to said plurality of decoding means (201, 202, 20m), and each obtaining an error generated through a corresponding coding means and a corresponding decoding means based on said input signal and an output of a corresponding decoding means; and a plurality of S/N obtaining means (21, 22, 23) being provided corresponding to said plurality of decoding means (201, 202, 20m), and each obtaining a signal-to-noise ratio based on the power of said input signal and the power of an output of a corresponding error obtaining means; wherein said predetermined standard includes a first requirement that a signal-to-noise ratio in the construction through a coding means and a corresponding decoding means is not to be smaller than a predetermined threshold, and a second requirement that an output of a coding means having a smallest number of output bits is to be selected as the output of the coder among the outputs of the coding means satisfying said first requirement when the power of the input signal is greater than a predetermined value; and said predetermined standard further includes of a third requirement that an error which is generated through a coding means and a corresponding decoding means is to be smaller than a predetermined maximum, and a fourth requirement that an output of a coding means having a smallest number of output bits is to be selected as the output of the coder among the outputs of the coding means satisfying said third requirement when said power of the input signal is not greater than said predetermined value.

Claim 6

A coder according to claim 5, wherein: said predetermined standard includes a fifth requirement that said signal-to-noise ratio satisfies said first requirement and said error satisfies said second requirement, and a sixth requirement that an output of a coding means having a smallest number of output bits is to be selected as the output of the coder, among the outputs of the coding means satisfying said fifth requirement then said power of the input signal is between said first and second predetermined values.

Claim 7

A coder for transforming an input signal into a low bit rate digital signal, comprising: a plurality of coding means (101, 102, 10m) in each of which a number of output bits can be variably set, and each of which can carry out a predictive coding independently; a plurality of prediction gain obtaining means (321, 322, 32m, 331, 332, 33m) each for obtaining a prediction gain in each of said plurality of coding means (101, 102, 10m); an bit rate calculating means (360) for determining a plurality of successive numbers as said numbers of output bits of said plurality of coding means (101, 102, 10m), based on a required value of the signal- to-noise ratio of the coder and said prediction gains in said plurality of coding means (101, 102, 10m); a bit rate setting means (380, 156) for setting said plurality of successive numbers in said plurality of coding means (101, 102, 10m) as said number of output bits, respectively; and an output selecting means (400, 500) for selecting an output of one of said plurality of coding means (101, 102, 10m), as the output of said coder, based on a predetermined standard.

Claim 8

A coder according to claim 7, further comprising: a plurality of decoding means (201, 202, 20m) being provided corresponding to said plurality of coding means (101, 102, 10m); a plurality of error obtaining means (61, 62, 63, 65, 66, 67) being provided corresponding to said plurality of decoding means (201, 202, 20m), and each obtaining an error which is generated through a corresponding coding means and a corresponding decoding means, based on said input signal and an output of a corresponding decoding means; and a plurality of S/N obtaining means (21, 22, 23) being provided corresponding to said plurality of decoding means (201, 202, 20m), and each obtaining a signal-to-noise ratio, based on the power of said input signal and the power of an output of a corresponding error obtaining means; wherein said predetermined standard comprises a first requirement that a signal-to-noise ratio in the construction through a coding means and a corresponding decoding means is not to be smaller than a predetermined threshold, and a second requirement that an output having a smallest number of output bits is to be selected among the outputs of the coding means satisfying said first requirement.

Claim 9

A coder for transforming an analog input signal into a low bit rate digital signal, comprising: an analog to digital converter provided in an input stage, and converting said analog input signal to a digital signal; a plurality of coding means (101, 102, 10m) in each of which a number of output bits can be variably set, and each of which can carry out a predictive coding independently; a plurality of prediction gain obtaining means (321, 322, 32m, 331, 332, 33m) each for obtaining a prediction gain in each of said plurality of coding means (101, 102, 10m); a bit rate calculating means (360) for determining a plurality of successive numbers as said numbers of output bits of said plurality of coding means (101, 102, 10m), based on a required value of the signal-to-noise ratio of the coder, an ADC signal-to-noise ratio determined by an input amplitude range and resolution in said analog to digital converter, and said prediction gains in the plurality of coding means (101, 102, 10m); a bit rate setting means (380, 156) for setting said plurality of successive numbers in said plurality of coding means (101, 102, 10m) as the number of output bits, respectively; and an output selecting means (400, 500) for selecting an output of one of said plurality of coding means (101, 102, 10m) as the output of said coder, based on a predetermined standard.

Claim 10

A coder according to claim 9, wherein said plurality of successive numbers are determined as outputs of 2k+1 coding means (101, 102, 10m) having number of output bits i-k to i+k with a center value of the number of output bits i, where the number i is determined by the equations, i=(S/N-PG+10.8-20log₁₀AR)/6 when S/N ≦ SNI, and i=(SNI-PG+10.8-20log₁₀AR)/6 when S/N ≧ SNI, S/N is said required value of the signal-to-noise ratio of the coder, SNI is said ADC signal-to-noise ratio in said analog to digital converter, PG is said prediction gain in each coding means, and AR is said input amplitude range expressed by a root mean square of the input signal level.

Claim 11

A coder according to claim 10, further comprising: a plurality of decoding means (201, 202, 20m) being provided corresponding to said plurality of coding means (101, 102, 10m); a plurality of error obtaining means (61, 62, 63, 65, 66, 67) being provided corresponding to said plurality of decoding means (201, 202, 20m), and each obtaining an error generated through a corresponding coding means and a corresponding decoding means based on said input signal and an output of a corresponding decoding means; and a plurality of S/N obtaining means (21, 22, 23) being provided corresponding to said plurality of decoding means (201, 202, 20m), and each obtaining a signal-to-noise ratio based on the power of said input signal and the power of an output of a corresponding error obtaining means; wherein said predetermined standard includes a first requirement that a signal-to-noise ratio in the construction through a coding means and a corresponding decoding means is not to be smaller than a predetermined threshold, and a second requirement that an output of a coding means having a smallest number of output bits is to be selected as the output of the coder among the outputs of the coding means satisfying said first requirement when the power of the input signal is greater than a predetermined value; and said predetermined standard further includes of a third requirement that an error which is generated through a coding means and a corresponding decoding means is to be smaller than a predetermined maximum, and a fourth requirement that an output of a coding means having a smallest number of output bits is to be selected as the output of the coder among the outputs of the coding means satisfying said third requirement when said power of the input signal is not greater than said predetermined value.

Claim 12

A coder according to claim 10, further comprising: a plurality of decoding means (201, 202, 20m) being provided corresponding to said plurality of coding means (101, 102, 10m); a plurality of error obtaining means (61, 62, 63, 65, 66, 67) being provided corresponding to said plurality of decoding means (201, 202, 20m), and each obtaining an error which is generated through a corresponding coding means and a corresponding decoding means, based on said input signal and an output of a corresponding decoding means; and a plurality of S/N obtaining means (21, 22, 23) being provided corresponding to said plurality of decoding means (201, 202, 20m), and each obtaining a signal-to-noise ratio based on the power of said input signal and the power of an output of a corresponding error obtaining means; wherein said predetermined standard includes a first requirement that a signal-to-noise ratio in the construction through a coding means and a corresponding decoding means is not to be smaller than a predetermined threshold, and a second requirement that an output of a coding means having a smallest number of output bits is to be selected as the output of the coder, among the outputs of the coding means satisfying said first requirement when the power of the input signal is greater than a first predetermined value; said predetermined standard includes a third requirement that an error which is generated through a coding means and a corresponding decoding means is to be smaller than a predetermined maximum, and a fourth requirement that an output of a coding means having a smallest number of output bits is to be selected as the output of the coder, among the outputs of the coding means satisfying said third requirement when said power of the input signal is not greater than a second predetermined value; and said predetermined standard includes a fifth requirement that said signal-to-noise ratio satisfies said first requirement and said error satisfies said second requirement, and a sixth requirement that an output of a coding means having a smallest number of output bits is to be selected as the output of the coder, among the outputs of the coding means satisfying said fifth requirement when said power of the input signal is between said first and second predetermined values.