KR0174695B1 - Multi-channel dual tone multi-frequency and special tone detection device and method - Google Patents

Abstract

1. The technical field to which the invention described in the claims belongs:

Exchanger.

2. The technical problem the invention is trying to solve:

One integrated circuit provides an apparatus and method for detecting multi-channel DTMF and special tones.

3. Summary of the Solution of the Invention:

The multi-channel dual tone multi-frequency and special tone detection device includes a means for allocating a user time slot in the switch, and means for linearizing the data of each channel pulse code modulation data of the compressed first predetermined bit into a second predetermined bit. Means for storing the linear data for each channel according to an assigned time slot, means for generating a control code for dividing the linear data into a dual tone multi-frequency mode or a special tone mode, and the linear data according to a set mode. A single integrated circuit comprising means for detecting a double tone multi-frequency signal or a special tone from said second tone and storing a digit corresponding thereto; It is characterized in that the main control unit for controlling the generation of the control code by setting the mode for each channel and the exchange main control unit for reading the digit from the integrated circuit.

4. Important uses of the invention:

One integrated circuit used to detect multi-channel DTMF and special tones.

Description

Multi-channel dual tone multi-frequency and special tone detection device and method

1 is a block diagram of a multi-channel double tone multi-frequency and special tone detection circuit according to the present invention.

2 is an operational waveform diagram according to the present invention.

3 is a flowchart illustrating a multi-channel double tone multi-frequency and special tone detection process according to the present invention.

4 is an energy graph for each frequency used in the present invention.

* Explanation of symbols for main parts of the drawings

110: parallel and parallel converter (SPC) 120: linear converter

140, 170: Registers 150, 210: Multiplexer

160: DSP cores 180 to 200: decoder

220: TSAC 300: CPU

The present invention relates to an apparatus and method for detecting a dual tone multi frequency (hereinafter referred to as DTMF) and a special tone in an exchange, and more particularly, to an apparatus and method for detecting DTMF and special tones of a multi-channel. It is about.

In general, DTMF receivers take a form of detecting a DTMF signal of one channel in an integrated circuit (hereinafter referred to as IC), and mainly by using an analog structure, that is, a bandpass filter structure. It is configured to detect the component. When the analog DTMF receiver is used for the digital exchange, one PCMF codec is additionally required for one DTMF receiver due to the current exchange structure. This is because the analog signal is converted into a PCM pulse and converted back into an analog signal according to the assigned user channel, and the DTMF signal is detected by inputting the changed signal.

Eventually, the structure of DTMF receiver was digitized to be suitable for all digital exchanges, and the bandpass filter was applied to a digital signal processor (hereinafter referred to as DSP), or a Goertzel algorithm was applied. The main form is to do.

However, all of these types of DTMF detection devices are designed in such a way that one IC can detect one channel or only DTMF. Therefore, a single IC can detect several channels at the same time or another special tone by mode switching. There was a disadvantage.

Accordingly, an object of the present invention is to provide an apparatus and method for detecting DTMF and tone signals of various channels in the form of a single chip.

The multi-channel double tone multi-frequency and special tone detection device for achieving the above object is characterized in that the switch comprises: means for allocating a user time slot, each channel pulse code modulation data of the compressed first predetermined bit into a second predetermined bit; Means for decompressing and linearizing data, means for storing the linear data for each channel according to an allocated time slot, means for generating a control code for dividing and processing the linear data into a dual tone multi-frequency mode or a special tone mode; A single integrated circuit comprising means for detecting a dual tone multi-frequency signal or a special tone from said linear data and storing a digit corresponding thereto; It is characterized in that the main control unit for controlling the generation of the control code by setting the mode for each channel and the exchange main control unit for reading the digit from the integrated circuit.

Hereinafter, exemplary embodiments of the present invention will be described in detail with reference to the accompanying drawings.

First, in adding reference numerals to the components of each drawing, it should be noted that the same reference numerals have the same reference numerals as much as possible even if displayed on different drawings. In addition, in the following description, many specific details such as components of specific circuits are shown, which are provided to help a more general understanding of the present invention, and the present invention may be practiced without these specific details. It will be obvious to those skilled in the art. In the following description of the present invention, if it is determined that a detailed description of a related known function or configuration may unnecessarily obscure the subject matter of the present invention, the detailed description thereof will be omitted.

1 is a configuration diagram of a multi-channel DTMF and a special tone detection apparatus according to the present invention, the operation of each part in detail as follows.

First, in this embodiment, a time slot assignment circuit is connected to a T1 (North American) and CEPT (European) exchange device in order to detect DTMF and tone using a DSP core. Communicates with the exchange main controller, i.e., CPU 300, in order to be able to program the 220 and to adjust the variable values by the exchange to better detect DTMF and tones. The interface unit 100 is provided.

In the serial-to-parallel converter (hereinafter referred to as SPC) 110, an 8-bit serial bit stream of each channel input to the highway L11 is converted into parallel 8-bit data. To perform this conversion, a reference clock is used. In the case of an audio signal, a periodic signal of 8 KHz is used, and all the operations are controlled by this synchronization signal.

The linear converter 120 converts the PCM data converted in parallel into a linear A-Law or Mu-Law method based on CCITTG.711 to form a linear data. As an example, 8-bit PCM data is input to output 14-bit linear data. The linear data is 13 ₁ , 13 ₂ ,... The signals of TSAC 220, which generate 13 _n pulses, are sequentially stored in registers 140 at the instant 'fall', that is, at the moment of falling from the high level to the low level.

2 shows 13 ₁ , 13 ₂ ,... Output from the sync clock SCLK, the reference clock RCLK, and the TSAC 220. , The shape of a 13 _n pulse is shown. The moment when the pulse falls from the high level to the low level corresponds to the 'a' part of the 131 pulses.

Referring back to FIG. 1, register 140 stores linear data channel by channel.

The DSP core 160 receives an interrupt signal every time the sync clock SCLK changes from a high level to a low level, and reads each channel data stored in the register 140, and each read data is transmitted to the DTMF and tone detection algorithms for each channel. Processed to find the DTMF or tone signal actually transmitted. A detailed operation process of this algorithm will be described later with reference to FIG. 3.

The algorithm embedded in the DSP core is processed through different processes according to the result of recognizing whether each channel mode, that is, DTMF mode or tone mode, is given from the CPU 300 by the interface unit 100.

The control code is sequentially generated to read the linear data stored in the register 140 using the synchronous clock SCLK inputted at regular intervals as an interrupt signal. The decoder 200 which inputs and decodes the control code distributes the output signal to the register 140 to enable the corresponding register.

Data of each channel input to the DSP core 160 is processed for each channel according to the flow chart of FIG. 3 to be described later, and outputs predetermined digits according to the finally detected DTMF and tone signals, which are output in the order of input. Each is stored in register 170. At this time, the signal for enabling the register 170 is output from the decoder 180 to decode the control code output from the DSP core 160.

Digits stored in the register 170 are read out from the CPU 300 of the exchange every predetermined time. The decoder 190 decodes the control code output from the CPU 300 and enables the register 170 so that the CPU 300 can read the digits of the register.

The decoder 200 decodes the control code of the DSP core 160 to read linearly converted data and generates a signal for each channel.

The multiplexer 150 and the multiplexer 210 open a path only for data of a predetermined channel at the moment of reading data of each channel.

Next, the process of detecting DTMF and tones will be described with reference to FIG.

When power is initially supplied to the DSP core 160, a program for executing the present invention begins to be executed. First, in step 3a, data is downloaded from the CPU 300 and stored in the internal RAM. This download data contains a lot of information, such as the number of samples by frequency, the energy threshold, the energy weight by frequency, the coefficient of distortion and the digital filter coefficients.

When the initial download is completed, the command data is received from the CPU 300 and the operation mode is set. That is, the program determines whether to operate in DTMF mode or tone mode. The reason for receiving the mode setting data from the CPU 100 is that when the DTMF detection device is used at the exchange, the DTMF and the special tone are not detected at the same time, and which mode should be set as one of the two modes. Setting this mode not only reduces the load on the DTMF detector, but also reduces the likelihood that the DSP will malfunction.

If you select DTMF mode, go to step 3f to remove the dial tone. This is to increase the accuracy of detection since the first digit of the DTMF is input together with the dial tone. As a method of eliminating dial tone, it is possible to use a notch filter or a bandpass filter that passes only the required frequency. In the case of the notch filter, the bandpass filter has a high band rejection effect but no frequency stability. use.

In case of the tone mode, the process proceeds directly to step 3c. In case of the DTMF mode, the process proceeds to step 3g through the step 3f to adjust the PCM data size to perform AGC (Automatic Gain Control). That is, when the filtered PCM data is input, the data is shifted according to the size of the input PCM data in order to widen the dynamic range. By doing this, the input level of PCM data can be detected from -0dB to -50dB.

The PCM data that has been resized like this is performed in step 3d and 3h to perform the algorithm. First, the energy factor is calculated for the DTMF of the input data and the frequency components used for the special tones. When calculating the energy factor for each frequency, the number of samples to be processed differs depending on the frequency, since the frequency range used for DTMF tones and special tones varies widely from 350 Hz (dial tones) to 1700 Hz (pager tones). . In other words, the phaser tone is about five times higher than the dial tone, and the same number of samples makes a large difference in the frequency resolution.

Explaining this phenomenon using Fig. 4, if the energy value of each frequency is calculated using the same sample number, the energy at 394 Hz, which is 1 Hz from the dial tone, and the energy at 1699 Hz, which is 1 Hz away from the pager tone, are the same. B) Tolerances in special tones are given in%. In this case, 1699Hz in phaser tone is 0.06% to 1700Hz, but the error in dial tone is about 4 times to 0.2%. In order to solve this characteristic, the detection energy level may be adjusted for each frequency. However, in the case of the pager tone, even if noise is input because the detection energy level is too small, an error may occur. Therefore, the problem is solved by adjusting the distribution of energy by varying the number of samples by frequency without using the above method. In this case, the energy factor is multiplied by each frequency energy to compensate for the lowering of the energy level at the center frequency. The coefficients needed for this operation are downloaded from the CPU 300 at the beginning of the program execution.

After that, the total energy of the input PCM data is calculated in steps 3e and 3o. In step 3i and 3p, the data shift factor is determined based on the calculated energy. This is to examine and deal with whether overflow may occur in the next sample processing. The data shift factor determined here indicates how much to adjust the size of the input PCM data in step 3c or 3g.

If all of the above-described operation process is performed, all calculations for one sample are completed. Next, compare the number of samples in steps 3j and 3q. Here, N1 represents the minimum number of samples that can determine the digits of the input data. The determination of N1 should consider the following two points.

First, it must be larger than the number of samples for each frequency determined in the 3D or 3h step algorithm.

Second, it should be less than the maximum detection time.

For example, the maximum detection time is 50 msec for DTMF, and 400 samples for 8 KHz. However, in order to detect a single accurate digit, DTMF is detected only when two detected digits are the same. Therefore, it should not exceed 200 samples, which is 1/2 of 400 samples. The problem caused by this is that in case of dial tone, the frequency should be low, so the number of samples should be 600, which is three times the number of 200. However, the number of samples of N1 is limited to 200, making it difficult to operate in tone mode. Therefore, in step 3k, 'N2 ≤ number of detection repetitions' is checked.

On the other hand, in the tone mode, the maximum detection time exceeds 200 msec, regardless of the number of sampling. In order to increase the sampling number for detecting a special tone once, the special tone is detected once while the DTMF detection process is repeated several times.

Next, the energy of each frequency is calculated to determine which DTMF or special tone signal is input in steps 3l and 3r. As a result of the calculation, a digit corresponding to a DTMF or a special tone detection result is output in steps 3m and 3s, and the digit is stored in the internal RAM of the DSP core 160 and read by the CPU 300 at regular intervals.

The form read by the CPU is configured to read the detected contents by accessing the DTMF and the tone receiver at regular intervals.

As described above, the present invention has an advantage of being able to detect DTMF and special tones of multiple channels as one integrated circuit without having a separate circuit for each channel.

Meanwhile, in the detailed description of the present invention, specific embodiments have been described, but various modifications may be made without departing from the scope of the present invention. Therefore, the scope of the present invention should not be limited to the described embodiments, but should be defined not only by the scope of the following claims, but also by the equivalents of the claims.

Claims (12)

In the exchange, means for allocating user time slots, means for linearly expanding each channel pulse code modulation data of a compressed first predetermined bit into a second predetermined bit, and linearly storing the linear data for each channel according to the allocated time slots. Means for storing, means for generating a control code for dividing the linear data into a dual tone multi-frequency mode or a special tone mode, and detecting a dual tone multi-frequency signal or a special tone from the linear data according to a set mode. An integrated circuit comprising means for storing corresponding digits; A multi-channel dual tone multi-frequency and special tone detection device, comprising: a main controller for controlling the generation of the control code by setting a mode for each channel and reading the digit from the integrated circuit. 2. The apparatus of claim 1, wherein the linear converter processes a mu-lo or a-lo pulse code modulation signal by mode selection. The apparatus of claim 1, further comprising a decoder for decoding the control code to control the first and second register units. The apparatus of claim 1, further comprising a multiplexer for selectively transferring linearly converted data output from the first register unit to the digital signal processor. The apparatus of claim 1, further comprising a multiplexer for selectively outputting a digit output from the second register unit. In the exchange, a time slot for allocating a user time slot according to the synchronization clock, the reference clock, the dual tone multi-frequency mode or the special tone mode, and the synchronization clock, the reference clock and the mode setting signal A serial / parallel converter for converting the serial pulse code modulation signal of the compressed first predetermined bit inputted through the highway into a parallel data form, and a linear converter for extending the parallel data into a second predetermined bit to linearize the data. And a first register unit for each channel that stores the linear data, and linear data corresponding to a selected channel from the first register unit to detect a dual tone multi-frequency signal or a special tone, and output a digit corresponding thereto. And a digital signal processor for generating a predetermined control code according to the mode setting signal, and the digital signal processor. A digit outputted from the call processor is a dual tone multi-frequency channels and special tone detection apparatus, characterized by a second portion configured to store a register for each channel corresponding to the channel in accordance with the control code. 7. The apparatus of claim 6, wherein the linear converter processes a mu-lo or a-lo pulse code modulation signal by mode selection. 7. The apparatus of claim 6, further comprising first and second decoders for decoding the control code and enabling a register of a corresponding channel of the first or second register part. 9. The apparatus of claim 8, further comprising a multiplexer for selecting linearly converted data output from an arbitrary register of the first register enabled by the first decoder and transferring the linearly converted data to the digital signal processor. . The apparatus of claim 8, further comprising a multiplexer for selecting and outputting a digit output from an arbitrary register of the second register part enabled by the second decoder. In the method of detecting multi-channel double tone multi-frequency and special tone as a single integrated circuit in the exchange, downloading the number of samples per frequency, the energy threshold, the energy weight per frequency, the high coefficient, the digital tone removal coefficient, etc. The second step of setting the operation mode of one of the dual tone multi-frequency mode or the special tone mode, the third step of removing the dial tone when the dual tone multi-frequency mode is selected, the tone mode or the dial tone A fourth step of automatically gain control by adjusting the size of the pulse code modulation data after the removal of the signal, a fifth step of detecting a used frequency component by performing a high-frequency algorithm for the adjusted data, and energy for the detected frequency component A sixth step of calculating the factor, after calculating the energy factor to the entirety of the pulse code modulation data First step of repeating the operation of looping to the fourth step by the minimum number of samples that can determine the digits of the input data after calculating a data shift factor using the calculated energy. A second step of calculating energy for each frequency to determine a double tone multi-frequency signal or a tone signal, and a third step of generating and storing a digit corresponding to the double tone multi-frequency signal or a special tone according to the calculation result; And a fourth process of reading the digits of each of the stored channels at regular intervals. 12. The method of claim 11, wherein the minimum number of samples that can determine the digits of the input data is greater than the number of samples for each frequency determined during the high-frequency algorithm and less than the maximum detection time.