JP3060578B2 - Digital signal encoding method - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a high efficiency digital signal encoding apparatus for encoding an input digital signal.

[0002]

BACKGROUND ART _Audio, in high-efficiency encoding of the signal of voice, etc., audio, adaptive with, the number of bits of each channel is divided into a plurality of channels of the input signal in the time axis or the frequency axis such as voice There is a coding technique based on bit allocation (bit allocation).
For example, the encoding technique based on the above-mentioned bit allocation of an audio signal or the like includes band division encoding (sub-coding) in which an audio signal or the like on a time axis is divided into a plurality of frequency bands and encoded.
Band coding: SBC) or so-called adaptive conversion coding (ATC) in which a signal on the time axis is converted into a signal on the frequency axis (orthogonal conversion), divided into a plurality of frequency bands, and adaptively encoded for each band. Alternatively, the SBC is combined with so-called adaptive prediction coding (APC), a signal on the time axis is divided into bands, and each band signal is converted into a baseband (low band), and then a multi-order linear prediction analysis is performed. There is an encoding technique such as so-called adaptive bit allocation (APC-AB) for predictive encoding.

In the above-mentioned high efficiency coding, an audio signal or the like on a time axis is transformed into an axis (frequency axis) orthogonal to the time axis by an orthogonal transformation such as a fast Fourier transform (FFT) every predetermined unit time. Then, the data is divided into a plurality of bands, and the FFT coefficient data of each of the divided bands is encoded by adaptive bit allocation. This encoded data is transmitted.

[0004]

When the FFT coefficient data for each band is encoded by the adaptive bit allocation, for example, the FFT coefficient data on the frequency axis is divided into blocks, and In many cases, bit compression is further performed by performing a so-called block floating process. For this reason, the configuration for later decoding includes FFT coefficient data that has been band-divided and subjected to block floating processing for each block,
The floating coefficient for each block and the sub information including the word length information corresponding to the number of allocated bits are transmitted.

However, in the high-efficiency coding, it is desired to further increase the compression efficiency.

Accordingly, the present invention has been proposed in view of the above-described circumstances, and has as its object to provide a digital signal encoding method capable of performing higher bit compression.

[0007]

SUMMARY OF THE INVENTION A digital signal encoding method according to the present invention has been proposed to achieve the above-mentioned object. An input digital signal is orthogonally transformed and divided into a critical band. Encoding the signal component of each critical band with the number of bits according to the level of the difference between the allowable noise level for each critical band and the energy for each critical band set based on the energy for each band, and A digital signal encoding method in which signal components after orthogonal transformation are divided into blocks, block floating processing is performed for each block, and floating coefficients for each block are transmitted, wherein the block floating processing is performed for a band narrower than the critical band. When the operation is performed in small block units, word length information corresponding to the number of bits allocated to each small block is transmitted. Together, in which so as to transmit the information about the allowable noise level shifted specified range from the signal level range in the critical band only to lower predetermined level content instead of the floating coefficient for each said critical band. Here, when determining the number of allocated bits, for example, a so-called masking amount is determined from the energy of each critical band in consideration of human auditory characteristics, and each critical band is determined using an allowable noise level based on the masking amount. Is desirably determined. Further, it is preferable that the predetermined level is an amount corresponding to the masking amount.

[0008]

According to the present invention, when the block floating processing is performed on a small block having a band narrower than the critical band, a plurality of small blocks exist in one critical band. By transmitting the information on the allowable noise level and the word length information for each critical band without transmitting the floating coefficient for each block, the number of bits for the floating coefficient can be reduced. At this time, the information on the allowable noise level can be further reduced by shifting the specified range of the allowable noise level from the signal level range within the critical band by a predetermined level.

[0009]

Embodiments of the present invention will be described below with reference to the drawings. The digital signal encoding method according to the present invention employs a so-called critical band in which an input digital signal is orthogonally transformed by, for example, a fast Fourier transform (FFT) and the bandwidth becomes wider in a higher frequency band in consideration of human auditory characteristics. And the signal component of each of the critical bands is determined by the number of bits according to the level of the difference between the allowable noise level of each critical band and the energy of each critical band set based on the energy of each critical band. This is a digital signal encoding method that encodes FFT coefficient data), blocks signal components after the orthogonal transform, performs block floating processing for each block, and transmits floating coefficients for each block.

In other words, in this embodiment, the processing shown in the flowchart of FIG. 1 is performed. First, step S1
Then, the signal components after the orthogonal transformation are divided into blocks, and a block floating process is performed for each block, and a floating coefficient (floating level) for each block is performed.
To determine. In step S2, the allowable noise level for each critical band set based on the energy for each critical band as described later is determined. In step S3, the allowable noise level for each critical band and the energy for each critical band are determined. Is determined based on the number of allocated bits set based on the level of the difference between the two.

Here, in the embodiment of the present invention, when the block floating processing is performed in units of small blocks in a band narrower than the critical band, in step S4, a word corresponding to the number of bits allocated to each small block is used. The length information is obtained and transmitted, and in steps S5 and S6, a predetermined level (a level based on a masking amount described later) from the signal level range within the critical band is used instead of the floating coefficient for each critical band. Information about an allowable noise level that is shifted from a specified range to a lower one is obtained and transmitted. More specifically, a quantization table storing a value that shifts the specified range from the signal level range within the critical band to the lower side by the predetermined level is used. A value corresponding to the obtained allowable noise level is output and transmitted.

The reason why the specified range of the allowable noise level is shifted as described above is as follows. First, when the information on the permissible noise level obtained in step S2 is transmitted as it is, it is often wasteful to use the same index as the dynamic range that an actual signal can take for the permissible noise level. That is,
As will be described later, the allowable noise level is obtained based on the masking amount in consideration of the human auditory characteristics, so that the allowable noise level is always lower by a certain level than the maximum value of the actual signal level. Things. For example, the allowable noise level is about 26 dB lower than the signal level. in this way,
It is very wasteful to use the dynamic range that the signal level can take as it is with respect to the allowable noise level that is always lower by a certain level than the signal level, and it is not preferable from the viewpoint of reducing the number of bits. . For this reason, in the present embodiment, as described above, a quantization table is used which shifts the designated range from the signal level range within the critical band to a lower level by a predetermined level. As a result, the allowable noise level can be expressed with a small number of bits. In other words, the permissible noise level can be represented with the same resolution (accuracy) as in the case where the shift is not performed, even with a small number of bits. Therefore, it is possible to reduce the number of bits for transmitting an allowable noise level.

By the way, as described above, the case where the floating process is performed in a small block having a band narrower than the critical band is assumed, for example, as shown in FIG.
FIG. 2 shows a high critical band, that is, a critical band B having a wide bandwidth, and an example in which a plurality of the blocks (for example, the four small blocks b1 to b4) exist in the critical band B. Is shown.

Normally, when the inside of the critical band B is divided into small blocks b1 to b4 and the block floating processing is performed for each of the small blocks as described above, each of the small blocks is used in the subsequent decoding processing. , A floating coefficient obtained by the floating process and word length information corresponding to the number of bits allocated to each small block are required. That is, for the configuration for later decoding, it depends on the information of the floating coefficient for each small block and the number of bits allocated based on the level difference between the level of the floating coefficient and the allowable noise level of the critical band. It is necessary to transmit the word length information for each small block. In other words, at the time of subsequent decoding, the most significant bit (MSB) in the block floating processing is determined from the information on the floating coefficient, and the least significant bit (LSB) is determined from the information on the word length. An allowable noise level is determined. Further, the signal size is determined from the FFT coefficient data (main data) of each small block.

Normally, the information of the floating coefficient is represented by, for example, 6 bits, and the information of the word length is represented by, for example, 4 bits. When the orthogonal transform is a DFT (Discrete Fourier Transform), the word length information indicates a magnitude (amplitude) and a phase or a real part and an imaginary part by the four bits. Thus, for example, 1
When one critical band B is divided into a plurality of floating blocks (small blocks b1 to b4), the total number of transmission bits of the critical band according to the number of small blocks (that is, the number of band divisions) of the block floating process is: As shown in Table 1.

[Table 1]

In Table 1, when the critical band is represented by one block (one division), a total of 10 bits of 6 bits for the floating coefficient and 4 bits for the word length are transmitted. When the critical band is represented by two small blocks (divided into two), the floating coefficient is 6
× 2 (= 12 bits) and a word length of 4 × 2 (= 8 bits), a total of 20 bits are transmitted. Similarly, when divided into three, the floating coefficient is 6 × 3
(= 18 bits) and the word length is 4 × 3 (= 12 bits)
, A total of 30 bits, and in the case of four divisions, a floating coefficient is 6 × 4 (= 24 bits) and a word length is 4 × 4
(= 16 bits), a total of 40 bits are transmitted. As described above, as the number of blocks in one critical band increases, the number of transmitted bits also increases.

On the other hand, in the example of FIG. 2 of the embodiment of the present invention, each of the small blocks b1 to b
4 without transmitting the information of the floating coefficients Fc1 to Fc4, the information of the allowable noise level NL set only for the critical band B and the information of the word lengths W1 to W4 corresponding to the number of allocated bits. I try to transmit. That is, in the subsequent decoding process, if information on the allowable noise level NL of the critical band B is transmitted,
Information on the permissible noise level NL and each of the small blocks b
The floating coefficient Fc for each of the small blocks b1 to b4 based on the information on the word lengths W1 to W4 of the small blocks b1 to b4.
Since the information of the floating coefficients Fc1 to Fc4 can be obtained, the information of the floating coefficients Fc1 to Fc4 is not transmitted. Accordingly, the number of bits for transmitting the floating coefficients Fc1 to Fc4 required for the critical band B can be reduced.

In FIG. 2, the word lengths W1 to W4 are shown in the figure for convenience because the level difference for obtaining the number of allocated bits and the word lengths W1 to W4 correspond to each other. I have.

Here, the permissible noise level NL is determined for each critical band in consideration of human auditory characteristics, as described above. In the critical band, the permissible noise level is substantially within one critical band. It can be considered constant. Therefore, it can be considered that the allowable noise level NL is the same in each of the small blocks b1 to b4 in the critical band B in FIG. However, if the entire dynamic range is, for example, 120 dB and the floating coefficient is represented by the 6 bits, the floating coefficient has an accuracy of about 2 dB, and if the word length information is represented by the 4 bits, The word length information has an accuracy of about 6 dB. For this reason, in each of the small blocks b1 to b4 in FIG. 2 described above, in the subsequent decoding, the floating coefficients Fc1 to Fc4 and the word length W1
The allowable noise level NL determined from the information of .about.W4 has a deviation of about 2 dB steps. However, the allowable noise level NL usually falls within a range of approximately ± 3 dB. For this reason, in the present embodiment, in the present embodiment, the allowable noise level NL is set to the common rough quantization in the critical band B, and the small blocks b1 to b4 of the floating process in the critical band B.
The permissible noise level NL is set as a common value with high precision in two stages, each of which has fine quantization. That is, in this embodiment, since the allowable noise level NL is a 4-bit log level, this 4-bit log level is used.
The allowable noise level NL that could not be expressed by
It is represented in detail by og. Therefore, in the present embodiment, it is possible to divide about 6 dB into four to achieve 1.5 dB accuracy. Thus, the allowable noise level NL
Is substantially equal across a plurality of small blocks b1 to b4, and the number of bits can be reduced by selecting the high-precision parameter from the floating coefficient and the allowable noise level. Table 2 shows how the number of bits is reduced in the example of FIG. 2 in comparison with Table 1 above.

[Table 2]

In Table 2, when the critical band B is represented by one block (one division), the allowable noise level NL is transmitted by 4 bits and the word length W is transmitted by 4 bits. However, in the allowable noise level NL, as described above, 2 bits for compensating for a shift of 2 dB are added (4 + 2 bits). Therefore, a total of 10 bits are transmitted in the one division. Similarly,
When the critical band B is represented by two small blocks (divided into two)
Means that 4 + 2 × 2 = 8 bits for the allowable noise level NL and 4 × 2 = 8 bits for the word length W, for a total of 16 bits. Similarly, when the signal is divided into three parts, the allowable noise level NL becomes 4 + 2 × 3 = 10 bits and the word length W
2 × 4 = 3 = 12 bits, totaling 22 bits, divided into four (FIG. 2
In the case of Example 2), the allowable noise level NL is 4 + 2 × 4 = 1.
2 bits, word length W, 4 × 4 = 16 bits, total 28
Bits will be transmitted. Therefore, comparing the example of Table 2 with the example of Table 2 where the number of transmission bits in the example of Table 1 is 100%, the example of Table 2 has the same value of 100% for one division, but 80% and 3% for the two divisions. As the number of divisions (the number of blocks) increases, such as 73% for division and 70% for four divisions, the bit reduction rate increases. Therefore, it can be understood that the method of this embodiment is very effective.

That is, in this embodiment, the allowable noise level NL is set for each critical band B as shown in FIG.
As described above, the output of the quantization table in which the specified range of the allowable noise level NL is shifted from the signal level range in the critical band B to the lower side by a predetermined level, as described above. I try to transmit. By doing so, the allowable noise level NL
It is possible to reduce the number of bits for transmission.

When the allowable noise level NL is represented by, for example, 4 bits as in the case of the example of FIG. 2, the range represented by the 4 bits is equal to the predetermined level as described above. Try to shift lower.
However, information on the actual allowable noise level NL is
A 2-bit deviation for compensating for the above-mentioned 2 dB deviation is added. In addition, word length W1
The quantization of the information of W4 is not adaptive but uniform.

Further, in the example of FIG. 2, among the information of the word lengths W1 to W in one critical band B, only the information of the word length W1 is transmitted, and the other word lengths W2 to W2 are transmitted. It is also possible not to transmit the information of W4. That is, information to be transmitted is information on the floating coefficients Fc1 to Fc4 in the critical band B and information on the word length W1. In other words, in the subsequent decoding process, if information of one word length W1 is transmitted,
The information on the remaining word lengths W2 to W4 can be obtained based on the information on the floating coefficients Fc1 to Fc4. Specifically, the allowable noise level NL can be obtained from the floating coefficient Fc1 and the word length W1, and if the allowable noise level NL can be obtained,
The allowable noise level NL and the floating coefficient Fc
2 to Fc4, the remaining word lengths W2 to W4 can be known. Therefore, it is possible to prevent the information of the remaining word lengths W2 to W4 from being transmitted.
The number of bits for transmitting information of two word lengths W2 to W4 can be reduced.

FIG. 3 shows an example of a configuration to which the above-described digital signal encoding method of the present embodiment is applied.

That is, in FIG. 3, the digital audio data on the time axis supplied to the input terminal 1 is
The signal is transmitted to the fast Fourier transform circuit 11. The fast Fourier transform circuit 11 converts the audio data on the time axis into data on the frequency axis for each unit time (unit block), and obtains FFT coefficient data including a real component value Re and an imaginary component value Im. Can be These FFT coefficient data are transmitted to the amplitude / phase information generation circuit 12, and the amplitude / phase information generation circuit 12 obtains amplitude information Am and phase information Ph from the real component value Re and the imaginary component value Im. The information of the information Am is output.
That is, although human hearing is generally sensitive to the amplitude (power) in the frequency domain, it is rather insensitive to the phase. Therefore, in the present embodiment, the above-mentioned allocated bit number information is obtained using only the amplitude information Am. I have to.

The amplitude information Am is first sent to the band dividing circuit 1
3 is transmitted. The band dividing circuit 13 divides the input digital signal represented by the amplitude information Am into a so-called critical bandwidth (critical band). The critical bandwidth is based on human auditory characteristics (frequency analysis capability). For example, 0 to 22 kHz is divided into 25 bands, and the higher the frequency band, the wider the bandwidth is selected. That is, human hearing has characteristics like a kind of band-pass filter, and the band divided by each filter is called a critical band.

The amplitude information Am for each band divided into critical bands by the band division circuit 13 is transmitted to the sum detection circuit 14, respectively. In the sum detection circuit 14, the energy for each band (spectral intensity in each band) is calculated by summing the amplitude information Am in each band (amplitude information Am).
By taking the peak or the average or the total energy). The output of the sum detection circuit 14, that is, the sum spectrum of each band is generally called a bark spectrum, and the bark spectrum SB of each band is, for example, as shown in FIG.
It becomes as shown in. However, in FIG. 4, the number of the critical bands is represented by 12 bands (B _{1 to} B ₁₂ ) for simplicity of illustration.

Here, in order to consider the influence of the bark spectrum SB on so-called masking, a predetermined weighting function is convolved with the bark spectrum SB (convolution). Therefore, the output of the sum detection circuit 14, that is, each value of the bark spectrum SB is sent to the filter circuit 15. The filter circuit 15 includes, for example, a plurality of delay elements for sequentially delaying input data and a plurality of multipliers (for example, 25 corresponding to each band) for multiplying an output from these delay elements by a filter coefficient (weighting function). Multipliers) and a sum adder for summing the outputs of the multipliers. This filter circuit 1
In each of the multipliers 5, for example, the multiplier M corresponding to an arbitrary band has the filter coefficient 1, the multiplier M- 1 has the filter coefficient 0.15, and the multiplier M- 2 has the filter coefficient 0.00.
19 is multiplied by a filter coefficient 0.000008 by the multiplier M-3.
6, the filter coefficient 0.4 by the multiplier M + 1 and the multiplier M
By multiplying the output of each delay element by the filter coefficient 0.06 by +2 and the filter coefficient 0.007 by the multiplier M + 3, the convolution process of the bark spectrum SB is performed. Here, M is an arbitrary integer of 1 to 25. By this convolution processing, the sum of the parts shown by the dotted lines in FIG. 4 is obtained. The masking refers to a phenomenon that a certain signal masks another signal and makes it inaudible due to human auditory characteristics.This masking effect includes a masking effect for an audio signal on a time axis. There is a masking effect on signals on the frequency axis.
That is, even if there is noise in the masked portion due to the masking effect, this noise will not be heard. Therefore, in an actual audio signal,
The noise in this masked portion is considered acceptable noise.

Thereafter, the output of the filter circuit 15 is sent to a subtractor 16. The subtracter 16 calculates a level α corresponding to an allowable noise level described later in the convolved area. The level α corresponding to the permissible noise level (permissible noise level) is, as described later, a level at which the permissible noise level of each critical band is obtained by performing inverse convolution processing. is there. Here, an allowance function (a function expressing a masking level) for obtaining the level α is supplied to the subtractor 16. The level α is controlled by increasing or decreasing the allowable function. The permissible function is supplied from a function generation circuit 29 described later.

That is, the level α corresponding to the allowable noise level can be obtained by the following equation, where i is a number sequentially given from the lower band of the critical bandwidth. α = S− (n−ai) In this equation, n and a are constants and a> 0, and S is the intensity of the convolution-processed Bark spectrum.
ai) is an allowable function. In this embodiment, n = 38, a =
It was set to 1. At this time, there was no sound quality deterioration, and good encoding was performed.

In this way, the level α is obtained, and this data is transmitted to the divider 17. The divider 17 is for inversely convolving the level α in the convolved region. Therefore, by performing this inverse convolution processing,
A masking spectrum can be obtained from the level α. That is, this masking spectrum becomes an allowable noise spectrum. Note that the above inverse convolution process requires a complicated operation, but in the present embodiment, inverse convolution is performed using a simplified divider 17.

Next, the masking spectrum is transmitted to a subtractor 19 via a synthesis circuit 18. Here, the subtractor 19 outputs the output of the sum detection circuit 14, that is, the bark spectrum S from the sum detection circuit 14 described above.
B is supplied via a delay circuit 21. Therefore, the subtractor 19 performs a subtraction operation between the masking spectrum and the bark spectrum SB, thereby obtaining a signal shown in FIG.
As shown in (1), the bark spectrum SB is masked below the level indicated by the level of the masking spectrum MS.

The output of the subtracter 19 is sent to the ROM 30 via the allowable noise level correction circuit 20. The ROM 30 stores a plurality of pieces of assigned bit number information used for quantization of the amplitude information Am, and outputs the information of the subtraction circuit 19 (the difference between the energy of each band and the output of the noise level setting means). ), And outputs information on the number of allocated bits according to the level. Based on the allocated bit number information from the ROM 30, the quantization circuit 24 outputs the amplitude information Am supplied via the delay circuit 23.
Is quantized. That is, the ROM 30 has a quantization table as described above. Also, from the output terminal 2, together with the quantized output of the amplitude information Am and the like,
The sub information including the word length information of each small block and the information on the shifted allowable noise level is also output. It should be noted that the delay circuit 21 takes into account the amount of delay in each circuit before the synthesis circuit 18 and the sum detection circuit 14
The delay circuit 23 is provided to delay the amplitude information Am in consideration of the delay amount in each circuit before the ROM 30.

In addition, at the time of synthesizing by the synthesizing circuit 18 described above, FIG.
And data indicating a so-called minimum audible curve RC which is a human auditory characteristic as shown in FIG.
And S can be synthesized. At this minimum audible curve, if the absolute noise level is below this minimum audible curve, the noise will not be heard. Further, even if the coding is the same, the minimum audible curve differs depending on, for example, a reproduction volume at the time of reproduction. However, in a realistic digital system, for example, there is not much difference in how to enter music into the 16-bit dynamic range. For example, if quantization noise in the most audible frequency band around 4 kHz is not heard, It is considered that quantization noise below the level of the minimum audible curve is not audible in other frequency bands. Therefore, assuming that the system is used so that noise around 4 kHz of the word length of the system cannot be heard, and an allowable noise level is obtained by synthesizing the minimum audible curve RC and the masking spectrum MS together, In this case, the permissible noise level can be set up to the shaded portion in the figure. In this embodiment, the 4 kHz level of the minimum audible curve is adjusted to the lowest level corresponding to, for example, 20 bits. Also, FIG.
The signal spectrum SS is also shown.

Here, the allowable noise level correction circuit 2
In the case of 0, the allowable noise level from the subtractor 19 is corrected based on information of a so-called equal loudness curve sent from the correction value determination circuit 28. That is, the correction value determination circuit 28 outputs correction value data for correcting the allowable noise level from the subtractor 19 based on information data of a so-called equal loudness curve. By being transmitted to the noise level correction circuit 20, the allowable noise level from the subtracter 19 is corrected in consideration of the equal loudness curve. The above-mentioned equal loudness curve relates to human auditory characteristics. For example, the equal loudness curve is obtained by obtaining sound pressures of sounds at each frequency that sounds as loud as a pure sound of 1 kHz and connecting them with a curve. Also called a curve. In addition, the equal loudness curve draws substantially the same curve as the minimum audible curve RC shown in FIG. In the equal loudness curve, for example, 1 k
1 kHz even if the sound pressure falls 8-10 dB below the Hz
Sounds the same size as z, 1k around 50kHz
Unless it is higher than the sound pressure at Hz by about 15 dB, the sound cannot be heard at the same level. For this reason, it can be seen that noise exceeding the level of the minimum audible curve (allowable noise level) preferably has a frequency characteristic given by a curve corresponding to the equal loudness curve. From this, it can be seen that correcting the allowable noise level in consideration of the equal loudness curve is suitable for human auditory characteristics.

In this embodiment, a configuration may be adopted in which the above-described minimum audible curve synthesizing process is not performed. That is, in this case, the minimum audible curve generating circuit 22 and the synthesizing circuit 18 become unnecessary, and the output from the subtractor 16 is inversely convolved by the divider 17 and immediately transmitted to the subtractor 19. Will be done.

[0037]

According to the digital signal encoding method of the present invention, an input digital signal is orthogonally transformed and divided into critical bands, and each critical band is adaptively allocated based on the allowable noise level of each critical band. In addition to encoding the signal components of the band and transmitting the floating coefficient obtained by performing block floating processing on the signal components after the orthogonal transformation, when performing the floating processing on a small block of a band narrower than the critical band, In addition to transmitting word length information according to the number of bits allocated to each small block, the designated range is shifted downward by a predetermined level from the signal level range in the critical band instead of the floating coefficient for each critical band. By transmitting the information on the allowable noise level, the video for transmitting the information on the allowable noise level is transmitted. It is possible to reduce the betting amount, therefore, allows higher bit compression.

[Brief description of the drawings]

FIG. 1 is a flowchart of an embodiment of the present invention.

FIG. 2 is a diagram for explaining a case where floating processing is performed in a block narrower than a critical band.

FIG. 3 is a block circuit diagram showing a specific configuration of an embodiment of the present invention.

FIG. 4 is a diagram showing a bark spectrum.

FIG. 5 is a diagram showing a masking spectrum.

FIG. 6 is a diagram in which a minimum audible curve and a masking spectrum are combined.

[Explanation of symbols]

 B band b1 to b4 block W1 to W4 word length NL1 to NL4 allowable noise level

──────────────────────────────────────────────────続 き Continued on the front page (58) Field surveyed (Int.Cl. ⁷ , DB name) H03M 7/30

Claims (1)

(57) [Claims] 1. An input digital signal is orthogonally transformed and divided into critical bands, and a difference between an allowable noise level for each critical band set based on energy for each critical band and energy for each critical band is determined. The digital component that encodes the signal components of each critical band with the number of bits according to the level, blocks the signal components after the orthogonal transformation, performs block floating processing for each block, and transmits a floating coefficient for each block. In the signal encoding method, when the block floating process is performed in units of small blocks in a band narrower than the critical band, information of a word length according to the number of bits allocated to each of the small blocks is transmitted. A predetermined level from the signal level range in the critical band instead of the floating coefficient for each critical band. A digital signal encoding method for transmitting information on an allowable noise level shifted a designated range to a lower level.