JPH0818518A - Data compression device with bit rate changeover function by dsp and its method - Google Patents

Abstract

PURPOSE:To obtain the data compression device with bit rate changeover function by a DSP and its method in which one DSP copes with three kinds of bit rate and plural input and output channels are supported. CONSTITUTION:A 1st changeover means 7 and a 2nd changeover means 8 select input output channels IN1, IN2, IN3, IN4 and OUT1, OUT2, OUT3, OUT4 respectively. Furthermore, an internal program to compress PCM data in 64kbit/ s into data with the bit rate of 32kbit/s, 24kbit/s and 16kbit/s is incorporated into the DSP1 and any bit rate is freely selected by using any of control commands 10a, 10b, 10c and 10d externally.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a method for compressing data such as voice, and more particularly to D for compressing voice data using DSP1.
The present invention relates to a data compression method and device with a bit rate switching function by SP.

[0002]

2. Description of the Related Art A technique for deleting the amount of data by removing the redundancy of the data in order to reduce the size of the file for storing the data and the time for sending the data on the communication line, that is, The data compression technology includes run length coding for compressing consecutive identical characters or bit strings, Huffman coding for assigning short internal codes to frequently appearing characters and long codes for low letters, and n / 2n. There are encoding, a method of expressing frequently appearing patterns with symbols, and removal of meaningless character strings and bit strings. Conventionally, a general-purpose microprocessor is used to perform arithmetic processing as a means for achieving this. The method was often used.

However, recently, with the advent of DSP 1, a data compression method and apparatus using this IC have been proposed. The DSP (D igital S i
The gnal P rocessor) 1, a processor of the digital signal processing only, and performs FFT (Fast Fourier Transform) or digital signal processing, high speed an order of magnitude or more compared with a general-purpose microprocessor. Therefore, the DSP
In No. 1, a unique circuit configuration is adopted to process the repetition of multiply-accumulate operations that are frequently used in digital signal processing at high speed. A hardware multiplier and an adder (ALU) are connected in pipeline to operate in parallel. Thus, the product-sum operation is processed in one cycle. Further, in order to perform the multiply-accumulate operation continuously every cycle, it is necessary to load instructions and transfer operands (data) in one machine cycle.
Therefore, the DSP 1 has a built-in RAM / ROM of several K words for instructions and data. Addressing also has a mode suitable for digital signal processing, for example, FFT.
In order to send data to the arithmetic unit without interruption during arithmetic operation, a modulo method is adopted in which the address of a certain area is circulated. The data length to be handled is 16, 24, 32 bits or the like.

FIG. 2 shows the basic configuration of the DSP 1. 2 in the figure
Reference numerals 1 and 22 are an instruction decoder and an address modifier for reading, decoding and executing an instruction. Further, 23 is a memory for storing internal data and programs, 24 is an arithmetic operation unit having a high-speed multiplier and an adder (ALU), and 25 is an input / output interface for performing input / output with the outside. Are interconnected by.

[0005]

When the DSP 1 is used in the field of communication, the PBX is used.
In order to perform data compression by a device such as a (private branch exchange), 32,24,16 Kbit / S depending on the connecting medium such as voice mail box, ISDN line, and wireless communication system.
Since the bit rates for compression differ, it is necessary to compress to multiple bit rates in order to deal with these, but in the past, the data compression program built into the DSP 1 has only one compression rate. Therefore, one DSP1 can support only one bit rate,
Also, since only one channel is supported, these DSPs are used by being allocated to each bit rate, and even if the same bit rate exists, if there are multiple DSP for that number
1 may have to be used, and the use of a large number of expensive DSPs 1, which are still expensive, complicates the device and raises the price. And one bit rate with one DSP1,
Since it can support only one channel, its design and usage were limited.

The present invention has been made in view of the above circumstances, and an object thereof is to allow one DSP 1 to freely select a plurality of types of bit rates, and
It is an object of the present invention to provide a data compression method and device with a bit rate switching function by DSP capable of supporting a plurality of channels.

[0007]

To achieve the above object, the present invention provides a plurality of data compression methods and devices with a bit rate switching function by a DSP for voice compression of a voice mailbox, an ISDN line and a wireless communication system. Input channels IN1, IN2, IN3, I
Multiple codecs 3, 4, 5, connected for each N4
First switching means 7 for switching 6 to input to the DSP 1, and 3 PCM data of 64 Kbit / S each.
2Kbit / S, 24Kbit / S, 16Kbit / S
D having an internal program for compressing each bit rate of
SP1 and second switching means 8 for switching the output from the DSP1 for each of the output channels OUT1, OUT2, OUT3, OUT4 are provided, and the first and second switching means 7, 8 are controlled by an external control command. Input channels IN1, IN2, IN3, IN4 operated by 10d, 10c
And output channels OUT1, OUT2, OUT3
OUT4 is switched, and a program corresponding to each bit rate inside the DSP1 is externally supplied with a control command 10b.
One DSP1 supports a plurality of input / output channels and is compatible with three kinds of bit rates by operating and switching.

[0008]

The first switching means 7 and the DSP 1 and the second switching means 8 are connected by the bus line 2, and the control commands 10a, 1 are connected by the bus line 2.
0b, 10c, 10d and each data 11a, 11b, 12
a, 12b, 13a, 13b, 13c, 13d are given and received. The first switching means 7 has a plurality of input channels IN.
1, a plurality of codecs 3, 4, 5, 6 connected to IN2, IN3, IN4 are connected, and the output of these codecs 3, 4, 5, 6 is switched by a control command 10d from the outside to obtain 64 Kbit / S. PCM data 11
a and 11b are sent to the DSP1. Inside the DSP1, 64Kbit / S PCM data is 32 and 2 respectively.
There are programs for compressing compressed data of 4,16 Kbit / S, and which program is to be started can be selected by an external command 10b. Then, the DSP 1 sets the bit rate selected by the control command 10d to 64 Kbit / S PC.
Compress M data 11a, 11b to 32, 24, 16K
The bit / S compressed data 12a and 12b are sent to the second switching means 8 via the bus line 2. And
The second switching means 8 is 32 for each compressed bit rate.
Kbit / S Compressed data 13a, 24 Kbit / S
Compressed data 13b, 16Kbit / S Compressed data 13
c and 13d are switched respectively and sent to the corresponding output channels OUT1, OUT2, OUT3 and OUT4.

As a data compression method, for example, CCIT
Recommendation G.T. 32Kbit / S AD according to 721
There is a compression method by the PCM method. Since the level of the audio signal changes with time, it is desirable to quantize while changing the step size according to the level of the audio level, rather than quantizing with a fixed quantization unit when quantizing. , If the quantization is performed with a constant step size, the number of bits will be insufficient for a large voice signal, and the overload will cause distortion of the sound quality. on the other hand,
Human hearing, which is perceived as harsh noise (quantization noise) due to coarse quantization for small signals, is insensitive to quantization noise for loud sounds, but sensitive to small sounds. There is a property of becoming. Therefore, it is understood that a large signal may be quantized with a large step size, and a small signal may be quantized finely, and such quantization is called adaptive quantization. In addition, unlike a random data signal, a voice signal has a strong correlation between the sampled voice signal and the sampled voice signal, and therefore it is possible to use this to encode, and this correlation is called proximity correlation. And A
In the DPCM method, a predictor is used to estimate the amount of close correlation and to perform coding. By incorporating the arithmetic program by the data compression method into the internal memory of the DSP 1, the data is compressed at a predetermined compression rate.

[0010]

DESCRIPTION OF THE PREFERRED EMBODIMENTS Next, one embodiment of the present invention will be described with reference to the drawings. FIG. 1 is a diagram showing the basic configuration of an embodiment of the present invention. In the figure, 1 is a DSP, which may be any general-purpose DSP, 2 is a bus line,
The CPU 9 and the DSP 1 or the first or second switching means 7, 8 are connected. Also IN1, IN2, IN
Reference numerals 3 and IN4 denote first to fourth input channels, and the first to fourth input channels IN1 and I
For example, a telephone is connected to N2, IN3, and IN4, and a voice analog signal is input. CODECs 3, 4, 5 and 6 are for converting analog signals from the first to fourth input channels IN1, IN2, IN3 and IN4 into digital PCM signals, which are usually one-chip ICs. It is composed by. A first switching means 7 switches the PCM data from each of the codecs 3, 4, 5 and 6 by the control command 10d and sends it to the DSP 1. In practice, for example, each of the codecs 3,
This is done by giving a special control signal to the ICs of 4, 5 and 6 to control their outputs. Reference numeral 8 is a second switching means for switching the compressed data from the DSP 1 by a control command according to each compression rate or the like, and the first to fourth output channels OUT1, OUT2, OUT3, OUT.
4 and is composed of, for example, a gate array. A CPU 9 sends out a control command 10a for controlling the DSP 1, the first switching means, and the second switching means.

In this embodiment, however, the input channels are the first to fourth input channels IN1, IN2.
There are four input channels IN3 and IN4, and codecs 3, 4, 5 and 6 are connected to the first to fourth input channels IN1, IN2, IN3 and IN4, respectively, and analog signals based on the input voice are provided. 64 Kbit / SPC
It is designed to convert to M data signals, and
The outputs of the codecs 3, 4, 5 and 6 are connected to the first switching means 7, and the first switching means 7 is controlled by the control command 10d from the bus line 2 to the respective codecs 3, 4, 5 and 6. 64Kbit / S by switching 6 outputs
The PCM data 11a and 11b are sent to the DSP 1 via the bus line 2.

As described above, the DSP 1 has a program memory inside so that an arithmetic program can be built therein. Therefore, 32 Kbit / S and 24 Kbit are stored in advance.
A program for compressing each bit rate of / S and 16 Kbit / S is written in advance so that which program is started can be selected by a command from the outside.

Thus, 64K sent to the DSP1
The bit / S PCM data 11a, 11b is the DS selected by the control command 10b from the bus line 2.
32, 24, 16 Kbit / S compressed data 1 which is compressed to a predetermined compressed data by an arithmetic program inside P1
2a and 12b are sent to the second switching means 8 via the bus line 2.

The second switching means 8 has the first to fourth parts.
Output channels OUT1, OUT2, OUT3, OU
T4 is connected to each output channel OUT1, O
UT2, OUT3, OUT4, for example, voice mail box 1, voice mail box 2,
The wireless communication device and the ISDN B channel are respectively connected. And voicemail box 1 has 32
Kbit / S, 24 Kbi in voice mail box 2
t / S, 16 Kbit / S for wireless communication device, ISDN
Since 16 Kbit / S compressed data is used for the B channel, the second switching means 8 transfers the data designated by the control command 10c from the bus line 2 to the first output channel OUT1 at 32 Kbit / S compressed data. 13a, 24Kbi on the second output channel OUT2
t / S compressed data 13b, third output channel OUT
3 has 16 Kbit / S compressed data 13c, and fourth output channel OUT4 has 16 Kbit / S compressed data 1
3d, each output channel OUT1,
The output is switched to OUT2, OUT3, and OUT4.

A voice mail box is a system that provides various services by filing and storing human voice messages. For example, input voice messages are temporarily stored in a disk in the system,
This voice message can be transferred to a designated telephone or the like, or a message addressed to oneself can be called and heard from the telephone of the receiving side. Then, in order to reduce the amount of data stored in the voice mail box, the voice message is compressed into, for example, PCM data of 32 Kbit / S.

The ISDN B channel is a channel for transmitting information such as voice and is usually 64 Kbit /
S PCM data can be transmitted, but in this embodiment 16K
Since the voice data is compressed to bit / S, a plurality of voice data or voice data and other data can be sent together.

The CPU 9 sends a control command 1 to the bus line 2.
0a is sent to control the DSP, the first switching means 7, and the second switching means 8. Generally, a one-chip microprocessor mounted in the same device is used. It is good to give a command from the computer.

[0018]

As described above, according to the present invention, a program for switching input / output channels by the first switching means 7 and the second switching means 8 and compressing into three bit rates is built in the DSP 1. Since it is possible to select from the outside, one DSP1 can compress to 3 kinds of bit rates, and it is possible to support multiple channels, eliminating the need to use many expensive DSP1 and reducing the number of parts. Since the mounting area is reduced, the cost is reduced, and since there are three types of bit rates of the output data, the flexibility in design and use is increased.

[Brief description of drawings]

FIG. 1 is a diagram showing a basic configuration of an embodiment of the present invention.

FIG. 2 is a diagram showing a basic configuration of a DSP.

[Explanation of symbols]

 1 DSP 2 Bus line 3, 4, 5, 6 Codec 7 First switching means 8 Second switching means 9 CPU 10a, 10b, 10c, 10d Control command 11a, 11b 64Kbit / S PCM data 12a, 12b 32, 24 , 16Kbit compressed data 13a 32Kbit / S compressed data 13b 24Kbit / S compressed data 13c, 13d 16Kbit / S compressed data 21 instruction decoder 22 address modification 23 memory 24 arithmetic operation unit 25 input / output interface 26 common bus 27 external bus

─────────────────────────────────────────────────── ─── Continuation of the front page (51) Int.Cl. ⁶ Identification number Office reference number FI technical display location H03M 7/38 9382-5K

Claims (1)

[Claims] 1. A data compression method and device with a bit rate switching function by a DSP for voice compression of a voice mail box, an ISDN line and a wireless communication system, wherein a plurality of input channels (IN1, IN2, IN3) are provided.
IN4) connected to a plurality of codecs (3, 4, 5,
The first switching means (7) for switching 6) and inputting to the DSP (1) and the PCM data of 64 Kbit / S are 32 Kbit / S, 24 Kbit / S and 16 K, respectively.
A DSP (1) having an internal program for compressing each bit rate of bit / S and an output from this DSP (1) is output channel (OUT1, OUT2, OUT3,
A second switching means (8) for switching each OUT4), and operating the first and second switching means (7, 8) by a control command (10d, 10c) from the outside to input channel (IN1). , IN2, IN3, IN4) and output channels (OUT1, OUT2, OUT3, OUT4)
To switch the program corresponding to each bit rate inside the DSP (1) from the external control command (10b).
A data compression method with a bit rate switching function by a DSP, wherein a single DSP (1) supports a plurality of input / output channels and is compatible with three types of bit rates. And equipment