JPH0748890B2 - Multi-frequency signal receiver and amplitude limiting method for outer peripheral wave signal - Google Patents

Description

The present invention relates to a multi-frequency signal receiver and a multi-frequency signal amplitude limiting method, and in a process of limiting the amplitude of an input signal, eliminating a period error due to sampling accompanying digital processing. For the purpose, each sample obtained by sampling and A / D converting the multi-frequency analog input signal is amplitude-limited into a digital signal with a constant positive and negative amplitude that preserves the cycle of the original multi-frequency analog signal,
In an amplitude limiting method of a multi-frequency signal in a multi-frequency signal receiver to be added to a band-pass filter for frequency detection, the amplitudes of samples before and after the polarity of the digital signal is changed to the analog input signal and positive and negative threshold values. It is adjusted according to the timing position of the intersection with and to remove the periodic error.

[Industrial application field]

The present invention relates to a multi-frequency signal receiver used in a push-button (PB) signal receiver used in a telephone exchange or the like, and a multi-frequency signal amplitude limiting method.

PB signals of push-button telephones are 12 types of multi-frequency signals consisting of one of each L side 697,770,852,941 Hz and one of each H side 1209,1336,1477 Hz.
Assigned to 9, #, *. The multi-frequency signal receiver receives the multi-frequency signal, and decodes from the frequency component which push button is generated by pressing. This is possible by using a bandpass filter (BPF).

For PB signals, it is necessary to selectively detect near-blank frequencies with a minimum frequency difference of 73 Hz. Therefore, the PB signal receiver has a precise frequency detection standard of dead band ± 3.0%. Further, in order to detect a frequency in a narrow band, it is necessary to limit the amplitude of the input signal within a certain range and remove the adverse effect due to the amplitude fluctuation. In other words, even if the output is -3 dB for the 0 dB input, if the input of 3 dB is applied, the output will be 0 dB, and the impression level will be output, causing a malfunction. Therefore, if the amplitude of the input signal can be limited to a fixed value, this concern will disappear. The present invention relates to this amplitude limitation.

[Prior art]

In the prior art, when detecting element frequencies contained in a multi-frequency analog input signal, amplitude limitation is first performed as is generally done by demodulating an FM signal. In multi-frequency signal receiver, when the amplitude of multi-frequency analog input signal exceeds a certain positive or negative constant value (L _Th ), the polarity of the positive or negative constant value (L _O ) of the output is inverted and Is converted into an output signal in the form of a phase signal of. In the output signal converted by this processing, the amplitude becomes constant and the cycle is preserved. Fourth
The figure shows this outline. In this figure, the input signal is shown as a sine wave, and the output signal takes a positive constant value L _O when the sine wave exceeds the positive threshold L _Th, and when the sine wave exceeds the negative threshold −L _Th Negative constant value-Inverts to L _O , and so on.
As is clear from the drawing, the cycle of the input signal is preserved in the output signal by this process.

In the field of telephones, digitization is prosperous, and therefore, digital processing is desired instead of analog processing as shown in FIG. When the multi-frequency signal receiver realized by digital signal processing, the input signal is the sampling frequency (F _S)
Each sample is obtained by sampling the analog input signal. If the same amplitude limiting process as before is performed on each of these samples, the period of the input signal and the output signal becomes maximum. Error of. FIG. 5A shows the digital signal processing by the conventional method, and FIG.

The input signal of FIG. 5 (a) is obtained by sampling the sine wave (input signal) of FIG. 4 at the period 1 / F _S and arranging the samples on the time (t) axis. In the sample of
When L _Th or more is generated, the output is a constant value L _O (L _O
Each sample and then), and - an output when L _Th following are generated and each sample constant value -L _O. With this, the same result as in FIG. 4 can be obtained, but since it is sampled at 1 / F _S , it is not possible to accurately grasp the time when the thresholds L _Th and −L _Th are exceeded, which leads to an error.

When described in FIG. 5 (b), S ₁ is then S ₂ immediately before becomes equal to or higher than the threshold L _Th is the sample of the input signal immediately after becomes equal to or higher than the threshold L _Th, the analog signal of this portion is the threshold L _Th Is exceeded at time t ₁ . Although S ₃ is a sample of the output signal for S ₁ and S ₄ for S ₂ , it is determined from these output samples S ₃ , S ₄ that the reversal (threshold crossing) has occurred at the central t ₂ time point. I will end up. Therefore, a periodic error of ΔT = t _{2 to} t ₁ occurs. This periodic error is, for example, 697Hz for the input signal,
In a PB signal receiver with F _S = 8 KHz, ΔT = 62.5 ns. If the input signal is 697Hz, this error causes 667.9 ~
It will vary to 728.7. To receive this,
It is necessary to widen the impression band. However, since the dead band is ± 3.0% in the frequency detection standard, do not be impressed with 718 or more and 676 or less. Therefore, such a receiver cannot be configured.

The digital output signal whose amplitude is limited in FIG. 4 (a) is used for frequency detection in addition to the digital BPF in the receiver of the PB signal. However, if there is a cycle error ΔT, the output of the BPF will have ripples. . That is, the amplitude of the input signal to the BPF is limited, so that the amplitude of the output signal of the BPF should be constant, but the amplitude fluctuates due to the period error ΔT. Therefore, BP
In order to deal with this in the post-stage circuit of F, it is necessary to widen the moving band so that it is not affected by the amplitude fluctuation. However, if the moving band is widened, it becomes impossible to detect the narrow band, which hinders the detection of the PB multi-frequency signal.

An object of the present invention is to eliminate a periodic error in the process of limiting the amplitude of an input signal.

[Means for Solving the Problems]

FIG. 1 shows the principle of the present invention. Also in the present invention, when each sample of the input signal exceeds the positive threshold value L _Th , the output signal is set to the positive constant value L _O , and when the sample falls below the negative threshold value −L _Th , the negative constant value −L _O is set. , The output signal corresponding to the sample above the positive threshold and the sample immediately before it, and the sample below the negative threshold and the sample immediately before it, must be a positive constant value L _O or a negative constant value -L _O. First, the timing position at which the input signal crosses the positive and negative threshold values is adjusted according to the size deviated from the midpoint of the interval between the immediately preceding sample position and the immediately following sample position. That is, the excess amount of the sample S ₂ immediately after exceeding the positive threshold value L _Th is b,
The shortage of S ₁ as a, the output sample S ₃ with respect to the input samples S ₁ is the -L not _{O -aL O / (a + b)} , the output samples for the input sample S ₂ is + L not _O + bL _O /
Set to (a + b).

According to the configuration of the multi-frequency signal receiver of the present invention, the input samples of the digital signal after sampling and A / D converting the multi-frequency analog input signal are converted into positive and negative threshold values (L _Th , -L
_Th ), and as a result, the value of the input sample is positive for each input sample between the positive threshold (L _Th ) and the next negative threshold (-L _Th ), Produces an output sample of one of the negative constant values (L _O , -L _O ), and the next positive threshold (L _Th ) after the value of the input sample exceeds the negative threshold (-L _Th ). the positive for each input sample until exceeding, resulting negative constant value (L _O, -L _O) the output samples of the other values of positive, negative constant value (L _O, - In a multi-frequency signal receiver that detects the frequency of each element of a multi-frequency signal by adding each output sample whose amplitude is limited to L _O ), the value of the output sample applied to the band-pass filter is the value immediately before and immediately after the output sample (S _3, S ₄₎ changing the polarity, it output samples immediately before and straight (S _3, S ₄₎ Is the value the positive corresponding input samples (S _1, S _2), to the sum of the negative threshold (L _Th, -L _Th) shortage and their overrun seeking excess and amounts short of the each proportion of the excess amount and insufficient amount, the positive, negative constant value (L _O, -L _O) comprising means for adjusting the value obtained by multiplying each of the multi-frequency analog input signal is the positive, negative threshold (L _Th ,
-L _Th ), and the time when the straight line connecting the respective values of the output samples (S ₃ , S ₄ ) immediately before and immediately after changing the polarity adjusted by the means intersects the 0 level. It is characterized by.

Further, according to the amplitude limiting method for a multi-frequency signal of the present invention, each input sample of the digital signal after sampling and A / D converting the multi-frequency analog input signal is processed into positive and negative threshold values (L _Th , -L
_Th ), and as a result, the value of the input sample is positive for each input sample between the positive threshold (L _Th ) and the next negative threshold (-L _Th ), Produces an output sample of one of the negative constant values (L _O , -L _O ), and the next positive threshold (L _Th ) after the value of the input sample exceeds the negative threshold (-L _Th ). the positive for each input sample until exceeding, resulting negative constant value (L _O, -L _O) the output samples of the other values of positive, negative constant value (L _O, - In the amplitude limiting method of a multi-frequency signal for obtaining an output sample whose amplitude is limited to L _O ), the values of the output samples (S ₃ , S ₄ ) immediately before and after the polarity of the value of the output sample changes And input samples (S ₁ , S ₂ ) corresponding to the output samples (S ₃ , S ₄ ) immediately after
Of the positive and negative thresholds (L _Th , -L _Th ) to obtain the excess and deficiency, and calculate the excess and deficiency ratios for the sum of the excess and deficiency. , Negative constant values (L _O , −L _O ) are respectively adjusted so that the multi-frequency analog input signal has the positive and negative threshold values (L _Th ,
-L _Th ), and the time when the straight line connecting the respective values of the output samples (S ₃ , S ₄ ) immediately before and immediately after changing the polarity adjusted by the means intersects the 0 level. It is characterized by.

[Action]

According to the configuration of the multi-frequency signal receiver and the amplitude limiting method of the multi-frequency signal of the present invention described above, in the conventional technique, the time when the signal value represented by the output sample becomes substantially 0 (at the time of interpolation) is within the sample interval. What was a point is the same as when the analog input signal crosses the threshold. This point is clear from the fact that each value of S ₃ and S ₄ is proportional to the excess or deficiency of each value of S ₂ and S ₁ with respect to the threshold value, and the intervals are both equal to 1 / F _S. Note that the analog input signal has its harmonics removed in advance by a BEF (band elimination filter), and has only one multi-frequency signal, that is, a sine wave, so the analog input signal threshold crossing point and the output sample zero point If and match, there is no difference between the periods of the input signal and the output signal, and the frequency of the input signal is accurately stored in the output signal.

〔Example〕

FIG. 2 shows a processing flow as an embodiment of the present invention. x _i is the input sample value, x _i-1 is the previous input sample value,
L _Th and −L _Th are the above-mentioned positive and negative threshold values, L _O is a constant output value, and A, B, C0, and C1 are variables.

First, it is checked whether it is x _i L _Th . If L _Th or more,
Check if x _i-1 L _Th . If x _i-1 is smaller than L _Th, the process of the present invention is performed, and L _Th −x _i−1 is A, x _i −L _Th is B, −A / (A + B) is calculated, and this is C0, Or
Calculate B / (A + B) and set this as C1 and set C0 · L ₀ as y _i-1
And Further, since the next output (y _i ) is set to BL ₀ / (A + B), the value of C0 is set to C1 and the processing is ended.

Then x _i, if x _i-1 is more than L _Th together (, but the case of YES), so that the next output is L _0, put 1 to C1, and outputs the C0 · L ₀ of the previous , And for the next output, change the value of C0 to C1.
And finish.

When x _i is smaller than L _Th, it is checked whether it is x _i -L _Th . If x _i is less than −L _Th, check whether it is x _i−1 −L _Th . When x _i-1 is larger than -L _{Th, the} process according to the present invention is performed, and -L _Th + x _i-1 is A and -x _i + L _Th is B,
Calculate A / (A + B) and set it as C0, and -B / (A +
B) is calculated, and this is designated as C1, and C0 · L ₀ is designated as y _i-1 .
Further, the value of C0 is set to C1 in order to obtain the next output C1 · L ₀ .

Next, if both x _i and x _i-1 are less than -L _Th (when, is YES), put -1 in C1 so that the next output will be -L ₀ ,
Outputs previous C0 · L _0, further C1 values of C0 for the next output
And finish.

-When L _Th <x _i <L _Th (when, is NO), the same y _i-1 as the previous time is output and C1 → C0 ends.

The processing for the input samples x ₁ , x ₂ , x ₃ ... X _{12 in} FIG. 1 (c) is collectively shown as follows.

x ₁ ... is Y, is N, and output = -A / (A + B).
L ₀ x ₂ ... is Y, is Y, and output = B / (A + B).
L ₀ x ₃ ... is Y, is Y, and is output = L ₀ x ₄ ... is N, is N, is output = L ₀ x ₅ ... Same as above x ₆ ... Same as above x ₇ ... is N, is Y, is N , Output = A / (A +
B) · L ₀ x ₈ ... is N, is Y, and output = -B / (A +
B) -L ₀ x ₉ ... is N, is Y, and is output = -L ₀ x ₁₀ ... is N, is N, is output = -L ₀ x ₁₁ ... Same as above x ₁₂ ... Same as above Fig. 3 (a) As shown, the PB receiver PBREC is connected to the network (switch) NW and has the configuration shown in FIG. DT is a telephone terminal (push-button telephone), and SC is a subscriber circuit. DTR is a circuit that rejects a dial tone of 400Hz, 3F is a filter for city noise (3KHz), BEF ₁ , BEF
₂ is a filter for removing the other side of high and low frequency PB multi-frequency signals, and BPF _{1 to} BPF ₇ are band pass filters for passing 3 types of high frequency (H) signals and 4 types of low frequency (L) signals. Is. The outputs of these BPFs are processed so that the output terminals simply show 1,0. It is the limiter circuit LIM that outputs the input signal of FIG. 1 and the like whose amplitude is limited, and the amplitude of the sample before and after the polarity reversal of the present invention is adjusted.
FIG. 3B is actually composed of a DSP.

If the input of the BPF fluctuates, for example, the output curve C ₁ becomes C ₂ , and the frequency has a level that is not f ₁ but f ₂ so that the determination result by the threshold value may be erroneous. BPF characteristics C ₁
This can be prevented by sharpening, but this complicates the BPF configuration. If the amplitude of the output does not fluctuate, the filter structure can be simplified and the occurrence of erroneous output can be avoided.

〔The invention's effect〕

As described above, according to the present invention, the amplitude limit of the digital signal obtained by sampling the analog input signal is adjusted before and after the polarity conversion, and the point where the output signal after the amplitude limit becomes 0 is the amplitude limit value of the input signal. Since it coincides with the intersection point of, the difference between the cycles of the output signal and the input signal can be eliminated. That is, it is possible to realize the amplitude limitation in which only the amplitude is limited by the digital signal and the frequency is preserved.

[Brief description of drawings]

FIG. 1 is an explanatory view of the principle of the present invention, FIG. 2 is a flow chart showing an embodiment of the present invention, FIG. 3 is a block diagram showing an application example of the present invention, and FIG. 4 is an explanatory view of amplitude limitation of an analog signal. FIG. 5 is an explanatory diagram of the amplitude limitation of the digital signal. In FIG. 1, L _Th , -L _Th are positive and negative thresholds, S _{1 to} S ₄ are samples,
L ₀ -L ₀ is a positive and negative constant amplitude.

Claims (2)

[Claims] 1. Sampling a multi-frequency analog input signal
Each input sample of the digital signal after A / D conversion,
Positive, negative threshold (L _Th, -L _Th) compared to its results from the input sample value exceeds the positive threshold (L _Th) until the next exceeds a negative threshold (-L _Th) For each input sample in between, an output sample of one of positive and negative constant values (L _O , -L _O ) is generated, and the value of the input sample is a negative threshold (-
The positive for each input sample of between exceed L _Th) until the next exceeds a positive threshold (L _Th), a constant negative value (L _O, output of the other values of -L _O) A multi-frequency detecting a frequency of each element of a multi-frequency signal by generating each sample and adding each output sample whose amplitude is limited to the positive and negative constant values (L _O , -L _O ) to a plurality of band pass filters. In the signal receiver, the values of the output samples (S ₃ , S ₄ ) immediately before and immediately after the value of the output sample added to the band pass filter changes the polarity, and the values of the output samples immediately before and after the output samples (S ₃ , S ₄ ) The values of the input samples (S ₁ , S ₂ ) corresponding to the above are calculated for the amount of excess and deficiency with respect to the positive and negative thresholds (L _Th , -L _Th ), and the sum of these excess and deficiency is calculated. hand each ratio of the excess amount and insufficient amount, the positive, negative constant value (L _O, -L _O) adjusted to a value obtained by multiplying each relative Wherein the multi-frequency analog input signal is the positive, negative threshold (L _Th,
-L _Th ), and the time when the straight line connecting the respective values of the output samples (S ₃ , S ₄ ) immediately before and immediately after changing the polarity adjusted by the means intersects the 0 level. Multi-frequency signal receiver characterized by. 2. Sampling a multi-frequency analog input signal
Each input sample of the digital signal after A / D conversion,
Positive, negative threshold (L _Th, -L _Th) compared to its results from the input sample value exceeds the positive threshold (L _Th) until the next exceeds a negative threshold (-L _Th) For each input sample in between, an output sample of one of positive and negative constant values (L _O , -L _O ) is generated, and the value of the input sample is a negative threshold (-
The positive for each input sample of between exceed L _Th) until the next exceeds a positive threshold (L _Th), a constant negative value (L _O, output of the other values of -L _O) In a method of limiting the amplitude of a multi-frequency signal, in which a sample is generated and an output sample whose amplitude is limited to the positive and negative constant values (L _O , -L _O ) is obtained, immediately before and after the polarity of the value of the output sample changes. Of the output samples (S ₃ , S ₄ ) of the input sample (S ₁ , S ₂ ) corresponding to the output samples (S ₃ , S ₄ ) immediately before and after the output sample (S ₁ , S ₂ ), respectively.
Of the positive and negative thresholds (L _Th , -L _Th ) to obtain the excess and deficiency, and calculate the excess and deficiency ratios for the sum of the excess and deficiency. , Negative constant values (L _O , −L _O ) are respectively adjusted so that the multi-frequency analog input signal has the positive and negative threshold values (L _Th ,
-L _Th ), and the time when the straight line connecting the respective values of the output samples (S ₃ , S ₄ ) immediately before and immediately after changing the polarity adjusted by the means intersects the 0 level. A method for limiting the amplitude of a multi-frequency signal characterized by.