JP2005151327A - Pcm data compressing/decompressing method and apparatus - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To compress data so as to be smaller than a conventional PCM block compressing/decompressing system while keeping reversibility. <P>SOLUTION: PCM data resulting from encoding using quantizing bits 2<SP>m</SP>((m) is a positive integer.) are divided into each set of (n) ((n) is a positive integer of ≥2.) to generate a sample block, and a set of bits in the 2<SP>m</SP>bits is divided into a first set of (i) bits ((i) is 0 or an integer of ≤(2<SP>m</SP>-1).) and a second set of 2<SP>m</SP>-i bits for each sample in the sample block, and divided into a sample of which the value can be represented only with the second set of bits, and a sample of which the value can not be represented only with the second set of bits. The number of required bits is (2<SP>m</SP>-i) for the first one and is 2<SP>m</SP>for the latter one and by calculating the total number of required bits for the block while changing the value of the number (i) of bits, the value of the number (i) of bits is determined to minimize the data amount of the entire block, and the number of bits is converted to represent the value of the first sample only with the second set of bits. <P>COPYRIGHT: (C)2005,JPO&NCIPI

Description

  The present invention relates to PCM data compression / decompression processing. More specifically, the present invention relates to a PCM data compression / decompression method capable of compressing / decompressing data encoded by the PCM method while maintaining lossless (lossless) performance. And an apparatus.

  Here, PCM means pulse code modulation, specifically, converting an analog electric signal into a digital electric signal by sampling, quantizing and encoding. “Sampling” means that the amplitude value of a continuous analog signal is divided by a discrete period T (T = 1 / fs, where fs is a sampling frequency), and an instantaneous value impulse of the amplitude is taken out to obtain an impulse. To create a sequence. Each of the generated impulses is called a sample. “Quantization” means that the maximum amplitude value of a sample sequence generated by sampling is divided into unit amplitudes from the positive value to the negative value to set a reference value, and the analog amplitude of the sample sequence is set to the reference value. It means to approximate to the closest one of them. “Encoding” means that each sample (having a unique amplitude value) in a sample sequence whose amplitude is digitized by quantization is a code (a pulse sequence having a constant amplitude or a binary amplitude of “0” and “1”). This is expressed in terms of a bit string.

  The compression / decompression processing of PCM data was mainly performed by “uncompressed encoding method” by pulse code modulation (PCM) represented by CD (Compact Disk) when the audio equipment was digitized. . However, in recent years, “compression encoding methods” such as MP3 and WMA are becoming mainstream. This is because, assuming that audio data is recorded on a magnetic disk or transmitted via a line, the amount of data (data size) when encoded by the uncompressed encoding method is too large. On the other hand, the compression encoding method can reduce the amount of data, but has a problem of deterioration in sound quality. That is, in the compression encoding method, the data amount and the sound quality are in a trade-off relationship. For this reason, in the past, many users were satisfied with the sound quality obtained from audio data processed by the compression coding method, with an emphasis on the advantage of easy data recording and transmission. There seemed to be many users who were dissatisfied. Recently, the data recording / transmission function has been improved due to the advancement of technology. Therefore, the amount of data can be reduced by compression while maintaining the sound quality when processed by the non-compression coding method or the good sound quality close to it. There is an increasing demand for reducing the data size as much as possible (that is, a request for a lossless compression method capable of reducing the amount of data).

  Conventionally, as a technique for compressing data encoded by the PCM method (hereinafter referred to as PCM data) without damaging the amount of information, and decompressing and reproducing the compressed PCM data (that is, lossless compression / decompression technique) “PCM block compression / decompression method” is known. In this method, “samples” of a digital data (PCM data) obtained by sampling, quantizing, and encoding an analog signal with a predetermined maximum number of quantization bits are collectively block-formed. The block generated in this way is called a “sample block” (hereinafter also simply referred to as a block). After that, the number of quantization bits necessary to express the maximum value (maximum amplitude) of all samples included in each of these sample blocks is calculated for each block. Then, the calculated number of quantization bits is defined as the number of quantization bits for the block, and all the samples in the block are encoded again, thereby reducing the total number of bits (data amount) in the block. . In this way, by changing and optimizing the number of quantization bits used for each sample block, the data can be compressed without impairing the amount of information contained in the PCM data. This method has an advantage that it can be easily realized with a small amount of calculation.

  The conventional “PCM block compression / decompression method” described above has the following problems.

  First, there is at least one sample data in the sample block that requires the maximum number of quantization bits used for PCM coding, in other words, the maximum of all samples contained in a sample block. If the number of quantization bits required to represent a value is equal to or close to the maximum number of quantization bits used for PCM encoding, the sample block is “no compression” or “low compression”. It is to end up. For this reason, “non-compressed blocks” and “low-compressed blocks” tend to increase, and as a result, the compression ratio of the entire PCM data (data amount after compression / original data amount) tends to decrease.

  The reason for the first problem is that the number of quantization bits used for each sample block is uniquely determined by the number of quantization bits necessary to express the maximum value of all samples in the sample block. There is to be. If the maximum number of quantization bits is needed to represent the value of one sample in a sample block, it will be ignored even if the majority of other samples are small enough. The sample block becomes an “uncompressed block”.

  The second problem is that the compression ratio (compression performance) varies greatly depending on the sample block length (hereinafter also simply referred to as the block length), that is, the total number of samples included in the sample block. It is difficult to determine the optimal block length for.

  The reason for the occurrence of the second problem is that if the block length is increased, the followability to changes in the value of each sample (amplitude value) deteriorates. As a result, as described in the first problem, for example, This is because if there is even one sample that requires the maximum number of quantization bits in a sample block including a silent section, the entire sample block becomes an uncompressed block. Conversely, if the block length is reduced, the ratio of the management data amount to the data amount in the sample block increases, and as a result, the intra-block compression efficiency decreases.

  The present invention has been made in consideration of the above-mentioned two problems, and its main object is to maintain the reversibility and to increase the data volume compared to the conventional PCM block compression / decompression method. PCM data compressing method and apparatus capable of compressing the PCM data, and a PCM data decompressing method and apparatus capable of decompressing PCM data compressed by the method.

  Another object of the present invention is that the dependency of the compression ratio (compression performance) on the sample block length is smaller than that in the case of the conventional PCM block compression / decompression method, and an optimum block length for compression processing is set. It is an object of the present invention to provide an easy and reversible PCM data compression method and apparatus, and a PCM data decompression method and apparatus capable of decompressing PCM data compressed by the method.

(1) The PCM data compression method of the present invention includes:
N pieces of PCM data obtained by encoding the original data by the PCM method with a quantization bit number of 2 ^m (m is a positive integer) divided into n samples (n is a positive integer of 2 or more). Generating a sample block of samples of
For each sample in the sample block, a bit group having a bit number of 2 ^{m is} a first bit group having a bit number i (i is an integer equal to or less than 0 or (2 ^m −1)) and a bit number (2 ^m −i). Dividing into a second bit group of:
Dividing all the samples in the sample block into ones whose values can be expressed only by the second bit group and those whose values cannot be expressed only by the second bit group;
For the sample whose value can be expressed only by the second bit group, the required number of bits is the number of bits of the second bit group (2 ^m -i), and the value cannot be expressed only by the second bit group. For samples, the required number of bits is 2 ^m , and the total required number of bits of the sample block is calculated while changing the value of the number of bits i;
Determining the value of the bit number i based on the total required number of bits of the sample block calculated while changing the value of the bit number i so that the total data amount of the sample block is minimized; ,
Of all samples in the sample block, those values that can be expressed only by the second bit group using the determined value of the number of bits i are expressed only by the second bit group. And a step of converting the number of bits.

In the PCM data compression method of the present invention, n pieces (n is a positive integer of 2 or more) of PCM data obtained by encoding the original data with the quantization bit number 2 ^m (m is a positive integer) by the PCM method. ) To generate a sample block consisting of n samples. Next, for each sample in the sample block, a bit group having a bit number of 2 ^m is converted into a first bit group having a bit number i (i is an integer equal to or less than 0 or (2 ^m −1)) and the bit number (2 ^m -I) is divided into the second bit group. In addition, all the samples in the sample block are divided into ones whose values can be expressed only by the second bit group and those whose values cannot be expressed only by the second bit group. For the sample whose value can be expressed only by the second bit group, the required number of bits is set to the number of bits of the second bit group (2 ^m -i), and the value is expressed only by the second bit group. For the samples that cannot be used, the required number of bits is set to 2 ^m , and the total necessary number of bits of the sample block is calculated while changing the value of the number of bits i. Thereafter, based on the total required number of bits of the sample block calculated while changing the value of the number of bits i, the value of the number of bits i is determined so that the total data amount of the sample block is minimized, Using the determined value of the number of bits i, the number of bits for which the value can be expressed only by the second bit group is converted so that the sample can be expressed only by the second bit group. Therefore, the data compression efficiency is improved in each sample block, and the compression rate of the entire PCM data can be increased, that is, the data amount can be compressed smaller than in the case of the conventional PCM block compression / decompression method.

Further, when decompressing the compressed PCM data, the set value of the number of bits i of the first bit group is determined for the sample block, and the first value for each of the samples in the sample block is determined. By discriminating whether or not a bit group is used, each of the samples using the first bit group can be reproduced as a representation expressed by a quantization bit number of 2 ^m . There is no loss of information during compression, so reversibility is maintained.

Furthermore, in the PCM data compression method of the present invention, for the sample whose value can be expressed only by the second bit group, the required number of bits is set to the number of bits (2 ^m −i) of the second bit group, and the second For the sample whose value cannot be expressed only by the 2-bit group, the required number of bits is set to 2 ^m , the total number of necessary bits of the sample block is calculated while changing the value of the number of bits i, and then the number of bits Based on the total necessary number of bits of the sample block calculated while changing the value of i, the value of the number of bits i is determined so that the total data amount of the sample block is minimized. Therefore, the conventional PCM block compression / decompression method in which the number of quantization bits of each sample block is uniquely determined by the number of quantization bits necessary to express the maximum value of the sample data in the sample block Therefore, the dependency of the compression ratio (compression performance) on the sample block length becomes smaller, and it becomes easier to set the optimum block length for the compression processing.

  As described above, according to the PCM data compression method of the present invention, while maintaining reversibility, the data amount can be compressed smaller than in the case of the conventional PCM block compression / decompression method, In addition, the dependence of the compression ratio (compression performance) on the sample block length can be reduced, and the setting of the optimum block length for the compression processing can be facilitated.

  When a large number of samples constituting PCM data are divided into sample blocks having a predetermined block length, the number of quantization bits necessary to express the value of a certain sample in the sample block is the maximum quantization bit. Even if the number of bits is equal to or close to the number, the number of quantization bits required to represent the values of the other samples may be much smaller. In such a case, the conventional “PCM block compression / decompression method” uses the number of bits equal to or close to the maximum number of quantization bits as the number of quantization bits of the entire block. If the maximum number of quantization bits is required to represent one sample data in the block, it is ignored even if the value of the majority of other samples is sufficiently small. Becomes “uncompressed block” or “low-compressed block”. Therefore, the compression ratio of the entire PCM data (data amount after compression / original data amount) tends to be low.

On the other hand, according to the PCM data compression method of the present invention, each sample included in the sample block is represented by a first bit group having a bit number i (i is 0 or an integer equal to or less than (2 ^m −1)). The sample is divided into the second bit group of the number of bits (2 ^m −i), and the sample whose value can be expressed only by the second bit group, and the sample whose value cannot be expressed only by the second bit group, Divide into For the sample whose value can be expressed only by the second bit group, the required number of bits is (2 ^m -i), and for the sample whose value cannot be expressed only by the second bit group, the required number of bits is set. As 2 ^m , the total necessary number of bits of the sample block is calculated while changing the value of the number of bits i. Based on the calculated required total number of bits of the sample block, the value of the number of bits i is determined so that the total data amount of the sample block is minimized. For this reason, the total data amount of the sample block can be reduced. In other words, considering the necessity of the maximum number of quantization bits for each sample included in the sample block, the value of the number of bits i so that the total data amount of the sample block is minimized. Can be defined (determined). For this reason, the possibility that the above-described sample blocks become “uncompressed blocks” or “low-compressed blocks” is reduced, and as a result, the compression ratio of the entire PCM data is increased.

  In a preferred example of the PCM data compression method of the present invention, the sample whose value cannot be expressed only by the second bit group is expressed by using both the first bit group and the second bit group.

  In another preferred example of the PCM data compression method of the present invention, the sample expressed using only the second bit group, and the sample expressed using both the first bit group and the second bit group. And generating a bit stream for the sample block using the samples. The bit stream includes a first bit group bit number display unit indicating a set value of the bit number i, a first bit group use display unit indicating whether or not the first bit group is used for each sample, And a data bit storage unit in which data bits for the sample block are stored.

(2) The PCM data compression apparatus of the present invention
N pieces of PCM data obtained by encoding the original data by the PCM method with a quantization bit number of 2 ^m (m is a positive integer) divided into n samples (n is a positive integer of 2 or more). Means for generating a sample block consisting of
For each sample in the sample block, a bit group having a bit number of 2 ^{m is} a first bit group having a bit number i (i is an integer equal to or less than 0 or (2 ^m −1)) and a bit number (2 ^m −i). Means for dividing into a second bit group of
Means for dividing all samples in the sample block into ones whose values can be expressed only by the second bit group and ones whose value cannot be expressed only by the second bit group;
For the sample whose value can be expressed only by the second bit group, the required number of bits is the number of bits of the second bit group (2 ^m -i), and the value cannot be expressed only by the second bit group. For the sample, a means for calculating the total required number of bits of the sample block while changing the value of the number of bits i, with the required number of bits being 2 ^m ,
Means for determining the value of the bit number i based on the total necessary number of bits of the sample block calculated while changing the value of the bit number i so that the total data amount of the sample block is minimized;
Of all samples in the sample block, those values that can be expressed only by the second bit group using the determined value of the number of bits i are expressed only by the second bit group. And a means for converting the number of bits.

  In the PCM data compression apparatus of the present invention, when the PCM data compression method of the present invention is based on the conventional PCM block compression / decompression method while maintaining reversibility for the same reason as described in (1) above. In addition, the data amount can be compressed smaller than that, and the dependence of the compression ratio (compression performance) on the sample block length can be made smaller than that of the same method to facilitate the setting of the optimum block length for the compression processing.

  In a preferred example of the PCM data compression apparatus of the present invention, the means for converting the number of bits so that the sample whose value can be expressed only by the second bit group is expressed by only the second bit group, The sample whose value cannot be expressed only by the bit group is expressed by using both the first bit group and the second bit group.

  In another preferred example of the PCM data compression device of the present invention, the means for converting the number of bits so that the sample whose value can be expressed only by the second bit group is expressed by only the second bit group, A bit stream for the sample block is generated using the samples represented by only the second bit group and the samples represented by both the first bit group and the second bit group. The bit stream includes a first bit group bit number display unit indicating a set value of the bit number i, a first bit group use display unit indicating whether or not the first bit group is used for each sample, And a data bit storage unit in which data bits for the sample block are stored.

(3) A PCM data decompression method of the present invention is a method of decompressing the PCM data compressed by the PCM data compression method of the present invention described in (1) above.
Determining a set value of the number of bits i for the sample block from the compressed PCM data;
Determining from the compressed PCM data whether the first bit group is used for each of the samples in the sample block; and
A step of decompressing the compressed PCM data using the determined set value of the number of bits i and the determination result of whether or not the first bit group is used.

  In the PCM data decompression method of the present invention, the set value of the number of bits i of the first bit group is determined for the sample block from the compressed PCM data, and each of the samples in the sample block is determined. Whether or not the first bit group is used is determined. Then, the compressed PCM data is decompressed using the determined set value of the number of bits i and the determination result of whether or not the first bit group is used. Therefore, the PCM data compressed by the PCM data compression method of the present invention described in (1) above can be expanded.

  In a preferred example of the PCM data decompression method of the present invention, the compressed PCM data includes a first bit group bit number display unit indicating a set value of the bit number i, and whether or not the first bit group is used. A first bit group use display section shown for each sample; and a data bit storage section in which data bits for the sample block are stored, and reading the first bit group bit number display section A set value of the number of bits i of the first bit group is discriminated, and the first bit group use display section is read to determine whether or not the first bit group is used for each sample.

(4) A PCM data decompression apparatus of the present invention is an apparatus for decompressing the PCM data compressed by the PCM data compression method of the present invention described in (1) above.
A bit number discriminating means for discriminating a set value of the bit number i for the sample block from the compressed PCM data;
First bit group usage determining means for determining whether or not the first bit group is used for each of the samples in the sample block from the compressed PCM data;
Data decompression means for decompressing the compressed PCM data using the determined set value of the number of bits i and the determination result of whether or not the first bit group is used is provided.

  In the PCM data decompression apparatus of the present invention, the PCM data compressed by the PCM data compression method of the present invention can be decompressed for the same reason as described in the above (3) for the PCM data decompression method of the present invention. .

  In a preferred example of the PCM data decompression apparatus according to the present invention, the compressed PCM data includes a first bit group bit number display unit indicating the set value of the bit number i, and whether or not the first bit group is used. A first bit group use display section for each sample; and a data bit storage section for storing the data bits for the sample block, wherein the bit number discrimination means includes the first bit group bit number The display unit is read to determine the set value of the number of bits i, and the first bit group use determination unit reads the first bit group use display unit to determine whether or not the first bit group is used.

  In the PCM data compression method and apparatus of the present invention, while maintaining reversibility, it is possible to compress the data amount smaller than in the case of the conventional PCM block compression / decompression method, and to the sample block length than the same method. Setting the optimum block length for the compression processing can be facilitated by reducing the dependency of the compression rate (compression performance).

  In the PCM data decompression method and apparatus of the present invention, the PCM data compressed by the PCM data compression method of the present invention can be decompressed.

  DESCRIPTION OF EXEMPLARY EMBODIMENTS Hereinafter, preferred embodiments of the invention will be described in detail with reference to the accompanying drawings.

(overall structure)
FIG. 5 is a functional block diagram showing a configuration of an audio data processing system in which a PCM data compression apparatus and a PCM data decompression apparatus according to an embodiment of the present invention are incorporated.

  The audio data processing system of FIG. 5 includes an analog sound input device 1, a digital modulator 2, a compression processing device 3, a data recording device or data transmission device 4, an expansion processing device 5, and an analog demodulator 6. And an analog sound output device 7. This system is capable of PCM encoding analog sound (sound wave) data such as voice and music, then compressing and recording or transmitting while maintaining reversibility, and further compressing and recording or transmitting data The original analog sound (sound wave) can be reproduced by expanding as necessary.

  The analog sound input device 1 converts analog sound data (sound wave) such as voice or music into an analog electric signal, and generates and outputs analog audio data. In other words, the analog sound input device 1 performs the same function as the microphone.

  The digital modulator 2 is an analog-digital converter (ADC), and performs digital modulation (PCM encoding) on analog audio data sent from the analog sound input device 1 to generate PCM audio data. Output.

  The compression processing device 3 is a PCM data compression device according to an embodiment of the present invention. The compression processing device 3 performs lossless compression on the PCM audio data sent from the digital modulator 2, so that bits of the audio data are obtained. -Generate and output a stream.

  The data recording device or data transmission device 4 records the bit stream of the audio data sent from the compression processing device 3 on a predetermined recording medium (memory, hard disk, floppy disk, compact disk, etc.), or The stream is transmitted to a predetermined location through a transmission path such as a wired / wireless analog circuit network or digital circuit network.

  The decompression processing device 5 is a PCM data decompression device according to an embodiment of the present invention. The decompression processing device 5 performs decompression processing on a bit stream of audio data sent from the data recording device or the data transmission device 4 to obtain PCM audio.・ Generate and output data. The PCM audio data generated by the decompression is completely the same as the PCM audio data output from the digital modulator 2 (bit match).

  The analog demodulator 6 is a digital-analog converter (DAC) that performs the reverse operation of the digital modulator 2, and performs digital demodulation (PCM decoding) on the PCM audio data sent from the decompression processing unit 5. Generates and outputs analog audio data.

  The analog sound output device 7 converts the analog audio data (analog electric signal) output from the analog demodulator 6 into analog sound data (sound wave) and reproduces the original sound. In other words, the analog sound output device 7 performs the same function as the speaker.

(Configuration of compression processing apparatus 3)
Next, the configuration of the compression processing apparatus (PCM data compression apparatus) 3 will be described with reference to FIG.

  As shown in FIG. 3, the compression processing device 3 includes a sample block generation unit 31, a sample division unit 32, and a data compression unit 33.

As shown in FIG. 6, the sample block generation means 31 receives n pieces of PCM audio data (2 ^m- bit PCM encoded digital acoustic data) (m is a positive integer) sent from the digital modulator 2. (N is a positive integer greater than or equal to 2) Collect to generate a sample block. In this way, a sequence of sample blocks of PCM audio data, such as a first block, a second block,... Is generated.

As shown in FIG. 7, the sample bit group dividing means 32 converts 2 ^m bit data (bit group of 2 ^m bits) of each sample in each sample block into “higher bit group ( 1st bit group) ”and“ lower bit group (second bit group) ”of the number of bits (2 ^m −i). A sample that can be represented by (2 ^m -i) bits is represented only by a lower bit group, and a sample that cannot be represented by (2 ^m -i) bits is both an upper bit group and a lower bit group. Use to express. The number of bits i is determined as follows.

First, all the samples in the sample block are divided into those in which the value can be expressed only by the lower bit group and those in which the value cannot be expressed only by the lower bit group. For samples whose value can be represented only by the lower bit group, the required number of bits is set to the number of bits of the lower bit group (2 ^m -i). The number of bits is 2 ^m , and the total necessary number of bits of the sample block is calculated while changing the value of the bit number i. Thereafter, the total required number of bits of the sample block calculated while changing the value of the number of bits i are compared with each other so that the total required number of bits of the sample block, that is, the total amount of data is minimized. The value of is determined. When the value of the number of bits i is determined, it is used to convert each sample in the block into a combination of an upper bit group and a lower bit group.

That is, as shown in FIG. 8, 0,1,2 values of i from 0 to ⁽² m ^{-1), ····,} by changing the order and ⁽² m -1), all in the block The total required number of samples is calculated, and the value of i when the obtained total required number of bits is minimized is selected and set to the number of bits i of the upper bit group of the block. Then, using the determined number of bits i, the data bit group of each sample in the block is divided into an i-bit upper bit group and a (2 ^m -i) -bit lower bit group. In other words, the data bit group of each sample is converted into a combination of an upper bit group and a lower bit group.

The data compression means 33 uses the determined number of bits i to divide each sample divided into an upper bit group of i bits and a lower bit group of (2 ^m -i) bits into (2 ^m -i ) Determine whether or not it can be expressed in bits. Then, for a sample determined to be expressible with (2 ^m -i) bits, the upper bit group is deleted and converted to be expressed only with the lower bit group of (2 ^m -i) bits. Thus, data compression is performed on the sample. On the other hand, samples that are determined not to be expressed with (2 ^m -i) bits are not subjected to such conversion, and are expressed using both the upper bit group and the lower bit group. That is, it is held as expressed in 2 ^m bits, which is the maximum number of quantization bits, and data compression is not performed. In this way, the block is compressed, and a bit stream having a data structure as shown in FIG. 9 is generated. In this data structure, an “upper bit group bit number display section” is arranged at the top of the bit stream, followed by an “upper bit group use display section”, and then a “data bit storage section”. Is arranged. Note that the arrangement order of these three regions is not limited to this.

In the “upper bit group bit number display section”, a set value [any integer of 0 to (2 ^m −1)] of “bit number i” of the sample block is arranged. The set value of the number of bits i is m-bit data. In the “upper bit group use display portion”, an “upper bit group use flag” is arranged. Since the upper bit group use flag is expressed by 1 bit per sample, it is n-bit data. For example, if a certain sample is “use upper bit group”, the value of the corresponding upper bit group use flag is 1, and if it is not “use upper bit group”, the value of the flag is 0. In the “data bit storage unit”, a “data bit stream” that is the main body of the audio data is arranged. This data bit stream is a group of bits obtained by compressing data bits of n samples and arranging them in a predetermined order (eg, temporal order of n samples). The minimum number of bits of data in the data bit stream is (1 × n) bits, and the maximum number of bits is (2 ^m × n) bits. Therefore, the total amount of data (total number of bits) of the block consisting of n samples has a minimum value of [m + n + (1 × n)] bits and a maximum value of [m + n + (2 ^m × n)] bits.

(Operation of the compression processing device 3)
Next, the operation (PCM data compression method) of the compression processing apparatus (PCM data compression apparatus) 3 having the above configuration will be described with reference to the flowchart of FIG.

  First, in step E1, a sample sequence of PCM data sent from the digital demodulator 2 is sequentially input. In step E2, it is determined whether n samples forming one sample block have been accumulated. If a determination result is No, it will return to step E1, and if a determination result is Yes, it will progress to step E3. In this way, by repeating step E1 and step E2 as many times as necessary, a sample block including n samples is generated. These steps E1 to E2 are executed by the sample block generating means 31.

In step E3, the value of the loop counter i is initialized, i.e., i = 0. In the next step E4, it is determined whether or not the value of the loop counter i exceeds the upper limit value (2 ^m −1). In this case, since the determination result is No, the process proceeds to step E5, and it is necessary to express the values of these samples on the assumption that the bit number i of the upper bit group is set to i = 0 for each sample in the block. The total number of bits is calculated for each sample. Then, the process proceeds to step E6, where the value of the loop counter i is set to i = i + 1 (that is, i = 1), and then the process returns to step E4 to determine whether the value of the loop counter i exceeds (2 ^m −1). Determine whether or not. Since the determination result in step E4 is No, the process proceeds again to step E5, and it is necessary to express the value of the sample on the assumption that the upper bit number i is set to i = 1 for each sample in the block. The total number of bits is calculated for each sample.

At this time, all samples in the sample block are divided into those that can represent the value only by the lower bit group of the number of bits (2 ^m -i) and those that cannot represent the value only by the lower bit group. For a sample whose value can be expressed only by the lower bit group, the required number of bits is the number of bits of the lower bit group (2 ^m -i). For a sample whose value cannot be expressed only by the lower bit group, the required number of bits Is 2 ^m , and the total necessary number of bits of the sample block is calculated. Thereafter, the same operation is repeated until it is determined that the value of the loop counter i exceeds (2 ^m −1), i.e., i = (2 ^m −1).

This situation is shown in FIG. When the bit number i of the “upper bit group” is i = 0, as shown in the upper diagram of FIG. 8, all the samples of the block are expressed only by the lower bit group of 2 ^m bits. . At this time, the total amount of data (total number of required bits) necessary to express all the samples in the block is calculated. When i = 1, 2,... (<(2 ^m −1)), the result is as shown in FIG. 2 ^m -i) and the lower bit group. In the case of i = (2 ^m −1), as shown in the lower diagram of FIG. 8, all the samples of the block are divided into an upper bit group having a bit number (2 ^m −1) and a lower bit group having a bit number 1. Will be expressed in combination. Also when i = 1, 2,... (2 ^m −1), the total amount of data (total number of necessary bits) necessary to express all the samples in the block is calculated.

If the determination result in step E4 is Yes, that is, if it is determined that the value of the loop counter i exceeds the upper limit value (2 ^m −1), the process proceeds to step E7, and the value of i used for the block is determined. The value of i is set to the value of i when the total sum (required total number of bits) of all the samples in the block obtained by executing steps E3 to E6 is minimized. The value of i set in this way is the number of bits i of the “upper bit group” applied to all samples of the sample block. The value of i is an integer from 0 to (2 ^m −1). Steps E3 to E7 are executed by the sample / bit group dividing means 32.

When the value of the number of bits i of the “upper bit group” used for the sample block is determined in step E7, the process proceeds to the next step E8. In step E8, each sample divided into an i-bit high-order bit group and a (2 ^m -i) -bit low-order bit group using the determined i value is represented by (2 ^m -i) bits. Determine whether it is possible. For samples determined to be expressible with (2 ^m -i) bits, the upper bit group is deleted and converted to be expressed only with the lower bit group of (2 ^m -i) bits. Thus, data compression is performed on the sample. On the other hand, samples that are determined not to be expressed with (2 ^m -i) bits are not subjected to such conversion, and are expressed using both the upper bit group and the lower bit group. That is, it is held as expressed in 2 ^m bits, which is the maximum number of quantization bits, and data compression is not performed. Step E8 is executed by the data compression means 33.

The set value [any integer of 0 to (2 ^m −1)] of “number of bits i” determined in this way is stored in the “upper bit group bit number display section” of FIG. Since the value of 0 to (2 ^m −1) can be expressed by m bits, the data amount (total number of bits) of the “upper bit group bit number display portion” is m bits.

The “upper bit group use flag” arranged for each sample in the “upper bit group use display section” of FIG. 9 is determined as follows. That is, for the samples determined to be expressed by (2 m ^-i) bits, since the (2 m ^-i) is represented by only the lower bit group of bits, the upper bit group is not being used, thus "Top The value of the “bit group use flag” is set to “0”. A sample determined not to be expressed in (2 ^m −i) bits is expressed using both the upper bit group and the lower bit group, so the value of the “upper bit group use flag” is set to “1”. The Since n upper bit group use flags are arranged in the “upper bit group use display section”, the data amount (total number of bits) of the “upper bit group use display section” is n bits.

  The original PCM audio data is compressed by performing optimization for data compression for each sample in the block as described above. The data body (data bit stream) of the sample block thus compressed is stored in the “data bit storage unit” in FIG. 9 in a predetermined order (for example, the temporal order of n samples). Thereafter, it is formed into a bit stream having the data structure shown in FIG. 9 and output (step E9).

  Since the bit stream of FIG. 9 is generated and output for each sample block, the data structure of FIG. 9 is actually repeated for the total number of sample blocks.

(Configuration of decompression processing device 5)
Next, the configuration of the decompression processing device (PCM data decompression device) 5 will be described with reference to FIG.

  As shown in FIG. 4, the decompression processing device (PCM data decompression device) 5 includes a bit number discrimination means 51, an upper bit group use judgment means 52, and a data decompression means 53.

  The bit number discriminating means 51 checks the first m bits “upper bit group bit number display section” for each sample block for the bit stream having the data structure of FIG. The value represented by the data is read, and the set value of the upper bit number i of the block is determined.

  For each sample block, the upper bit group use determining means 52 checks the n-bit “upper bit group use flag” following the first m bits, reads the value represented by each bit, and thereby reads each sample in the block. Whether or not upper bits are used is determined. The “upper bit group use flag” is arranged in the “upper bit group use display section”.

The data decompression unit 53 examines the data bit stream that is the main body of the PCM data stored in the “data bit storage unit” following the “upper bit use flag” for each sample block. Each sample determined to be “use of upper bit group” in each sample block is represented by only the lower bit group of (2 ^m −i) bits, and (2 ^m −i ) Since it is compressed into bits, for example, by adding i bits having the value “0” to the upper part of the sample, the original 2 ^m- bit sample is expanded and reproduced. On the other hand, each sample determined as “no use of upper bit group” in each sample block is expressed by both an upper bit group of i bits and a lower bit group of (2 ^m −i) bits. Since it is not compressed, the original 2 ^m- bit sample is reproduced by arranging the lower bit group as it is after the upper bit group. The samples expanded and reproduced for each sample block in this way are again arranged in order for each sample block, converted into the original PCM audio data, and output.

(Operation of the decompression processing device 5)
Next, the operation (PCM data decompression method) of the decompression processing apparatus 5 having the above configuration will be described with reference to the flowchart of FIG.

  First, in step D1, the bit stream having the data structure shown in FIG. 9 sent from the data recording device or the data transmission device 4 is sequentially input. In step D2, the first m bits of the “upper bit group bit number display section” for each sample block are checked for the bit stream, the value represented by the m bit data is read, and the upper bits of the block are read. The set value of the number i is determined. Steps D1 and D2 are executed by the bit number discrimination means 51.

  In the next step D3, for each sample block, the n-bit “upper bit use flag” following the first m bits is checked to read the value represented by each bit, and thus each sample in the block is assigned to the upper bit group. Whether or not is used. This step D3 is executed by the upper bit group use determining means 52.

In the next step D4, for each sample block, the data bit stream which is the PCM data body stored in the “data bit storage unit” following the “upper bit use flag” is examined. Since each sample determined to be “use of upper bit group” from the value of “upper bit use flag” is compressed to (2 ^m −i) bits, for example, the value “ By adding i bits having “0”, the original 2 ^mbit sample is expanded and reproduced. Since each sample determined as “no use of upper bit group” from the value of the “upper bit use flag” is not compressed, by arranging the lower bit group as it is after the upper bit group, the original 2 ^m Play a bit sample. Step D4 is executed by the sample data decompression means 53.

  In the next step D5, it is determined whether or not n sample data forming one sample block have been accumulated. If the determination result is No, the process returns to Step D4, and if the determination result is Yes, the process proceeds to the next Step D6. By repeating step D4 and step D5 as many times as necessary, a sample block including n sample data is generated. Step D5 is also executed by the sample data decompression means 53.

  In step D6, sample blocks of PCM data thus expanded / reproduced are output. At that time, these sample blocks are arranged in a predetermined order and converted to original PCM audio data. Step D6 is also executed by the sample data decompression means 53.

As described above in detail, in the PCM data compression apparatus (compression processing apparatus) 3 according to an embodiment of the present invention, the original data is encoded by the PCM method with the quantization bit number 2 ^m (m is a positive integer). The PCM audio data obtained in this way is divided into n samples (n is a positive integer equal to or greater than 2) to generate a sample block consisting of n samples. Then, for each sample in the sample block, a bit group having a bit number of 2 ^m is converted into an upper bit group having a bit number i (first bit group) and a lower bit group having a bit number (2 ^m -i) (second bit group). Bit group). The value of the number of bits i of the upper bit group is set so that the total data amount (total number of required bits) of the sample block is minimized. Thereafter, the sample that can be expressed only by the second bit group using the set value of the number of bits i is converted to be expressed only by the second bit group. In this way, for each sample included in the sample block, the number of bits (2) using the bit number i set so that the total data amount (required total number of bits) of the sample block is minimized. ^{Since the} sample that can be expressed only by the second bit group of ( ^m− i) is converted so as to be expressed by only the second bit group, the data compression efficiency is improved in each sample block, and the entire PCM data The compression rate can be increased, that is, the data amount can be compressed smaller than in the conventional PCM block compression / decompression method.

When decompressing the compressed PCM audio data, the set value of the number of bits i of the upper bit group for the sample block is determined, and the upper bit group for each of the samples in the sample block By discriminating the presence / absence of use, each of the samples using the higher-order bit group can be reproduced to be expressed by a quantization bit number of 2 ^m . There is no loss of information during compression, so reversibility is maintained. Therefore, the data amount can be reduced while maintaining good sound quality.

  Furthermore, in the PCM data compression apparatus 3 according to an embodiment of the present invention, for each sample included in the sample block, the bit is set so that the total data amount (required total number of bits) of the sample block is minimized. Since the value of the number i is determined, the conventional PCM in which the number of quantization bits of each sample block is uniquely determined by the number of quantization bits necessary to represent the maximum value of the sample data in the sample block The dependence of the compression rate (compression performance) on the sample block length becomes smaller than in the case of using the block compression / decompression method, and it becomes easy to set the optimum block length for the compression processing.

Further, in the PCM data decompression device 5, with respect to a bit stream composed of a bit group having the data structure of FIG. 9 for each sample block, the first m bits are examined for each sample block and the m-bit data is expressed. The value is read to determine the set value of the number of bits i of the upper bit group for the block. Then, by checking n bits following the first m bits and reading a value represented by each bit, it is determined whether or not the upper bit group is used for each sample in the block. The compressed sample is then decompressed to 2 ^m bits and reproduced by examining the bit stream following those n bits. Since the sample determined as “no use of upper bit group” is not compressed, the original 2 ^mbit sample is reproduced by arranging the lower bit group as it is after the upper bit group. Therefore, the PCM audio data compressed by the PCM data compression device 3 can be expanded / reproduced while maintaining good sound quality.

(Concrete example)
Next, in order to make the understanding of the present invention easier, a specific example in which m = 4 and n = 44100 in the above-described embodiment will be described.

The digital modulator 2 is a general digital music data format for analog audio data sent from the analog sound input device 1, and has a sampling frequency of 44.1 kHz and a quantization bit number (quantization accuracy) of 16. A case will be described in which PCM audio data digitally modulated (PCM encoded) with (= 2 ⁴ ) bits is generated and output.

  In this case, the compression processing device (PCM data compression device) 3 converts 441 samples into one sample block (10 mm) for digital audio data input at 705600 bits (44100 samples × 16 bits) per second. (Corresponding to the number of samples per second) (see FIG. 10). The 16-bit data of each sample in each sample block is divided into an upper bit group having a bit number i and a lower bit group having a bit number (16-i), and can be expressed by (16-i) bits. A compression process is performed for (FIG. 11). The value of the upper bit number i is set in the range of 0 to 15 so that the sum of the data amount (bit number) of all the samples in the block is minimized for each sample block.

In step E2 of FIG. 1, it is determined whether or not 441 samples forming one sample block have been accumulated. If a determination result is No, it will return to step E1, and if a determination result is Yes, it will progress to step E3. In this way, by repeating Step E1 and Step E2 as many times as necessary, a sample block including 441 sample data is generated. In step E4, it is determined whether or not the value of the loop counter i exceeds (2 ⁴ -1) = 16-1 = 15. The same operation is repeated until it is determined that the value of the loop counter i exceeds 15.

  If the determination result in step E4 is Yes, that is, if it is determined that the value of the loop counter i exceeds 15, the value of i to be used for the block is determined in step E7. The value of i is the number of bits i of the upper bit group of the sample block, and takes any value in the range of 0 to 15 (see FIG. 12).

  Samples quantized with 16 bits can take values from -32768 to +32767. For example, when i = 5, the number of bits of the upper bit group is 5 bits, and the number of bits of the lower bit group is 11 bits. Therefore, the lower bit group can take values from −1024 to +1023. Therefore, if the value (amplitude) of a sample is within the range of -1024 to +1023, the sample can be expressed by 11 bits, that is, only the lower bits. That is, data can be compressed. When the sample value exceeds the range of -1024 to +1023, it is expressed as it is in 16 bits (see FIG. 11). These processes are performed in units of one block (441 samples).

For example, in a sample block, if 200 samples out of all 441 samples fall within the range of -1024 to +1023, and the remaining 241 samples are outside the range, all samples in the block are represented. The total number of bits required for
200 samples × 11 bits + 241 samples × 16 bits = 4851 bits.

  In steps E4 to E6, the same calculation is performed 16 times while changing the value of i by 1 from i = 0 to i = 15, and the total number of bits necessary for the block is calculated for each of i = 0 to 15 To do. Then, the value of i at which the value is the smallest among the total number of bits is adopted and set to the number of bits of the upper bit group of the block (step E7).

  In steps E8 to E9, a bit stream of the block having the data structure shown in FIG. 13 is generated and output using the determined value of the number of bits i. As shown in FIG. 13, this bit stream is composed of three areas.

  That is, the first 4 bits are an “upper bit group bit number display portion” indicating a set value (0 to 15) of the bit number i of the upper bit group. The subsequent 441 bits indicate whether or not the upper bits are used in the “upper bit use flag (upper bit group use display section)”. In other words, it indicates whether or not the value of each sample is within a range that can be expressed by (16-i) bits (one bit is used for each sample). The remaining part is the main body of the compressed audio data, and the first sample to the 441th sample are sequentially packed from the left (side closer to the head) and stored. For example, when i = 5, if the bit group of the upper bit group use flag continues as 1, 0, 1, 0, 0, 1,..., The audio data (main body) is the upper bit group. Adjacent to the end of the use flag, 16 bits, 11 bits, 16 bits, 11 bits, 11 bits, 16 bits,... Note that this is a case where “higher bit use” is defined when the value of the upper bit use flag is “1”, and “no higher bit use” is defined when “0”. Similarly, data can be stored even when “upper bit use” is defined when the value of the upper bit use flag is “0” and “no upper bit use” is defined when “1”.

  In step D1, the decompression processing device (PCM data decompression device) 5 sequentially inputs a bit stream having the data structure of FIG. 13 sent from the data recording device or the data transmission device 4. In step D2, for the bit stream, the first 4 bits are examined for each sample block, and the value represented by it is read to determine the set value of the number of upper bits i of the block. In the next step D3, the “higher bit group use flag” of 441 bits following the first four bits is examined to read the value represented by each bit, and thereby whether or not the upper bit group is used for each sample in the block is determined. To do.

  In the next step D4, the data bit stream which is the PCM data body following the “upper bit group use flag” is checked, and (16-i) bits (i = 1 to 15), each sample data compressed to 16 bits is decompressed. Then, in the next step D5, it is determined whether or not 441 sample data forming one sample block have been accumulated. If the determination result is No, the process returns to Step D4, and if the determination result is Yes, the process proceeds to the next Step D6. By repeating step D4 and step D5 as many times as necessary, a sample block containing 441 samples is generated. In step D6, the PCM audio data samples thus generated are output in the form of sample blocks.

(Modification)
The embodiment described above shows an example in which the present invention is embodied. Therefore, the present invention is not limited to this embodiment, and various modifications can be made without departing from the spirit of the present invention. Needless to say. For example, in the above-described embodiment, a specific example is given of PCM audio data digitally modulated (PCM encoded) with a sampling frequency of 44.1 kHz and a quantization bit number of 16 bits, which is a general digital music data format. However, according to the present invention, reversibility is ensured in the compression / decompression process, and therefore it can be applied to any data. For example, it is possible to obtain higher compression efficiency by applying to data having an AC component including zero cross, such as differential PCM data and adaptive differential PCM data, or data having a small overall amplitude value.

In the above-described embodiment, audio data compression / decompression is described. However, the present invention is not limited to this, and PCM data obtained by encoding original data with a quantization bit number of 2 ^m by the PCM method. If so, the present invention can be applied to arbitrary data other than audio data.

It is a flowchart which shows operation | movement of the PCM data compression processing apparatus (compression processing apparatus) which concerns on one Embodiment of this invention. It is a flowchart which shows operation | movement of the PCM data expansion | extension processing apparatus (expansion processing apparatus) which concerns on one Embodiment of this invention. It is a functional block diagram which shows the structure of the PCM data compression processing apparatus (compression processing apparatus) which concerns on one Embodiment of this invention. It is a functional block diagram which shows the structure of the PCM data expansion | extension processing apparatus (decompression processing apparatus) which concerns on one Embodiment of this invention. 1 is a functional block diagram showing a configuration of an audio data processing system in which a PCM data compression processing device and a PCM data decompression processing device according to an embodiment of the present invention are incorporated. It is a conceptual diagram which shows generally the block formation of the sample of PCM data in the PCM data compression processing apparatus (compression processing apparatus) which concerns on one Embodiment of this invention. It is a conceptual diagram which shows generally the method to divide the sample of PCM data into two in the PCM data compression processing apparatus (compression processing apparatus) which concerns on one Embodiment of this invention. It is a conceptual diagram which shows generally the method of calculating the bit number i of the high-order bit group of the sample of PCM data in the PCM data compression processing apparatus (compression processing apparatus) which concerns on one Embodiment of this invention. It is a conceptual diagram which shows generally the data structure of the sample block of the PCM data compressed with the PCM data compression processing apparatus (compression processing apparatus) which concerns on one Embodiment of this invention. It is a conceptual diagram which concretely shows block formation of the sample of PCM data in the PCM data compression processing apparatus (compression processing apparatus) which concerns on one Embodiment of this invention. It is a conceptual diagram which shows concretely the method to divide the sample of PCM data into two in the PCM data compression processing apparatus (compression processing apparatus) which concerns on one Embodiment of this invention. It is a conceptual diagram which shows concretely the method to calculate the bit number i of the high-order bit group of the sample of PCM data in the PCM data compression processing apparatus (compression processing apparatus) which concerns on one Embodiment of this invention. It is a conceptual diagram which shows concretely the data structure of the sample block of PCM data compressed with the PCM data compression processing apparatus (compression processing apparatus) which concerns on one Embodiment of this invention.

Explanation of symbols

DESCRIPTION OF SYMBOLS 1 Analog sound input device 2 Digital modulator 3 Compression processing apparatus (PCM data compression apparatus)
31 Sample block generation means 32 Sample bit group division means 33 Data compression means 4 Data recording device or data transmission device 5 Decompression processing device (PCM data expansion device)
51 Bit number discriminating means 52 Upper bit group use judging means 53 Data decompressing means 6 Analog demodulator 7 Analog sound output device

Claims (10)

N pieces of PCM data obtained by encoding the original data by the PCM method with a quantization bit number of 2 ^m (m is a positive integer) divided into n samples (n is a positive integer of 2 or more). Generating a sample block of samples of
For each sample in the sample block, a bit group having a bit number of 2 ^{m is} a first bit group having a bit number i (i is an integer equal to or less than 0 or (2 ^m −1)) and a bit number (2 ^m −i). Dividing into a second bit group of:
Dividing all the samples in the sample block into ones whose values can be expressed only by the second bit group and those whose values cannot be expressed only by the second bit group;
For the sample whose value can be expressed only by the second bit group, the required number of bits is the number of bits of the second bit group (2 ^m -i), and the value cannot be expressed only by the second bit group. For samples, the required number of bits is 2 ^m , and the total required number of bits of the sample block is calculated while changing the value of the number of bits i;
Determining the value of the bit number i based on the total required number of bits of the sample block calculated while changing the value of the bit number i so that the total data amount of the sample block is minimized; ,
Of all samples in the sample block, those values that can be expressed only by the second bit group using the determined value of the number of bits i are expressed only by the second bit group. A PCM data compression method comprising: converting the number of bits. The PCM data compression method according to claim 1, wherein the sample whose value cannot be expressed only by the second bit group is expressed by using both the first bit group and the second bit group. A bit stream for the sample block is generated using the samples represented only by the second bit group and the samples represented by both the first bit group and the second bit group. Further comprising steps,
The bit stream includes a first bit group bit number display unit indicating a set value of the bit number i, a first bit group use display unit indicating whether or not the first bit group is used for each sample, 3. The PCM data compression method according to claim 1, further comprising a data bit storage unit in which data bits for the sample block are stored. N pieces of PCM data obtained by encoding the original data by the PCM method with a quantization bit number of 2 ^m (m is a positive integer) divided into n samples (n is a positive integer of 2 or more). Means for generating a sample block consisting of
For each sample in the sample block, a bit group having a bit number of 2 ^{m is} a first bit group having a bit number i (i is an integer equal to or less than 0 or (2 ^m −1)) and a bit number (2 ^m −i). Means for dividing into a second bit group of
Means for dividing all samples in the sample block into ones whose values can be expressed only by the second bit group and ones whose value cannot be expressed only by the second bit group;
For the sample whose value can be expressed only by the second bit group, the required number of bits is the number of bits of the second bit group (2 ^m -i), and the value cannot be expressed only by the second bit group. For the sample, a means for calculating the total required number of bits of the sample block while changing the value of the number of bits i, with the required number of bits being 2 ^m ,
Means for determining the value of the bit number i based on the total necessary number of bits of the sample block calculated while changing the value of the bit number i so that the total data amount of the sample block is minimized;
Of all samples in the sample block, those values that can be expressed only by the second bit group using the determined value of the number of bits i are expressed only by the second bit group. A PCM data compression apparatus comprising means for converting the number of bits. For the sample whose value can be expressed only by the second bit group, the means for converting the number of bits so that the value can be expressed only by the second bit group. The PCM data compression apparatus according to claim 4, wherein the PCM data compression apparatus is expressed using both the first bit group and the second bit group. The means for converting the number of bits so that the sample whose value can be expressed only by the second bit group is expressed only by the second bit group, the sample expressed only by the second bit group, Generating a bit stream for the sample block using the samples represented by both the first bit group and the second bit group;
The bit stream includes a first bit group bit number display unit indicating a set value of the bit number i, a first bit group use display unit indicating whether or not the first bit group is used for each sample, 6. The PCM data compression apparatus according to claim 4, further comprising a data bit storage unit in which data bits for the sample block are stored. A method for decompressing the PCM data compressed by the PCM data compression method according to claim 1, comprising:
Determining a set value of the number of bits i for the sample block from the compressed PCM data;
Determining from the compressed PCM data whether the first bit group is used for each of the samples in the sample block; and
Decompressing the compressed PCM data using the determined set value of the number of bits i and the determination result of whether or not the first bit group is used. PCM data decompression comprising: Method. The compressed PCM data includes a first bit group bit number display unit indicating the set value of the bit number i, and a first bit group use display unit indicating whether or not the first bit group is used for each sample. A data bit storage unit in which data bits for the sample block are stored,
The first bit group bit number display section is read to determine the set value of the bit number i of the first bit group, and the first bit group use display section is read to determine whether the first bit group is used. The PCM data decompression method according to claim 7, wherein determination is performed for each sample. An apparatus for decompressing the PCM data compressed by the PCM data compression method according to claim 1,
A bit number discriminating means for discriminating a set value of the bit number i for the sample block from the compressed PCM data;
First bit group usage determining means for determining whether or not the first bit group is used for each of the samples in the sample block from the compressed PCM data;
PCM comprising data decompression means for decompressing the compressed PCM data using the determined set value of the number of bits i and the determination result of whether or not the first bit group is used Data decompressor. The compressed PCM data includes a first bit group bit number display unit indicating the set value of the bit number i, and a first bit group use display unit indicating whether or not the first bit group is used for each sample. A data bit storage unit in which data bits for the sample block are stored,
The bit number determining means reads the first bit group bit number display unit to determine a set value of the bit number i of the first bit group;
10. The PCM data decompression apparatus according to claim 9, wherein the first bit group use determining means reads the first bit group use display section and determines whether or not the first bit group is used for each sample.

Images