Source coding scheme using entropy coding to code a quantized signal on a determined number of bits

Decoder, comprising:

• a coded signal parser configured to parse a coded signal in order to acquire a plurality of entropy encoded values and at least one quantized residual value;
• an entropy decoder configured to decode the plurality of entropy encoded values using an entropy decoding scheme in order to acquire a plurality of quantized values;
• and
• an inverse quantization stage configured to inverse quantize the plurality of quantized values in order to acquire an output signal;
• wherein the inverse quantization stage is configured to refine an inverse quantization level used for acquiring the output signal in dependence on the quantized residual value;
• wherein the inverse quantization stage is configured to refine an inverse quantization level for a non-zero quantized value in dependence on a quantized residual value and a width of the dead zone;
• wherein the inverse quantization stage (126) is configured to refine the inverse quantization level (172) by determining a refined inverse quantization level (174, 176) in dependence on the dead zone;
• wherein the inverse quantization stage (126) is configured to determine two refined inverse quantization levels (174, 176) for a non-zero quantized value, wherein the inverse quantization stage (126) is configured to obtain the output signal (150) by using one out of the two refined inverse quantization levels (174, 176) indicated by quantized residual value;
• wherein the inverse quantization stage (126) is configured to determine the two refined inverse quantization levels (174, 176) for a non-zero quantized value based on the two factors:
• fac_p=0.25*dz
• fac_m=0.5*(1−0.5*dz)
• wherein fac_p indicates a normalized absolute value by which a normalized absolute value of the inverse quantization level (172) is to be increased in order to obtain a first of the two refined inverse quantization levels (174), wherein fac_m indicates a normalized absolute value by which the normalized absolute value of the inverse quantization level (172) is to be decreased in order to obtain a second of the two refined inverse quantization levels (176), and wherein dz is a normalized width of the dead zone;
• wherein the inverse quantization stage (126) is configured to determine two refined inverse quantization levels (174, 176) for a zero quantized value, wherein the inverse quantization stage (126) is configured to obtain the output signal (150) by using one out of the inverse quantization level (172) and two refined inverse quantization levels (174, 176) indicated by quantized residual value;
• wherein the inverse quantization stage (126) is configured to determine the two refined inverse quantization levels (174, 176) for the zero quantized value based on the factor:
• fac_z=dz/3
• wherein fac_z indicates a normalized absolute value by which a normalized absolute value of the inverse quantization level (172) is to be increased in order to obtain a first of the two refined inverse quantization levels (174) and a normalized absolute value by which a normalized absolute value of the inverse quantization level (172) is to be decreased in order to obtain a second of the two refined inverse quantization levels (176), and wherein dz is a normalized width of the dead zone;
• wherein at least one out of the coded signal parser, the entropy decoder and the inverse quantization stage is implemented by means of a processor or electronic device.

System, comprising:

• an encoder;
• and
• a decoder;
• wherein the encoder comprises a quantization stage configured to quantize an input signal using a dead zone in order to acquire a plurality of quantized values;
• wherein the encoder comprises an entropy encoder configured to encode the plurality of quantized values using an entropy encoding scheme in order to acquire a plurality of entropy encoded values;
• wherein the encoder comprises a residual quantization stage configured to quantize a residual signal caused by the quantization stage, wherein the residual quantization stage is configured to determine, for a non-zero quantized value, at least one quantized residual value and a width of a residual quantization cell in dependence on the dead zone of the quantization stage;
• and
• wherein the encoder comprises a coded signal former configured to form a coded signal from the plurality of entropy encoded values and the at least one quantized residual value;
• wherein the coded signal former is configured to form the coded signal by appending the at least one quantized residual value or a plurality of quantized residual values to the plurality of entropy encoded values until the coded signal comprises a maximum length available for a transfer to a decoder;
• wherein the at least one quantized residual value is not entropy encoded in order to stop the coding whenever the coded signal comprises the maximum length available for the transfer to the decoder;
• wherein the residual quantization stage is configured to re-quantize the quantized spectrum line by line, by
• first, determining the at least one quantized residual value such that the quantized residual value comprises 1 bit for a non-zero quantized value, wherein the residual quantization stage is configured to determine the quantized residual value based on the syntax
• prm
• ⁡
• [
• n
• ]
• =
• {
• 0
• if
• ⁢
• ⁢
• x
• ⁡
• [
• i
• ]
• <
• x_Q
• ⁡
• [
• i
• ]
• 1
• otherwise
• wherein prm is a bit stream generated by the residual quantization stage using the quantized residual value, wherein x[i] is the input signal, wherein x_Q[i] is the inverse quantized input signal, wherein n is an index that is incremented by 1 for each non-zero quantized value, and wherein i is an index that is incremented by 1 for each obtained quantized value;
• and second, if remaining bits allow, determining the quantized residual value such that the quantized residual value comprises 2 bits for a zero quantized value, wherein the residual quantization stage is configured to determine the quantized residual value based on the syntax
• prm
• ⁡
• [
• n
• ]
• =
• {
• 0
• if
• ⁢
• ⁢
• 
• x
• ⁡
• [
• i
• ]
• 
• <
• (
• C
• ·
• x_Q
• ⁡
• [
• i
• ]
• )
• 1
• otherwise
• ⁢
• ⁢
• if
• ⁢
• ⁢
• prm
• ⁡
• [
• n
• ]
• ==
• 1
• ⁢
• ⁢
• then
• ⁢
• ⁢
• prm
• ⁡
• [
• n
• +
• 1
• ]
• =
• {
• 0
• if
• ⁢
• ⁢
• x
• ⁡
• [
• i
• ]
• <
• 0
• 1
• otherwise
• wherein C depends on the dead zone of the quantization stage, wherein prm is a bit stream generated by the residual quantization stage using the quantized residual value, wherein x[i] is the input signal, wherein x_Q[i] is the inverse quantized input signal, wherein n is an index that is incremented by 1 for each zero quantized value that is requantized to a zero quantized value and incremented by 2 for each zero quantized value that is requantized to a non-zero quantized value, and wherein i is an index that is incremented by 1 for each obtained quantized value;
• wherein at least one out of the quantization stage, the entropy encoder, the residual quantization stage and the coded signal former is implemented by means of a processor or electronic device;
• wherein the decoder comprises a coded signal parser configured to parse a coded signal in order to acquire a plurality of entropy encoded values and at least one quantized residual value;
• wherein the decoder comprises an entropy decoder configured to decode the plurality of entropy encoded values using an entropy decoding scheme in order to acquire a plurality of quantized values;
• and
• wherein the decoder comprises an inverse quantization stage configured to inverse quantize the plurality of quantized values in order to acquire an output signal;
• wherein the inverse quantization stage is configured to refine an inverse quantization level used for acquiring the output signal in dependence on the quantized residual value;
• wherein the inverse quantization stage is configured to refine an inverse quantization level for a non-zero quantized value in dependence on a quantized residual value and a width of the dead zone;
• wherein the inverse quantization stage (126) is configured to refine the inverse quantization level (172) by determining a refined inverse quantization level (174, 176) in dependence on the dead zone;
• wherein the inverse quantization stage (126) is configured to determine two refined inverse quantization levels (174, 176) for a non-zero quantized value, wherein the inverse quantization stage (126) is configured to obtain the output signal (150) by using one out of the two refined inverse quantization levels (174, 176) indicated by quantized residual value;
• wherein the inverse quantization stage (126) is configured to determine the two refined inverse quantization levels (174, 176) for a non-zero quantized value based on the two factors:
• fac_p=0.25*dz
• fac_m=0.5*(1−0.5*dz)
• wherein fac_p indicates a normalized absolute value by which a normalized absolute value of the inverse quantization level (172) is to be increased in order to obtain a first of the two refined inverse quantization levels (174), wherein fac_m indicates a normalized absolute value by which the normalized absolute value of the inverse quantization level (172) is to be decreased in order to obtain a second of the two refined inverse quantization levels (176), and wherein dz is a normalized width of the dead zone;
• wherein the inverse quantization stage (126) is configured to determine two refined inverse quantization levels (174, 176) for a zero quantized value, wherein the inverse quantization stage (126) is configured to obtain the output signal (150) by using one out of the inverse quantization level (172) and two refined inverse quantization levels (174, 176) indicated by quantized residual value;
• wherein the inverse quantization stage (126) is configured to determine the two refined inverse quantization levels (174, 176) for the zero quantized value based on the factor:
• fac_z=dz/3
• wherein fac_z indicates a normalized absolute value by which a normalized absolute value of the inverse quantization level (172) is to be increased in order to obtain a first of the two refined inverse quantization levels (174) and a normalized absolute value by which a normalized absolute value of the inverse quantization level (172) is to be decreased in order to obtain a second of the two refined inverse quantization levels (176), and wherein dz is a normalized width of the dead zone;
• wherein at least one out of the coded signal parser, the entropy decoder and the inverse quantization stage is implemented by means of a processor or electronic device.

Method for encoding, the method comprising:

• quantizing an input signal in order to acquire a plurality of quantized values using a dead zone;
• encoding the plurality of quantized values using an entropy encoding scheme in order to acquire a plurality of entropy encoded values;
• quantizing a residual signal caused by quantizing the input signal, wherein quantizing the residual signal comprises determining, for a non-zero quantized value, at least one quantized residual value and a width of a residual quantization cell in dependence on the dead zone of the quantization stage;
• and
• forming a bit stream from the plurality of entropy encoded values and the plurality of quantized residual values;
• wherein forming the bitstream comprises appending the at least one quantized residual value or a plurality of quantized residual values to the plurality of entropy encoded values until the coded signal comprises a maximum length available for a transfer to a decoder;
• wherein the at least one quantized residual value is not entropy encoded in order to stop the coding whenever the coded signal comprises the maximum length available for the transfer to the decoder;
• wherein quantizing the residual signal comprises re-quantize the quantized spectrum line by line, by
• first, determining the at least one quantized residual value such that the quantized residual value comprises 1 bit for a non-zero quantized value, wherein the residual quantization stage is configured to determine the quantized residual value based on the syntax
• prm
• ⁡
• [
• n
• ]
• =
• {
• 0
• if
• ⁢
• ⁢
• x
• ⁡
• [
• i
• ]
• <
• x_Q
• ⁡
• [
• i
• ]
• 1
• otherwise
• wherein prm is a bit stream generated by the residual quantization stage using the quantized residual value, wherein x[i] is the input signal, wherein x_Q[i] is the inverse quantized input signal, wherein n is an index that is incremented by 1 for each non-zero quantized value, and wherein i is an index that is incremented by 1 for each obtained quantized value;
• and second, if remaining bits allow, determining the quantized residual value such that the quantized residual value comprises 2 bits for a zero quantized value, wherein the residual quantization stage is configured to determine the quantized residual value (146) based on the syntax
• prm
• ⁡
• [
• n
• ]
• =
• {
• 0
• if
• ⁢
• ⁢
• 
• x
• ⁡
• [
• i
• ]
• 
• <
• (
• C
• ·
• x_Q
• ⁡
• [
• i
• ]
• )
• 1
• otherwise
• ⁢
• ⁢
• if
• ⁢
• ⁢
• prm
• ⁡
• [
• n
• ]
• ==
• 1
• ⁢
• ⁢
• then
• ⁢
• ⁢
• prm
• ⁡
• [
• n
• +
• 1
• ]
• =
• {
• 0
• if
• ⁢
• ⁢
• x
• ⁡
• [
• i
• ]
• <
• 0
• 1
• otherwise
• wherein C depends on the dead zone of the quantization stage, wherein prm is a bit stream generated by the residual quantization stage using the quantized residual value, wherein x[i] is the input signal, wherein x_Q[i] is the inverse quantized input signal, wherein n is an index that is incremented by 1 for each zero quantized value that is requantized to a zero quantized value and incremented by 2 for each zero quantized value that is requantized to a non-zero quantized value, and wherein i is an index that is incremented by 1 for each obtained quantized value.

Method for decoding, the method comprising:

• parsing a coded signal in order to acquire a plurality of entropy encoded values and a quantized residual value;
• decoding the plurality of entropy encoded values using an entropy decoding scheme in order to acquire a plurality of quantized values;
• inverse quantizing the plurality of quantized values in order to acquire an output signal;
• and
• refining an inverse quantization level used for acquiring the output signal in dependence on a width of a dead zone and the quantized residual value;
• wherein refining the inverse quantization level comprises refining an inverse quantization level for a non-zero quantized value in dependence on a quantized residual value and a width of the dead zone;
• wherein refining the inverse quantization level comprises refining the inverse quantization level by determining a refined inverse quantization level in dependence on the dead zone;
• wherein refining the inverse quantization level comprises determining two refined inverse quantization levels for a non-zero quantized value, wherein the inverse quantization stage is configured to obtain the output signal by using one out of the two refined inverse quantization levels indicated by quantized residual value;
• wherein refining the inverse quantization level comprises determining the two refined inverse quantization levels for a non-zero quantized value based on the two factors:
• fac_p=0.25*dz
• fac_m=0.5*(1−0.5*dz)
• wherein fac_p indicates a normalized absolute value by which a normalized absolute value of the inverse quantization level is to be increased in order to obtain a first of the two refined inverse quantization levels, wherein fac_m indicates a normalized absolute value by which the normalized absolute value of the inverse quantization level is to be decreased in order to obtain a second of the two refined inverse quantization levels, and wherein dz is a normalized width of the dead zone;
• wherein refining the inverse quantization level comprises determining the two refined inverse quantization levels for a zero quantized value, wherein the inverse quantization stage is configured to obtain the output signal by using one out of the inverse quantization level and two refined inverse quantization levels indicated by quantized residual value;
• wherein refining the inverse quantization level comprises determining the two refined inverse quantization levels for the zero quantized value based on the factor:
• fac_z=dz/3
• wherein fac_z indicates a normalized absolute value by which a normalized absolute value of the inverse quantization level is to be increased in order to obtain a first of the two refined inverse quantization levels and a normalized absolute value by which a normalized absolute value of the inverse quantization level is to be decreased in order to obtain a second of the two refined inverse quantization levels, and wherein dz is a normalized width of the dead zone.

A non-transitory digital storage medium having a computer program stored thereon to perform the method for encoding, the method comprising:

• quantizing an input signal in order to acquire a plurality of quantized values using a dead zone;
• encoding the plurality of quantized values using an entropy encoding scheme in order to acquire a plurality of entropy encoded values;
• quantizing a residual signal caused by quantizing the input signal, wherein quantizing the residual signal comprises determining, for a non-zero quantized value, at least one quantized residual value and a width of a residual quantization cell in dependence on the dead zone of the quantization stage;
• and
• forming a bit stream from the plurality of entropy encoded values and the plurality of quantized residual values;
• wherein forming the bitstream comprises appending the at least one quantized residual value or a plurality of quantized residual values to the plurality of entropy encoded values until the coded signal comprises a maximum length available for a transfer to a decoder;
• wherein the at least one quantized residual value is not entropy encoded in order to stop the coding whenever the coded signal comprises the maximum length available for the transfer to the decoder;
• wherein quantizing the residual signal comprises re-quantize the quantized spectrum line by line, by
• first, determining the at least one quantized residual value such that the quantized residual value comprises 1 bit for a non-zero quantized value, wherein the residual quantization stage is configured to determine the quantized residual value based on the syntax
• prm
• ⁡
• [
• n
• ]
• =
• {
• 0
• if
• ⁢
• ⁢
• x
• ⁡
• [
• i
• ]
• <
• x_Q
• ⁡
• [
• i
• ]
• 1
• otherwise
• wherein prm is a bit stream generated by the residual quantization stage using the quantized residual value, wherein x[i] is the input signal, wherein x_Q[i] is the inverse quantized input signal, wherein n is an index that is incremented by 1 for each non-zero quantized value, and wherein i is an index that is incremented by 1 for each obtained quantized value;
• and second, if remaining bits allow, determining the quantized residual value such that the quantized residual value comprises 2 bits for a zero quantized value, wherein the residual quantization stage is configured to determine the quantized residual value (146) based on the syntax
• prm
• ⁡
• [
• n
• ]
• =
• {
• 0
• if
• ⁢
• ⁢
• 
• x
• ⁡
• [
• i
• ]
• 
• <
• (
• C
• ·
• x_Q
• ⁡
• [
• i
• ]
• )
• 1
• otherwise
• ⁢
• ⁢
• if
• ⁢
• ⁢
• prm
• ⁡
• [
• n
• ]
• ==
• 1
• ⁢
• ⁢
• then
• ⁢
• ⁢
• prm
• ⁡
• [
• n
• +
• 1
• ]
• =
• {
• 0
• if
• ⁢
• ⁢
• x
• ⁡
• [
• i
• ]
• <
• 0
• 1
• otherwise
• wherein C depends on the dead zone of the quantization stage, wherein prm is a bit stream generated by the residual quantization stage using the quantized residual value, wherein x[i] is the input signal, wherein x_Q[i] is the inverse quantized input signal, wherein n is an index that is incremented by 1 for each zero quantized value that is requantized to a zero quantized value and incremented by 2 for each zero quantized value that is requantized to a non-zero quantized value, and wherein i is an index that is incremented by 1 for each obtained quantized value,
• when said computer program is run by a computer.

A non-transitory digital storage medium having a computer program stored thereon to perform the method for decoding, the method comprising:

• parsing a coded signal in order to acquire a plurality of entropy encoded values and a quantized residual value;
• decoding the plurality of entropy encoded values using an entropy decoding scheme in order to acquire a plurality of quantized values;
• inverse quantizing the plurality of quantized values in order to acquire an output signal;
• and
• refining an inverse quantization level used for acquiring the output signal in dependence on a width of a dead zone and the quantized residual value;
• wherein refining the inverse quantization level comprises refining an inverse quantization level for a non-zero quantized value in dependence on a quantized residual value and a width of the dead zone;
• wherein refining the inverse quantization level comprises refining the inverse quantization level by determining a refined inverse quantization level in dependence on the dead zone;
• wherein refining the inverse quantization level comprises determining two refined inverse quantization levels for a non-zero quantized value, wherein the inverse quantization stage is configured to obtain the output signal by using one out of the two refined inverse quantization levels indicated by quantized residual value;
• wherein refining the inverse quantization level comprises determining the two refined inverse quantization levels for a non-zero quantized value based on the two factors:
• fac_p=0.25*dz
• fac_m=0.5*(1−0.5*dz)
• wherein fac_p indicates a normalized absolute value by which a normalized absolute value of the inverse quantization level is to be increased in order to obtain a first of the two refined inverse quantization levels, wherein fac_m indicates a normalized absolute value by which the normalized absolute value of the inverse quantization level is to be decreased in order to obtain a second of the two refined inverse quantization levels, and wherein dz is a normalized width of the dead zone;
• wherein refining the inverse quantization level comprises determining the two refined inverse quantization levels for a zero quantized value, wherein the inverse quantization stage is configured to obtain the output signal by using one out of the inverse quantization level and two refined inverse quantization levels indicated by quantized residual value;
• wherein refining the inverse quantization level comprises determining the two refined inverse quantization levels for the zero quantized value based on the factor:
• fac_z=dz/3
• wherein fac_z indicates a normalized absolute value by which a normalized absolute value of the inverse quantization level is to be increased in order to obtain a first of the two refined inverse quantization levels and a normalized absolute value by which a normalized absolute value of the inverse quantization level is to be decreased in order to obtain a second of the two refined inverse quantization levels, and wherein dz is a normalized width of the dead zone,
• when said computer program is run by a computer.