JP3021520B2 - Data compression encoder - Google Patents

Description

Description: BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a PCM signal data compression encoding apparatus suitable for use in, for example, a DSP (Digital Signal Processor).

[Summary of the Invention]

The present invention is directed to a data compression encoding apparatus which blocks an input signal for each fixed sample, performs requantization based on range information corresponding to the maximum value of each block, and performs noise shaping processing. By controlling the range information when an overflow occurs due to processing, it is possible to eliminate the overflow and to prevent the S / N ratio from deteriorating during reproduction or the like. This is a data compression encoding device that uses a value obtained by weighting the maximum value in the block as the range information in a predetermined manner and increases the weight of the paper at the time of the overflow to obtain the same effect as described above.

[Conventional technology]

In general, as a method of reducing the transmission bit rate by compressing the number of bits of the supplied PCM signal, for example, a filter that can obtain the highest compression ratio in block units of a plurality of samples is selected from among a plurality of filters prepared in advance. There is known a so-called filter-selection type bit compression encoding system for selection.

To such a filter selection type bit compression encoding system, for example, an input signal having 16 blocks as one block is supplied for each block. The input signal for each block is a plurality of filters, for example, a straight
The signals are supplied to three filters: a zero-order filter that outputs PCM, a first-order difference filter that outputs a first-order difference, and a second-order difference filter that outputs a second-order difference. Then, the maximum absolute value in the block is detected for each filter, and the 16-bit block data that has passed through the filter that minimizes the maximum absolute value in the block is selected and supplied to the quantizer.
The quantizer is supplied with range information detected based on the maximum absolute value in the block, and the quantizer sees from the upper bits of the supplied 16-bit block data based on this range information. First, for example, 4 bits are extracted from the bits for which “1” has just been set, that is, so-called requantization is performed and output.

At the time of the requantization, a so-called noise shaping process is performed in which the difference between the input and output of the quantizer is fed back to the input side of the quantizer and added to data newly supplied to the quantizer. Have been done. At the time of the output, filter information indicating a filter through which the data of the recorded block has passed, range information indicating a range of the data, and the like are output together with the data, and the decoder side responds to the information. Data is reproduced.

[Problems to be solved by the invention]

However, the above-described conventional data compression encoding apparatus adds the data fed back by the noise shaping to new data on the input side of the quantizer, so that the added data becomes data beyond the range. When an overflow occurs, the data in which the overflow occurs is output as it is, which leads to a deterioration in the SN ratio on the reproducing side.

In order to prevent this overflow, a so-called weighting is performed by multiplying the maximum value in the block by a predetermined coefficient,
For example, data is taken out from the bit one bit higher than the bit that should be taken out. However, since this weighting is always performed, the value of the encoded data is always smaller, and the SN at the time of reproduction is also The ratio is worse.

Further, when the weighting is performed, the value of the data fed back by the noise shaping is slightly increased, and the overflow is likely to occur. Note that this has a greater effect on an input signal whose input spectrum distribution is not uniform, such as a sine wave.

SUMMARY OF THE INVENTION The present invention has been made in view of the above-described problem, and eliminates an overflow and does not output data in which the overflow has occurred.
It is an object of the present invention to provide a data compression encoding device capable of preventing the deterioration of the ratio.

[Means for solving the problem]

The present invention relates to a data compression encoding apparatus which blocks an input signal for each fixed sample, performs requantization based on range information corresponding to the maximum value of each block, and performs noise shaping processing. Solves the above-mentioned problem by controlling the range information when an overflow occurs, and also uses a value obtained by weighting the maximum value in the block as a predetermined value as the range information. The above problem is solved by increasing the value of the weight.

[Action]

The data compression encoding device according to the present invention can control range information when an overflow occurs in the noise shaping, and this uses a value obtained by weighting the maximum value in the block with a predetermined weight. This can be simplified by increasing the value of the weight at the time of the overflow.

〔Example〕

Hereinafter, an embodiment of a data compression encoding apparatus according to the present invention will be described with reference to the drawings.

FIG. 1 is a functional block diagram showing functions of an embodiment of a data compression encoding apparatus according to the present invention in a block manner.

In FIG. 1, the audio data generation block 1 is such that reading can be stopped at an arbitrary point and reproduction can be instantaneously performed from the stopped point.
It can be configured using a device that generates digital data such as a CD (compact disk) player or a DAT (digital audio tape recorder), a memory, and the like.

Reading from the audio data generating block 1 is controlled by a control signal from a reading control circuit 22. For example, one sample is converted into 16 bits (1 word),
An audio PCM signal is generated as one block in the sample. The audio PCM signal is stored in a buffer 3 via a switch 2 described later, and is also stored in a predictor 4.
Are supplied to the zero-order filter 5, primary filter 6, secondary filter 7, and counter 8. The encode filter has a configuration of a so-called difference filter. The audio PCM signal of one block supplied to the zero-order filter 5 is used as a so-called straight PCM signal as it is without any difference. Supplied to

In the range detector 9, the maximum absolute value in one block of the supplied straight PCM signal is detected, the maximum absolute value in the block is compared to a comparison (minimum value detection) block 14, and all data of the block is selected by a selector.
Supplied to 15.

Next, the audio PCM signal of one block supplied to the primary filter 6 has a primary difference in the block, and the primary difference data is supplied to the range detector 10. The range detector 10 detects the maximum absolute value in the block from the primary difference data of one block.
The maximum absolute value of the primary difference data in the block is supplied to the weighting unit 12, and the primary difference data of the one block is supplied to the selector 15. The maximum absolute value in the block of the primary difference data supplied to the weighter 12 is
So-called weighting, which is multiplied by a predetermined coefficient (for example, 1.5), is supplied to the comparison block 14.

Next, the audio PCM signal of one block supplied to the secondary filter 7 is subjected to a second-order difference in the block and supplied to the range detector 11. Range detector
In step 11, the maximum absolute value in the block is detected from the supplied secondary difference data of one block. The maximum absolute value of the secondary difference data in the block is supplied to the weighting unit 13, and the secondary difference data of the one block is supplied to the selector 15. Then, the maximum absolute value in the block of the secondary difference data supplied to the weighting unit 13 is determined by a predetermined coefficient (for example,
The weight is multiplied by 1.5), and is supplied to the comparison block 14.

The comparison block 14 has a maximum absolute value in the block of the supplied straight PCM data,
The maximum absolute value in the block of the next difference data and the maximum absolute value in the block of the weighted secondary difference data are compared to detect the one with the smallest value, and the range of the detected minimum value is output as range information. At the same time, it detects which of the three filters the minimum value has passed, and outputs this as filter information.

The range information is supplied to the quantizer 18 and the multiplexer 17, and the filter information is supplied to the selector 15 and the noise shaping circuit 19 and to the multiplexer 17.

The selector 15 selects an audio PCM signal through a filter specified by the filter information from the comparison block 14 by, for example, switching a switch, and converts the selected one block of audio PCM signal into a quantizer 18.
To supply.

As described above, the selector 15 selects an optimum one from the audio PCM signals passed through the three filters and supplies the selected signal to the quantizer 18. The counter 8 determines the number of blocks of the supplied audio PCM signal. Counting is performed, and when a predetermined count number (for example, several blocks to several tens of blocks) is reached, a count reaching signal is supplied to each of the weighters 12 and 13.

When the count arrival signal is supplied, each of the weighters 12 and 13 increases the weighting coefficient, which was 1.5, for example, to a sufficiently large value (for example, 100). As described above, since the selector 15 selects the block having the largest absolute value in the minimum block from the three supplied audio PCM signals, the selector 15 necessarily blocks the straight PCM signal through the zero-order filter 5. Is selected, and even in the worst case, the block of the straight PCM signal can be forcibly arranged in the fixed block number count.

Here, the constant interval (constant block count)
Is N, and the number of samples in one block is n (for example, n = 28 in CDI and n = 16 in APU), it is guaranteed that the 0th-order filter 9 is always selected for every nN samples, and BRR encoding is performed. and when reproducing data from an arbitrary point, after nNt _{s [sec]} at the longest (t _s = sampling interval _[sec]) can obtain a complete reproduced sound. Specifically, f _s = 32 kHz, 1 when the block is referred to as 16 samples the time required for one block of about 0.5 _[msec], and the 20 to 30 even if arranged straight PCM signal every several tens blocks _{s [msec]} They can be arranged at intervals of the order. This interval is not a problem because it is not recognizable by human hearing.

The above until the block of the straight PCM signal is disposed is so that the resulting some errors may be allowed to mask data during the nNt _{s [sec]} from the start.

The 16-bit audio PCM signal thus selected and supplied to the quantizer 18 is used for the prediction range adaptive circuit 1
According to the range information supplied from 4, for example, 4 bits are taken out from the first 16 bits of the 16 bits as viewed from the upper bits, for example, 4 bits are extracted. Multiplexer 17
Supplied to The noise shaping path 19 feeds back a so-called quantization error, which is an error between an input and an output of the quantizer 18, to the input of the quantizer 18 according to the filter information from the prediction range adaptation circuit 14. I have. That is, the data of the lower 4 bits extracted at the time of the requantization is fed back to the input of the quantizer 18 and added to the data newly supplied to the quantizer 18.

Here, as described above, the quantizer 18 outputs, for example, 4 bits counting from the first bit of “1” in the 16 bits as viewed from the upper bits, based on the range information. The addition of the conversion error and the new data causes a carry, which may cause a so-called overflow in which "1" shifts to a bit higher than the extracted bit specified by the range information.

When this overflow occurs, there is a problem that a DC error component occurs during reproduction and good reproduction cannot be performed.
Usually, in order to prevent this, in the weighters 12 and 13, the maximum absolute value in the block through the primary filter and the secondary filter is weighted by multiplying the maximum absolute value in the block by, for example, 1.5, for example, by one bit rather than the bit to be originally extracted Data is extracted from the higher bit.

However, even if the above weighting is performed, an overflow may occur.

Therefore, when the overflow occurs, the noise shaping circuit 19 sends the switch 2, the buffer 3, the mask circuit 21 and the read control circuit 22 to the overflow amount and the current filter number (0-order, primary or secondary filter). The information is supplied, and the output of the audio PCM signal in which the overflow has occurred is temporarily stopped.

When the overflow information is supplied, the multiplexer 17 stops outputting data, the read control circuit 22 stops reading data from the audio data generation circuit 1, and the switch 2 switches the selection terminal 2c from the selection terminal 2a to the selection terminal 2a. Switch to the selection terminal 2b and temporarily switch to the audio data generation circuit 1
Stop taking data from. As described above, several blocks of audio data are stored in the buffer 3, and the buffer 3 stores the audio PC of the block in which the overflow has occurred according to the overflow information.
A so-called retry of supplying the M signal to the predictor 4 again is performed.

When the retry is started, the mask circuit 21 knows the current filter number from the supplied overflow information, and supplies a coefficient increment signal to a weighter connected to the filter. The weighter supplied with the coefficient increment signal has a coefficient value of 1.5
Is increased, for example, to 1.6. The re-supplied audio PCM signal of the block in which overflow has occurred is supplied to each of the range detectors 9, 10, and 11 through the respective filters as described above, and the maximum absolute value in the block is detected. Is supplied to the selector 15. Among the detected maximum absolute values in the block, the maximum absolute value in the block of the straight PCM signal via the range detector 9 is:
The maximum absolute value in the block of the primary difference data via the range detector 10 is weighted by the weighter 12 and the maximum absolute value in the block of the secondary difference data via the range detector 11 is weighted. Is supplied to the vessel 14. As described above, the coefficient value of one of the weighters that caused the overflow is incremented, and the maximum absolute value in each of the blocks is multiplied by the coefficient value and weighted and supplied to the comparison block 14. . The comparison block 14 compares the maximum absolute value in each block including the maximum absolute value in the block re-weighted by the incremented coefficient to detect the one with the smallest value, and based on the detection result, The range information is supplied to the quantizer 18. Based on this range information, the quantizer 18 performs requantization as described above. When the overflow is eliminated by performing the requantization, the data in which the overflow has been eliminated by the requantization is output, and the mask circuit 21 calculates the coefficient value based on the incremented coefficient value of the weighting device. And the switch 2 switches the selection terminal 2c to the selected terminal 2a and restarts the fetching of new data.

Note that this retry is performed until the overflow is eliminated, and if an overflow occurs despite the retry, the weighting coefficient value is incremented again.

The 4-bit data thus requantized is supplied to the multiplexer 17.

The multiplexer 17 outputs the supplied data and the above-described range information and filter information. These data and information are taken out via the output terminal 20.

As output data taken out from the output terminal 20, one block, for example, is as shown in FIG. 2. One byte of header information (parameter information related to compression or attached information) RF and eight bytes of sample data Data D _{A0 to} D _B3 . The header information RF is 4
Bit range information, 2-bit mode selection information or filter selection information, 1-bit two flag information, for example, information LI indicating the presence or absence of a loop, and whether or not the end block (end block) of the waveform. And the information EI shown. As described above, the data of one sample is compressed from, for example, 16 bits to 4 bits, and the data D _{A0 to} D _B3 include the 4-bit data D _{A0H to} D _{B3L for} 16 samples.

Then, such data and each information are recorded on a recording medium or the like, or transmitted directly to a decoding side to be reproduced.

As is clear from the above description, the data compression encoding apparatus according to the present embodiment divides an input signal into blocks, for example, every 16 samples to form a zero-order filter 5, a primary filter 6, and a secondary filter 7.
When the filter which can obtain the highest compression rate is selected from the above and re-quantized and the output signal is transmitted, the counter 8
By counting the number of blocks input in step (1), when a predetermined count value is reached, the coefficients of the weighting units 12, 13 provided in the primary filter 6 and the secondary filter 7 are set to predetermined large values. Since the 0th order filter is forcibly selected and the straight PCM signal can be arranged,
In addition to being able to reproduce data from any point, errors occurring during reproduction can be eliminated by arranging the straight PCM signals at the above-mentioned fixed intervals forcibly. Filters can be used.

Further, in a data compression encoding apparatus that changes the weighting parameter for each fixed block and indirectly controls the filter selection, the data compression encoding according to the present invention can be performed by only slightly changing such a system. Can be changed to device.

Further, the data compression encoding apparatus according to the present embodiment stops the output and the input when the overflow is detected in the noise shaping circuit 19, increments the coefficient values of the weighters 12 and 13, and detects the overflow again. By performing the retry for processing the data of the block which has been processed, data in which an overflow has occurred is not supplied, so that it is possible to prevent the SN ratio from deteriorating during reproduction.

Note that the coefficient value of the weighting device is only an example, and various changes are possible, for example, the base coefficient is set to 1.4 and the increment at the time of retry is set to 1.6, and it is needless to say that the value is not limited to the above value. It is.

〔The invention's effect〕

A data compression encoding device according to the present invention is a data compression encoding device that blocks an input signal into fixed samples, performs requantization based on range information corresponding to the maximum value of each block, and performs noise shaping processing. By controlling the range information when an overflow occurs due to the noise shaping processing in the decoding device, the overflow can be eliminated, and the data in which the overflow has occurred is not supplied to the decoding side. In addition, it is possible to prevent the deterioration of the S / N ratio in the same manner as described above by using a value obtained by weighting the maximum value in the block as the range information with a predetermined weight, and increasing the weight value at the time of the overflow. The effect of can be obtained.

[Brief description of the drawings]

FIG. 1 is a functional block diagram showing functions of a data compression encoding apparatus according to the present invention in a block diagram, and FIG. 2 is a schematic diagram showing a format of output data of the data compression encoding apparatus according to the present invention. . 1 ... audio data generation block 2 ... switch 3 ... buffer 4 ... predictor 5 ... 0th order filter 6 ... primary filter 7 ... secondary filter 8 ... counter 9, 10, 11 ... range Detector 12, 13 Weighting device 14 Comparison (minimum value detection) block 155 Selector 16, 17 Output terminal 18 Quantizer 19 Noise shaping circuit 20 Output terminal 21 Mask Circuit 22 …… Readout control circuit

Continuation of front page (58) Field surveyed (Int.Cl. ⁷ , DB name) H03M 7/32

Claims (2)

(57) [Claims] 1. A data compression encoding apparatus wherein an input signal is divided into blocks at a fixed number of samples, and quantization and noise shaping are performed based on range information corresponding to the maximum value of each block. A data compression encoding device for controlling the range information when an overflow occurs due to a shaping process. 2. The data compression according to claim 1, wherein a value obtained by weighting the maximum value in the block with a predetermined weight is used as the range information, and the value of the weight is increased at the time of the overflow. Encoding device.