JPH03226017A - Code conversion automatic changeover control system - Google Patents

Abstract

PURPOSE:To attain the correct code conversion in compliance with the companding rule by detecting a nonsound signal in a nonlinear code signal with a nonsound detection section and discriminating the A rule or the mu rule depending whether 2nd-4th bits of the nonsound signal are all 0 or all 1 so as to switch the code conversion. CONSTITUTION:A discrimination section 15 discriminates the rule to be the A rule when 2nd-4th bits of a nonlinear code signal fed as a nonsound signal, that is, segment bits SG are all '0' and discriminates the rule to be the mu rule when the 2nd-4th bits are all '1' and gives its discrimination signal to selectors 19, 25. Then the selector 19 selectively outputs an output signal of a code conversion section 17 by using a discrimination signal discriminated to be the A rule at the discrimination section 15 and selectively outputs an output signal of a code conversion section 18 by using a discrimination signal discriminated to be the mu rule at the discrimination section 15.

Description

[Detailed Description of the Invention] [Summary] Regarding a code conversion automatic switching control method that converts a non-linear code signal into a linear code signal by automatically switching to correspond to the companding law between A-law and μ-law. The purpose is to determine whether the companding law is the A-law or the μ-law, and to perform correct code conversion compatible with the companding law. A silence detector detects a silence signal, and determines whether the silence signal is A law or μ law based on the silence signal detected by the silence detector, and whether the upper second to fourth bits of the nonlinear code signal are all "0" or All '°1''
and a code conversion unit that converts the non-linear code signal according to either the A law or the μ law into a linear code signal, It is configured to switch between A-law and μ-law.

[Industrial application field]

The present invention relates to an automatic code conversion switching control system that automatically switches to correspond to companding laws between A-law and μ-law and converts a nonlinear code signal into a linear code signal.

2. Description of the Related Art Systems are known in which an analog audio signal is converted into a 13- to 16-bit linear code signal, which is compressed according to the A-law or μ-law companding law and then transmitted. In addition, in the case of echo cancellation processing, conference call processing, etc., a non-linear code signal that follows the companding law is converted to a linear code signal,
Arithmetic processing such as addition will be performed.

In such a system, compared to the case where a device dedicated to A-law and a device dedicated to μ-law are manufactured,
In the case of a device in which parts for regular use are switched and used, costs can be reduced because devices with the same configuration can be manufactured. In that case, it is necessary to switch without error in accordance with the A-law and the μ-law.

[Conventional technology]

FIG. 4 is a block diagram of main parts of a conventional example, in which 40 is a signal processing circuit, 41 is a compression encoding section, 42 is a code conversion section, 4
3 is a signal processing section, 44 is a timing signal generation section, and 45 is a changeover switch. A synchronization signal SYN and a clock signal CLK are applied from the timing signal generation section 44 to the compression encoding section 41 and the code conversion section 42, and the compression encoding section 4
At step 1, the audio analog signal is compressed and encoded to become a non-linear code signal NLC, which is applied to a code converter 42 for signal processing and is converted into a linear code signal LC.

This code converter 42 converts the 8-bit non-linear code signal NLC into, for example, a 16-bit linear code signal LC, and when the changeover switch 45 is turned on, the A law is used, and when it is turned off, the μ law is used. code conversion.

The audio signal converted into the linear code signal LC is applied to a signal processing unit 43 such as an echo canceller or a conference telephone trunk, and is processed therein.

[Problem that the invention seeks to solve]

The code converter 42 turns on the changeover switch 45.

The code conversion is switched to A-law and μ-law according to the off state, and the non-linear code signal NLC input from the compression encoding unit 41 directly or via a transmission path to the signal processing circuit 40 is based on the A-law. A maintenance person, etc., determines whether the changeover switch 4 is
5 was operated manually.

If the changeover switch 45 is operated incorrectly, the signal cannot be converted into a correct linear code signal, so the signal processing section 45
3, the audio signal cannot be processed correctly.

An object of the present invention is to determine whether a nonlinear code signal is A-law or μ-law, and perform correct code conversion compatible with companding law.

[Failure to solve the problem]

The code conversion automatic switching control system of the present invention uses the characteristics of the upper bits of a nonlinear code signal to determine whether it is A-law or μ-law, and will be explained with reference to FIG. .

A silence detection section 1 detects a silence signal included in a nonlinear code signal compressed according to either the A-law or the μ-law. Whether the non-linear code signal is determined by whether the upper 2nd to 4th bits are all “0” or all °“
1', and a code converter 3 that converts a non-linear code signal of either A-law or μ-law into a linear code signal. This is used to switch between the A-law and the μ-law.

[Effect]

Silence signals in nonlinear code signals compressed according to the companding law are largely different between the A-law and the μ-law. That is, the most significant bit indicates the polarity, and the second to fourth bits indicate the polarity.
The bits are all "0"' in the case of the A law, and all "1" in the case of the μ law. Therefore, the silence detection section 1 detects a silence signal in the nonlinear code signal, and the bits are sent to the determination section 2. The second to fourth bits of the silence signal are all “0”.
” or all °゛1゛, it is determined whether the law is A law or μ law, and based on the determination result, the code conversion section 3 switches the code conversion to convert from a non-linear code signal to a linear code signal. It is something to do.

[Examples] Examples of the present invention will be described in detail below with reference to the drawings.

FIG. 2 is a block diagram of an embodiment of the present invention, in which 10 is a signal processing circuit, 11 is a compression encoding unit that compresses and encodes an audio analog signal according to A law or μ law, 12 is an echo canceller, and 13 is a serial a parallel conversion circuit (S/P); 14 is a silence detection unit; 15 is a determination unit that determines whether the nonlinear code signal is based on the A law or the μ law;
16 is a timing signal generator, and 17 is a code converter tH that converts an A-law nonlinear code signal into a linear code signal.
B (A/L), 1 B is a coat conversion unit (μ/L) that converts a μ-law nonlinear code signal into a linear code signal.
), 19 is a selector, 20.21 is a register, 22 is a buffer memory, and 23 is a code conversion unit (L/A) that converts a linear code signal into a non-linear code signal according to the A law.
, 24 is a code converter (L/μ) that converts a linear code signal into a non-linear code signal according to μ law, and 25 is a selector.

This embodiment shows a case where an audio signal is processed by an echo canceller 12, and an audio analog signal is applied to an input terminal AIN of a compression encoder 11, and a synchronization signal SYN and a clock signal CLK are supplied from a timing signal generator 16. and is converted into an 8-bit non-linear code signal.

This nonlinear code signal corresponds to a compressed 13 to 16 bit linear code signal, and is applied to the signal processing circuit 10 as shown in the figure, or is applied to a digital exchange via a transmission line and exchanged. Ru.

FIG. 3 is an explanatory diagram of code conversion, where (a) is a synchronization signal,
(b) shows a clock signal, (C) shows a non-linear code signal, and (d) shows a linear code signal with a 16-bit configuration.

In the non-linear code signal, the most significant focus MSB is the polarity bit, the second to fourth bits are the segment bit SG, and the fifth bit is the polarity bit.
~The 8th (LSB) bit becomes the step knob bit STP, the 3-bit segment bit SG specifies any segment among the 8, and the 4-bit stump bit STP specifies any step among the 16 of each segment. is specified.

Such a non-linear code signal is converted into serial-to-parallel converter circuit 1.
3 into an 8-bit parallel signal, and the silence detection section 1
4 and code converters 17 and 18. The silence detection section 14 detects a silence signal of the nonlinear code signal and adds it to the determination section 15. This can be done by setting 4 bits, ie, segment bits SG, to be all "1" or all '0'.

The determining unit 15 determines the second to fourth bits of the nonlinear code signal added as a silent signal, that is, segment bits SG.
When all are "O", it is determined to be the A law, and when all are "1", it is determined to be the μ law, and the determination signal is sent to the selector 19.25.
Add to.

Therefore, the selector 19 selects and outputs the output signal of the code conversion section 17 based on the determination signal determined by the determination section 15 to be the A-law, and also by the determination signal determined by the determination section 15 to be the μ-law. The output signal of the code converter 18 is selectively output. That is, 8 shown in FIG. 3(C)
The non-linear code signal with a bit structure is (1 shown in dJ)
The signal is converted into a 6-bit linear code signal.

This 16-bit linear code signal is stored in the register 20.
is read out by the output enable signal OE from the echo canceller 12, and the echo canceller 12
The input data DI becomes the input data DI, echo cancellation processing is performed according to the frame signal F from the timing signal generator 16, and the output data DS is added to the register 21 and written according to the write enable signal WE. The signal is added from this register 21 to a buffer memory 22 and read out as a linear code signal LC in accordance with an enable signal EN to a device (not shown). Alternatively, it is applied from the register 21 or the buffer memory 22 to the code conversion sections 23 and 24, and is converted into an 8-bit non-linear code signal according to the A law or μ law, and the selector 25 is controlled according to the judgment signal of the judgment section 15, In the case of the system, the nonlinear code signal NL converted by the code converter 23
In the case of a μ-law system, a nonlinear code signal NLC converted by the code converter 24 is output.

Therefore, the audio signal processed by the echo canceller 12 can be sent to another device or transmission path as a linear code signal LC or nonlinear code signal NLC.

In the embodiment described above, the code converter 1718 converts the 8-bit nonlinear code signal into 16 bits in parallel.
The determination unit 15 converts it into a linear code signal with a binoto configuration.
The case is shown in which the selector 19 is controlled by the determination signal of the determination section 15, but when the determination signal of the determination section 15 indicates the A law, the code conversion section 17 is operated, and when it indicates the μ law, the code conversion section 18 is operated.
It is also possible to switch to operate. Similarly, the code converters 23 and 24 can be switched.

〔Effect of the invention〕

As explained above, the present invention detects a silence signal of a non-linear code signal by the silence detector l, and determines the μ rule when the second to fourth bits of the non-linear code signal are all "1". , it is determined that the code is A-law when all "O"s are present, and the code conversion for converting from a non-linear code signal to a linear code signal in the code converter 3 is switched based on the determination result, and whether it is A-law or μ-law. Since it is possible to automatically switch the code conversion in the code converter 3 by determining whether there is a code, even if A-law and μ-law nonlinear code signals are mixed, switching can be performed without error. There are advantages.

[Brief explanation of drawings]

FIG. 1 is a diagram illustrating the principle of the present invention, FIG. 2 is a block diagram of an embodiment of the present invention, FIG. 3 is a diagram illustrating code conversion, and FIG. 4 is a block diagram of main parts of a conventional example. 1 is a silence detection section, 2 is a determination section, and 3 is a code conversion section.

Claims (1)

[Claims] A silence detection unit (1) that detects a silence signal included in a non-linear code signal compressed according to either the A-law or the μ-law companding law; and the silence detection unit (1). Based on the silence signal detected by A
Whether it is the law or the μ law, the upper second to
a determination unit (2) that determines whether the fourth bit is all “0” or all “1”; and a code conversion unit (2) that converts the non-linear code signal of either A-law or μ-law into a linear code signal. 3) A code conversion automatic switching control system, characterized in that the code conversion section (3) switches between the A law and the μ law according to the determination result by the determination section (2).