JPH0124381B2 - - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION The present invention relates to a transmission characteristic measuring device for measuring the transmission characteristics of a digital telephone line.

Conventional digital telephone line transmission characteristic measurement equipment measures the transmission characteristics by converting the input signal under test into an analog signal, filtering it, and detecting it, but it has the following drawbacks. It was hot.

B. Because the measurement circuit consists of analog elements, the parts used deteriorate due to temperature changes, aging, etc., making stable measurements impossible.

Subtle adjustments were required to the filter, detection circuit, etc.

C. It was not possible to create a highly accurate device due to noise within the device.

(2) Each type of filter requires separate hardware (inductors, capacitors, etc.), and since it is used in the low frequency audio band, the shape of the element becomes large, which directly leads to an increase in the size of the device.

For level measurement, a highly accurate linear detection circuit was required, and for noise measurement, a square law detection circuit, which required relatively little accuracy, was required independently.

F. Because the measured value needs to be displayed in decibels, the conversion method for converting the measured value into a BCD (binary coded decimal code) decibel value is complicated.

G. It was difficult to ensure accuracy due to errors that occurred during the conversion process from digital to analog.

Furthermore, the above-mentioned drawbacks will be explained in detail below with reference to FIG.

Figure 1 is a configuration diagram of a conventional transmission characteristic measurement device.
This measures the signal level of the PCM line, noise during no calls, and noise during calls. In the figure,
is an input terminal, 1 is a code conversion circuit, 2 is a DA conversion circuit, 3 is a low-pass filter, 4, 5, 6, 10,
11 and 14 are switches, 7 is a band removal filter,
8 is a channel filter, 9 is an evaluation filter, 1
2 is a linear detection circuit, 13 is a square law detection circuit, 15 is a hold circuit, 16 is a resistance attenuation circuit, 17 is a comparison circuit, 18 is a reference voltage generation circuit, 19 is an attenuation value designation circuit, and 20 is a display circuit.

First, a level measurement operation for measuring the signal level of a PCM line will be explained. In advance, switch 4 is set to a, switch 5 is set to c, and switch 6 is set to e.
, the switch 10 is set to g, and the switch 14 is set to k. When a PCM signal is applied to the input terminal X, the PCM signal is converted from a non-linear code to a linear code digital value by the code conversion circuit 1,
The digital value of the linear code is converted into a step-wave analog signal by a DA conversion circuit 2 that converts a digital amount into an analog amount, and further converted into an analog signal having a frequency only in the voice band by a low-pass filter 3.

Note that the operation of converting a PCM signal into an analog signal is the same for the following measurement of no-call noise and talk noise. Now, the converted analog signal is applied to a channel filter 8 via switches 4, 5, and 6 to limit the band, and the band-limited signal is applied to a linear detection line 12 via a switch 10 to provide a linear detection line. Detected. When measuring signal levels, a linear detection circuit is used to ensure measurement accuracy. The detection output signal of the linear detection circuit 12 is applied to the hold circuit 15 via the switch 14, converted into a BCD decibel value by the method described below, and displayed on the display circuit 20 as a decimal decibel value.

First, the detected output signal is sent to the hold circuit 15 as described above.
It is retained until it is converted to a BCD decibel value.
This held signal is attenuated by the value specified by the attenuation value designation circuit 19 in the resistance attenuation circuit 16, and this attenuated value is compared with the reference voltage value output from the reference voltage generation circuit 18 in the comparison circuit 17. be done. When the attenuated value and the reference voltage value match, the value designated by the attenuation value designation circuit 19 is
This is a BCD decibel value, but if they do not match, the value specified by the attenuation value designation circuit 19 is successively changed until they match. In this way, with conventional transmission characteristic measurement equipment, the detection output is
The equipment used to convert it to BCD decibel values was complicated and large.

Next, a description will be given of the operation of measuring noise during no-call time on a PCM line. In advance, switch 4 is set to a, switch 5 is set to c, switch 6 is set to f, and switch 1 is set to
1 is set to j, and switch 14 is set to l.
The no-call noise of the PCM signal applied to the input terminal (called "analog noise"). The analog noise is applied to the evaluation filter 9 via switches 4, 5, and 6, where the evaluation function is weighted, and then applied to the square law detection circuit 1 via the switch 11.
3 is applied. No-call noise power is outputted from the square law detection circuit 13, and the no-call noise power is applied to the hold circuit 15 via the switch 14. Hold circuit 15, resistance attenuation circuit 16,
The comparison circuit 17, reference voltage generation circuit 18, and attenuation value specification circuit 19 convert the power of the no-call noise into BCD.
This display operation, which is converted into a decibel value and displayed on the display circuit 20, is the same as in the level measurement described above.

Next, we will explain the measurement operation of noise during a PCM line call. Generally speaking, noise during a call is determined by transmitting a single frequency signal, measuring the signal power and the power of signals other than the signal components within the voice band, and calculating the ratio. Since the signal power measurement operation is the same as the level measurement operation described above, the measurement operation of power other than signal components within the audio band will be described here.
In advance, switch 4 is set to b and switch 5 is set to d.
The switch 6 is set to e, the switch 10 is set to h, the switch 11 is set to i, and the switch 14 is set to l. The PCM signal applied to the input terminal X is converted into an analog signal by the code conversion circuit 1, the DA conversion circuit 2, and the low-pass filter 3 in the same manner as in the above-mentioned level measurement and no-call noise measurement. Here, the analog signal is applied to a band elimination filter 7 via a switch 4, and signal components are removed. Therefore, the output of the band elimination filter 7 is noise during a call. The noise during a call is applied to a channel filter 8 via switches 5 and 6, where the band is limited, and applied via switches 10 and 11 to a square law detection circuit 13, which outputs noise power during a call. .

The noise power during a call is applied to the hold circuit 15 via the switch 14, converted into a BCD decibel value by the resistance attenuation circuit 16, comparison circuit 17, reference voltage generation circuit 18, and attenuation value designation circuit 19, and displayed in the display circuit 20. is displayed as a decimal decibel value.

Since the above-mentioned measurement circuit is usually constructed of analog elements, the measurement results are subject to temperature changes and changes over time, and it is also necessary to adjust the resistance, inductance, capacitance, etc. of the circuit. In other words, measurement performance is determined by the quality of the adjustment, so delicate adjustments were required. Furthermore, measurement accuracy is determined by internal noise of the device, such as power supply noise, signal leakage between wires, signal leakage between coils, etc., making it difficult to manufacture a highly accurate measuring device. Furthermore, the measurement circuit requires four filters: a low-pass filter, a band rejection filter, a channel filter, and an evaluation filter, but since the frequency band used by these filters is the audio frequency, the coils and capacitors used in the filters are large. As a result, the filters have also become larger, resulting in larger transmission characteristic measuring devices. As can be seen from the above explanation, the circuit for converting the measured signal level or noise power into a BCD decibel value has a complicated circuit configuration and is large in size. In addition, since the signal level must be measured with high precision, it is impossible to share a square law detection circuit for noise power measurement, which has relatively low accuracy requirements, so a linear detection circuit is used. This has caused the inconvenience of having two detection circuits in one transmission characteristic device. Furthermore, although PCM signals are originally digital signals, they are converted to analog signals, their transmission characteristics are measured, and then converted back to digital signals for display, which is an irrational process.
There were extra circuits: a circuit to convert PCM signals to analog signals and a circuit to convert measured values to digital values. Additionally, errors occurring during this conversion process made it difficult to ensure accuracy.

The present invention was devised to eliminate these drawbacks, and its purpose is to provide a highly stable, highly accurate, and compact transmission characteristic measuring device that does not require frequent adjustments.

The following points were considered in order to achieve this objective: (a) Analog signal processing was replaced with digital signal processing.

(b) By replacing the three filters of the band removal filter, channel filter, and evaluation filter with one digital filter, and selectively controlling the coefficients given to the digital filter, the three filters of the band removal filter, channel filter, and evaluation filter can be used.
It was made smaller by making it function as two filters.

C. The detection circuit is made smaller by using a single square law detection circuit.

The two-decibel value conversion circuit was replaced with a single memory circuit, making it more compact.

E. Eliminates the need for a circuit that converts the PCM signal into an analog signal, that is, a DA converter and low-pass filter.

That's true.

Hereinafter, the (a) general concept, (b) configuration, and (c) operation of the present invention will be explained in order.

B. General Concept Here, in order to supplement the purpose of the present invention described above, an overview of the digital signal processing system and its main components, such as a digital filter, a square-law detection circuit, and a logarithmic value conversion circuit, will be explained.

In conventional analog signal processing type transmission characteristic measurement devices, the signal to be measured is an analog signal, and the power of the analog signal from which unnecessary frequency bands have been removed is converted into a DC value using a linear detection circuit or a square law detection circuit. The power of the signal was measured using a voltmeter, etc. In contrast, in the present invention, the signal to be measured is converted into a numerical value (generally a binary number) representing the amount of the signal to be measured, and the digital signal of 0 and 1 representing the numerical value is processed using an arithmetic circuit such as an adder or a multiplier. In particular, it measures the power of the signal under test.

A digital filter is an arithmetic circuit that filters a digitized signal by numerical calculation.
The calculation formula is shown by the following difference equation.

y(nT)= _N 〓 ⁱ⁼¹ a _i・x{(n−i)T} − _M 〓 ^j=1 b _j・y{(n−j)T} ……(1) (However, x( nT) is the input numerical value string, y (nT) is the output numerical value sequence, T is the sampling period, a _i , b _j are the coefficients of the digital filter, M is the order of the transfer coefficient (MN). At this time, the transfer function of the filter is expressed as (2)
It is indicated by.

(However, Z=e ^jwT W=2πf) The circuit of the digital filter is an arithmetic circuit that executes the above calculation formula (1), but its specific circuit configuration is described in, for example, the Japanese Patent Laid-Open No. 47-31555 This is explained in detail in Donnell Hightley and others.

The feature of the digital filter is that, as can be seen from the above equation (2), by changing the coefficients a _i and b _j of the digital filter, different filter characteristics can be realized using the same circuit. The feature of the present invention is to skillfully utilize the variability of this digital filter characteristic to apply three types of filter functions, channel filter, evaluation filter, and band removal filter, to the coefficients of the above calculation formula (1) according to the application. By providing different numerical values, this can be achieved with a single digital filter, thereby making the transmission characteristic measuring device more compact. In addition, recent digital IC
With the development of technology, the adder circuits, multiplier circuits, etc. that make up digital filters have become smaller, and thus digital filters have become one analog filter,
For example, it is smaller than a channel filter.

Next, the square law detection circuit is expressed by the following calculation formula (3).

P= _M 〓 ^{j= 1} and then accumulate it. When squaring, in the analog signal processing method, the accuracy is determined by the input/output characteristics of the multiplier element itself, but in the digital signal method, the accuracy is determined by the number of digits of the numerical value, so if the number of digits is sufficient, the accuracy of the square-law detection circuit is determined. Accuracy is also good. Therefore, by using the present invention, the linear detection circuit conventionally used only for signal level measurement is no longer necessary.

Next, in the present invention, a memory circuit is used to convert the square law detection output into a BCD decibel value. The operation of the memory circuit is such that when a numerical value indicating a specific address is given to the memory circuit, the content stored in the storage location indicated by that address is outputted. Therefore, if you give the numerical value of the output of the square law detection circuit as an address to the memory circuit and store the BCD decibel value in the corresponding storage location, the BCD decibel value corresponding to the output of the square law detection circuit will be used as the output of the memory circuit. is obtained. For example, if the value of the output of the square law detection circuit is 2048 and the value is 0 dB, the contents shown in FIG. 3 are stored in the memory circuit. 2048 of memory circuit
"0000 0000" is stored at the address, which indicates that when address 2048 is specified in the memory circuit, a BCD "0" is output. “0000 0011” is stored at address 1024. This is because when address 1024 is specified in the memory circuit, half of 2048, that is, BCD "3" meaning -3 dB, is output. Similarly, "0001 0010" indicating -12 dB is stored at address 128.

In this way, conventional analog processing type transmission characteristic measurement equipment requires a hold circuit, a resistance attenuation circuit,
By using the present invention, the decibel converter, which was comprised of a comparison circuit, a reference voltage generation circuit, and an attenuation value designation circuit, has been reduced in size to a single storage circuit.

B. Configuration FIG. 2 is a configuration diagram of an embodiment of the present invention having the same function as the conventional transmission characteristic measuring device shown in FIG. 1.

In the figure, Y is an input terminal to which the PCM signal to be measured is applied; 21 is a code conversion circuit that converts the nonlinear code of PCM into a binary linear code that is easy to process; 22 is a coefficient storage circuit 23, which will be described later. 23 is a digital filter that applies an effect equivalent to a channel filter, band elimination filter, or evaluation filter in the prior art to the digital signal converted into a binary linear code; In order to determine the characteristics of 22, that is, to have the function as a conventional channel filter 8, band elimination filter 7, or evaluation filter 9, the coefficients required for the calculation formula 1 are stored, and the coefficients are stored according to the measurement items. A coefficient storage circuit for supplying the coefficient to the digital filter 22; 24 is a square circuit that squares the digital signal filtered by the digital filter 22 and outputs a digital signal representing the instantaneous power value; 25 is the square circuit an accumulating circuit that receives a digital signal representing the instantaneous power value from 24, successively accumulates it, and outputs it;
A decibel value storage circuit 6 receives the accumulated value from the accumulation circuit 25 and outputs a BCD decibel value according to the accumulated value; 27 receives a BCD decibel value from the decibel value storage circuit 26. 28 is a temporary storage circuit that temporarily stores and holds the digital signal that is the output of the digital filter 22; 29
is a switching device that operates when measuring noise during a PCM line call, and is a switch that is activated when the digital filter 22
This is a switch that selects and switches either the output of the code conversion circuit 21 or the output of the temporary storage circuit 28 as an input. Note that a control circuit (not shown) controls the coefficient storage circuit 23 for channel filters, band elimination filters,
Selection and switch 29 of each coefficient for evaluation filter
switching is controlled.

C. Operation The operation of the present invention will be explained according to the configuration shown in FIG.

First, we will explain how to measure the level of a PCM signal. Channel filter coefficients are given to the digital filter 22 from the coefficient storage circuit 23 in advance. The switch 29 is set to m. When a PCM signal is applied to the input terminal Y and this signal is received by the code conversion circuit 21, the code conversion circuit 21 converts the PCM non-linear code into a digital signal of a binary linear code. This 2
The digital signal of the decimal linear code is applied to the digital filter 22 via the switch 29, and is subjected to calculation processing equivalent to a channel filter, that is, processing for band limiting. (The above calculation formula (1) is calculated.) Then, the calculated signal is squared in a square circuit 24 and accumulated by a certain number of circuits (K times in the above calculation formula (3)). In other words, the calculation formula (3)
calculation process is executed, and the output of the accumulator 25 is PCM
A digital signal representing the power of the signal is output. Here, the calculation process from when the PCM signal is applied to the code conversion circuit 21 to when the output of the square circuit 24 is accumulated by the accumulation circuit 25 is performed until the next PCM signal arrives. Further, the output of the accumulator circuit 25 is cleared to zero each time it outputs a digital value representing power. A digital value indicating the power of the PCM signal is given to a decibel value storage circuit 26 and output as a BCD decibel value. This BCD decibel value is applied to the display circuit 27; each time the accumulator circuit 25 outputs a digital value representing the power of the PCM signal, it is displayed as a decimal decibel value.

Next, the case of measuring PCM no-call noise will be explained. The coefficients of the evaluation filter are stored in advance in the digital filter 22 from the coefficient storage circuit 23.
is given to. At this time, the switch 29 is set to m. The non-call noise of the PCM line applied to the input terminal Y is converted into a binary linear code digital signal by the code conversion circuit 21, and is applied to the digital filter 22 via the switch 29. The digital value indicating the no-call noise is weighted by an evaluation function in the digital filter 22, squared in the square circuit 24, and accumulated in the accumulator 25, and converted into a digital signal indicating the no-call noise power in decibels. The value is given to the value storage circuit 26 and displayed as a decimal decibel value on the display circuit 27 at regular intervals. This display operation is the same as the operation when measuring the signal level.

When measuring the noise during communication on a PCM line using the device of the present invention, first the power of a single frequency signal is measured, then the power of signal components other than those within the voice band is measured, and then the comparison is made. Calculate and find. Here, the operation of measuring the power of signal components other than those within the audio band will be described. First, the coefficients of the band elimination filter are given from the coefficient storage circuit 23 to the digital filter 22, and the switch 29
is set to m. applied to input terminal Y
The PCM signal is applied to the code conversion circuit 21, converted into a digital signal with a binary linear code, and applied to the digital filter 22 via the switch 29. A band-removal filter calculation process is applied, that is, only the signal components are removed,
It is converted into a digital signal indicating the noise during a call.
Here, a digital signal indicating noise during a call is applied to a temporary storage circuit 28 and held therein.

Next, the coefficients of the channel filter are applied from the coefficient storage circuit 23 to the digital filter 22, and the switch 29 is switched to n. Then, the output of the temporary storage circuit 28, that is, the digital value indicating the noise during the call, is again applied to the digital filter 22 via the switch 29. Here, the digital filter 22 operates as a channel filter. Therefore, the digital signal indicating the noise during a call is band-limited by a digital filter 22 and is applied to a squaring circuit 24. The digital signal indicating the band-limited speech noise is sent to the square circuit 2.
4, and the squared digital signal is further accumulated in an accumulating circuit 25 and converted into a digital signal indicating noise power during a call. 27 is displayed as a decimal decibel value.

As described above, compared to the prior art, the present invention has the following advantages: (1) Since the circuit components are digital arithmetic elements such as adders and multipliers, the arithmetic results do not change with temperature or change over time.

(b) The digital arithmetic circuit does not need to be adjusted.

C. Since the characteristics of a plurality of filters can be realized with one digital filter, even if a plurality of filters are required, only one digital filter is required, leading to miniaturization.

D. Since signals are handled in the form of numerical values and transmission characteristics are measured, a means for converting analog signals into digital signals consisting of a low-pass filter, a sample-hold circuit, and an AD conversion circuit is provided in place of the code conversion circuit 21. For example, the transmission characteristics of analog signals can be measured.

E. If a multiplexer is provided before the code conversion circuit 21, the transmission characteristics of multiple lines can be measured at the same time using the multiprocessing function, which is another feature of the digital signal processing method.

F. If an electronic computer is provided in place of the display circuit 27, it is possible to measure the transmission characteristics of multiple lines in a planned manner, and it is also possible to realize a transmission characteristics measurement system that statistically aggregates the measurement results.

G. Since digital-to-analog conversion is not required as in the past, errors in this conversion process are eliminated, and accuracy can be ensured.

There are advantages such as

[Brief explanation of drawings]

FIG. 1 is a block configuration diagram according to an embodiment of conventional transmission characteristic measurement; FIG. 2 is a block diagram according to an embodiment of the present invention; FIG. 3 is an explanation showing addresses in the decibel value storage circuit 26 and their contents. It is a diagram. In the figure, 1 and 21 are code conversion circuits, and 2 is a DA
Conversion circuit, 3 is a low pass filter, 4, 5, 6,
10, 11, 14, 29 are switches, 7 is a band rejection filter, 8 is a channel filter, 9 is an evaluation filter, 12 is a linear detection circuit, 13 is a square law detection circuit, 15 is a hold circuit, 16 is a resistance attenuation circuit, 17 is a comparison circuit, 18 is a reference voltage generation circuit,
19 is an attenuation value designation circuit, 20 and 27 are display circuits,
22 is a digital filter, 23 is a coefficient storage circuit, 24 is a square circuit, 25 is an accumulation circuit, 26 is a decibel value storage circuit, 28 is a temporary storage circuit, X, Y
is an input terminal.

Claims (1)

[Scope of Claims] 1. In a digital transmission characteristic measuring device that measures the PCM signal level, no-call noise, and call noise of a digital telephone line, code conversion that receives a non-linear code input signal and converts it into a linear code. a circuit 21; a switch 29 for switching the output or temporary storage output of the code conversion circuit 21; a coefficient storage circuit 23 storing coefficients for at least two filter processes; receiving the output of the switch 29; a digital filter 22 that performs filter processing determined by the coefficients output from the coefficient storage circuit; and a digital filter 22 that stores the output of the digital filter 22;
A temporary storage circuit 28 that outputs to the digital filter 22 as a temporary storage output during the next filtering process of the digital filter 22 via the switch 29; and a squaring circuit that squares the output of the digital filter and outputs an instantaneous power value. 24, an accumulation circuit 25 that accumulates the instantaneous power value output from the square circuit 24 for a predetermined period of time and outputs the accumulated value; and an accumulation circuit 25 that receives the output of the accumulation circuit 25 as an address and performs the accumulation. A digital transmission characteristic measuring device comprising: a decibel value storage circuit 26 that outputs a decibel value corresponding to the output value of the circuit 25;