SU1247775A1 - Device for recognizing single and group composite pulse signals - Google Patents

Abstract

This invention relates to electrical measuring technology. It can also be used for processing block or seismic signals. The purpose of the invention is to increase the likelihood of recognizing single and grouped pulse signals. For this purpose, a device containing an analog-to-digital converter 1, a fast Fourier transform block 2, a computation block 3 of a square module of the discrete Fourier transform coefficients,. switch 4, block 5 of memory, block 6 of division, block 7 of calculation and calculation of Ft.g sl with: VI VJ SP

Description

tanya, multiplication unit 8, weight function generator 9, inverse fast Fourier transform unit 10, digital-to-analogue converter (11AP) 14, registration unit 15, synchro-generator 16, first, second 11, 12 comparison blocks, display unit 13 for single and group impulses . Situations of the absence of a signal correspond to the codes O, O, 00 at the outputs of blocks 11-13, respectively, and the signal equal to O at the outputs of the DAC 1A. The presence of a single signal

The invention relates to electrical measuring equipment and can be used to process location or seismic signals.

The purpose of the invention is to increase the likelihood of recognizing single and grouped pulse signals.

Figure 1 shows the functional diagram of the device in figure 2 - time diagrams of the output signals of the synchronous generator in the preparatory mode; fig.Z - then the processing mode.

The device contains an analog-to-digital converter (ADC) 1, a fast Fourier transform (FFT) unit 2, a block 3 for calculating a square of a module of discrete Fourier transform (DFT) coefficients, a switch 4, a block 5 for memory, a block 6 for dividing, a block 7 for calculating and subtracting medium, multiplication unit 8, weight function generator 9, inverse Fourier transform unit (OBPF) t t first comparison unit 11, second comparison unit 12, single and group pulse indication unit 13, digital-to-analog converter 14, registration unit 15, clock generator 16.

The output of the ADC 1 is connected to the signal input of the FFT unit 2, the first and second outputs of which are connected to the first and second signal inputs of the block 3 for calculating the square of the DFT module, the output of which is connected to the signal input of the switch 4, the first output

codes 0 correspond., t, 01 at the outputs of blocks 11-13, respectively, and a signal equal to 1 at the output of the DAC 14. The presence of group pulses will reflect code 1, 1, 10 at the outputs of blocks 11-13 and signal 2 at the output of the DAC 14. The device makes it possible to increase the probability of a correct distribution of single and group pulsed signals in the presence of pulses and the spread of the shape of elementary pulses. 3 il.

which is connected to the signal input of the memory unit 5, and the second to the first signal input of the dividing unit 6, the second signal input of which

connected to the output of the memory block 5, and the output to the signal input of the calculating and subtracting unit 7, the output of which is connected to the first signal input of the multiplication unit 8

the second signal input of which is connected to the output of the generator 9 of the weighting function, and the output to the input of the OBFT unit 10.

Output 11AP 14 is connected to the input

block 15 registration. The first output of the clock generator 16 is connected to the control input A1SCH 1 and the first control input of the FFT unit 2. The second output is connected to the second control.

the input of the FFT unit 2, the third output with the third control input of the FFT unit 2 and the control input of the square calculator 3 DFT module. The fourth output of the synchro generator 16 is connected to the first control input of the memory block 5. The fifth output is connected to a control1 {by the input of the switch .4, the otherwire is connected to the second control input of the memory unit 5, the control input of the division unit 6 and the first control input of the calculation and subtraction unit 7. The seventh output of the clock generator 16 is connected to the second control input, and the eighth to the third control input of the calculation and subtraction block 7, de3

the output is connected to the fourth control input of the calculating and subtracting average unit 7, to the control inputs of the generator 9 of the weight function and the multiplication unit 8, and first to the control input of the OBFT block 10, the second and third control inputs of which are connected to the tenth and eleventh sync generator outputs 16. The inputs of the display unit 13 for single and group pulses are connected to the outputs of blocks 11 and 12 of the comparison, and the output is connected to the input of the DAC 14. The signal input of the first comparison unit 11 is connected to the output of the IFFT block 10 The twelfth output of the sync generator 16, the thirteenth output of which is connected to the first control input of the second comparison unit 12, and the fourteenth with the second control inputs of the first 11 and second 12 comparison blocks, the signal input of which is connected to the output of the ADC 1. The input of the device is input The ADC 1, the first output - the output of the block 10 OBPF, and the second - the output of the block 13 of the display of single and group pulses.

The device works as follows.

The signal at the input of the location detector, formed, for example, in a layered medium with contrasting interfaces, can be represented as a sum of reflections from each section boundary. The task is to distinguish the signal reflected by the medium with one contrast interface (for example, air - water) from the signal reflected by the medium with two or more contrast interfaces, including re-reflections, even in that case, if the reflected pulses overlap in time. Similarly, the task may set

with and for point objects: recognition of single and group point reflecting objects.

Let the signal S (t) at the output of the linear receiver, i.e. at the input of the ADC 1 (figure 1), contains two reflected pulses S (t) and S.j (t) has the form

S (t) S, (t) + S2 (t),

where S, (t) aS (t-f),

t is the mutual shift of the pulses S, (t) and S, (t),

a - relative amplitude

(a O and UTi 1).

If the spectral function of the first pulse S (t) is written as

S, (")

“,

where u is the circular frequency

 i - is the delay S (t) relative to the beginning of the time window or observation interval for which,

then the spectral function of the second reflection S, (t) is respectively equal to

aS., (u) e J (,

Thus, the spectral function of the composite signal S (t) has

view

S (i) S (i) (Na-), (1)

then

 IS (i) I 2 IS ((j) 2 (1 1-2a cos ut + a). (2)

It follows that the time delay t can be measured with high accuracy by applying the inverse Fourier transform if the factor | S (J) M is excluded from equation (2). This can be done by having prior information about the multiplier. For the case of probing from loili media or point objects, such information can be obtained from a signal with one reflection or even using a probing signal in the case of active location. A signal with a single reflection can be obtained from a single contrast interface in a layered medium (for example, air - water) or when reflected from a single point volume.

ekta. This signal can be used as a sample or elementary oscillation calorimeter for calculating and storing the 1b8 function (i) And, where b is the coefficient taking into account the mutual spread of the amplitude of the largest elementary pulse in the reflected signal and the reference pulse. If the right side of equation (2) is then divided into this function during the processing, then we can obtain a real even spectral function

A (s)

IS (u)) l 1 + a IbS (a) 2 1b | 2

(1 + 2a cosijL + a2). (3)

The inverse Fourier transform of A (w) on the time axis gives three pulses: one positive for amplitudes of (1 + a2) / | bp and two identical symmetrically located at points with amplitudes of a / 1b | 2. The value (1 + a2) / 1b in equation (3) is the average value. If this value is determined by averaging the expressions (3) over w, and then subtracting it from the original expression, then the inverse Fourier transform ambient will remain only two pulses at the points t tt. The time interval between these pulses can be measured with high accuracy by fixing the position of the maxima of the absolute values.

In addition, if 0, then the signal S (t) is actually a single elementary pulse (1 + a) S (t). In this case, A (u) const (1 + a) and subtracting the average value from A (and) (also equal to (1 + a) 2 / lbl2) results in a zero signal level after the inverse Fourier transform.

Using an automatic gain control circuit in the receiver, it is possible to provide an approximate equality of b1 const. Therefore, in the future can be considered.

The device works as follows. .

The purpose, principle of operation and communication between the transducer 1, the blocks 2,3,6-8,10, the generator 9 fully corresponds to the same blocks of the known device. This part of the device can be used to measure I from the output of block 10 by fixing the position of the two largest absolute values of equal peaks on either side of the - point or one of the largest peaks at half the determination interval from to where N is the size of the DFT. To recognize single and grouped pulse signals, and also to fix the presence or absence of a signal at the input of block 1 of the ADC and to record all this current information, blocks 11–13, 15, and converter 14 are used.

The device works in two modes. In the first, preparatory, mode, the samples of the square of the DFT module of the sample of the elementary oscillation are written into block 5 of the memory. In

In the main mode, the reflection signal is processed. The transition from one mode to another is secured by switch 4. This transition can be carried out either automatically or manually.

The switch 4 is controlled by a synchronous generator 16 (output d). In the second mode, the measurement of the shift t between the two highest absolute pulses at the output of block 10 can be performed, which corresponds to the measurement of the shift between the two largest amplitudes of the elementary pulses in the input signal, and also records the presence and composition of the same signal by block 15 registration.

During the sample processing time, the clock generator 16 connects the input of the switch 4 to its first output. When this occurs, the cells of the block 5 of the memory of samples of the square of the sample of the DFT sample are recorded. These samples are formed as follows. In accordance with the Kotel'nikov theorem with a frequency not less than 2 times the upper frequency in the spectrum of the processed signals, samples are taken from the analog signal of the sample entering the ADC 1, the amplitude of which is converted into a binary code that is fed to the input block 2 FFT. The frequency of taking these samples is determined by the repetition period of the clock pulses received at the control input of the ADC 1 from the output of the clock generator 16. These same clock pulses control the writing of the digital samples into the input register of the FFT unit 2. During the interval for determining the signal, N samples are taken, and for the FFT algorithm, typically N is power 2. The start of the FFT is controlled by pulses from the output b of the clock generator 16 to the second control input of the FFT unit 2 after writing to its input register N- digital signal sampling. From the output registers of FFT block 2, a pair of samples corresponding to the imaginary and real parts of each DFT coefficient is read within half of the frequency window. Use for further processing only half of the DFT patients from among those falling into the frequency window with normalized frequencies f from O to M / 2, where F tN is the number that determines the width of the frequency j window (if evaluated by the number of DFT coefficients, then the width when is equal to M + 1), due to the fact that the DFT coefficients of a real signal have a clear symmetry of modulus relative to 10

or depending on the choice of the determination interval for ./2, N / 2 or O, N. The data pair for each DCF coefficient is used to calculate the square of the mod- ule 15 l DFT in block 3. This data pair goes to the two inputs of block 3, Reading each pair from the output registers of the FFT block 2 and controlling the process of calculating the square of the 20 DFT modulus the pulses coming from the output to the synchro-generator 16. The samples of the square of the DFT module, then from the output of block 3, through switch 4 arrive at the recording input of memory block 5 and are stored in its cells for the duration of the device's operation in the second mode. To control the recording, pulses are fed from the output 30 of the clock generator 16 to the first control input of the unit 5.

In the second mode of operation, the input of the switch 4 is connected to its second output, to which the input of the divisible dividing unit 6 is connected. At the input 35 of the divider of dividing unit 6, simultaneously with the readings of the square of the DFT module of the composite signal under investigation, coming from the second output of switch 4, the readings of the square of the DFT sample of the sample with the same normalized frequency f that were recorded in the first mode. Reading them from the memory block 5 and the division process in block 6 controls the sequence of pulses from the output of the synchronous generator 16. The result of the division from the output of block 6 of the division is fed to the input of block 7 of calculation 50 and subtraction of the average. In block 7, the samples are memorized with the normal and | f frequencies from O to K / 2, which are controlled by the same impulses | The output of the synchronous generator 16 and 55 is the accumulation of these samples by the accumulation method taking into account the symmetry of the DFT module for obtaining the sum of quotients of the quotient within the entire frequency window. Then, the average value is calculated by dividing the sum by the number of samples equal to M + 1 deducted from the received counts of the quotient. These operations are controlled by pulses from outputs g, h, and a synchronous generator 16, of which pulses from output g are designed to read the accumulated sum and start the dividing circuit by M + 1, and from output g they are double-digit pulses M + 1, and from the output, they are used to subtract the average from the counts of the quotient and feed the difference counts to the output of block 7. These counts then go to one of the inputs of multiplication unit 8, to the other input of which the counts of the weight function are generated, generated by the generator 9 of the weight function. From the output of block 8 multiplication, the product counts are fed to the input of block 10 OBPF and stored there in the input register. The operation of block 8, generator 9, and writing to the input register of block 10 are also controlled by pulses from the output and clock generator 16. In the input register of block 10 OBPF, the source for OBPF array is formed. Each counting from the output of block 8 is recorded due to the even symmetry of the latter, except and (for), is made in two cells at once. For the determination interval from, Nj is a pair of cells corresponding to the normalized frequencies f and N-f. For M / 2 (N-M / 2), i.e. outside the frequency window, zeros are written in the cells of this register. The start of the IFFT is controlled by pulses coming from the output of the synchro-generator 16, which begin to be fed to the second control input of the block 10 after writing to all N cells of the input register of the corresponding counts. , if you use the definition interval 0, NJ. The samples of the output data of the OBPF block 10 are read by pulses coming from the output l of the synchronizing generator 16 and are fed to the first output (Out1) of the device in binary code and to the first input of the first block 11 of the comparison. WITH

Output 1 The binary code is used to measure and record the mutual time offset.

In the first comparison block 11, the samples received at the first input from the output of block 10 IFFT are normalized by a sign by assigning a positive sign to all of them and subtracting from each normalized reference the number corresponding to the specified threshold value 110ROG-1 The code of this number comes in the form the pulse sequences from the output of the sync generator 16 to the second input of the comparator block 11 are remembered here. By the sign of the difference, the excess of the threshold level nOPOr-l is recorded. So, for example, if the + sign in the sign difference of the difference corresponds to the logical 1 sign - the logical O level, then after performing the comparison operation with the threshold level of half of the samples from n from O to N / 2 (due to the parity, this is enough) , coming from the output of block 10 OVPF in the sign bit difference, at least one 1 is recorded, if at least one exceeding the level of the Threshold is 1. If there was no such excess, then for all counts from the output of the 10 V block. the sign bit of the difference will be O. The level O or 1 in the sign bit will be fixed, remembered and fed to the output of the comparison unit 11. At the same time, if there was at least one excess of the readout from the output of block 10 of the THRESHOLD level, then at the output of block 11 there is 1, and if this excess is not for all readings, then - O. The level O or 1 from the output of the first comparison unit 11 is read out and fed to one of the inputs of the block 13. This is controlled by the pulses from the output n of the clock generator 16. I

Similar to the first comparison unit 11, the second comparison unit 12 operates. At its first input, samples of the composite signal under study are fed from the output of ADC 1 when η varies from O to N-1. To the second input of the unit 12 from the output from the sync generator 16

The code of the number corresponding to the level of the THRESHOLD-2 is received, which is stored in the register of block 12. If this level is exceeded by at least one

5 10. 15 20, 25 30. 35 40 45

55

the signal normalized by the signal from the output of the ADC 1 at the output of block 12 is fixed 1, otherwise O. The level O or 1 from the output of the second comparison block 12 is detected and fed to another input of block 13, which is done by pulses at the output n of the clock generator 16. Block 13 is a two-bit adder. The one-bit binary code on each of its inputs is the code of the corresponding term.

The inputs of block 13, the output of which is the second output (Output 2) of the device, receive binary codes for the sum of single-digit numbers from the outputs of blocks 11 and 12. This code can be further used to digitally indicate the absence of reflected pulses, as well as information about their structure. (single or group signals) if present. In addition, using the DAC 14, the digital code from the output of block 13 is converted to analog voltage (or current) with three gradation levels corresponding to O, 1 and 2. This voltage can be recorded on the tape of registration block 15.

In this situation, the absence of a signal corresponds to the codes O, O, 00 at the outputs of blocks 11-13, respectively, and the signal, equal to O, at the output of the DAC 14. Situations of the presence of a single signal correspond to codes 0,1,01 at the outputs of blocks 11-13 respectively and a signal equal to 1 at the output of the DAC 14. Situations of the presence of group pulses correspond to codes 1,1,10 at the outputs of blocks 11-13 and a signal equal to 2 at the output of the DAC 14.

Thus, the proposed device makes it possible to increase the probability of a correct distribution of single and group pulsed signals in the presence of pulses and the spread of the shape of the elementary pulses.

Claims (1)

Invention Formula A device for recognizing single and grouped pulse signals containing an analog-to-digital converter, the output of which is connected to the signal input of the fast Fourier transform unit, the first and second outputs of which are connected to the first and the second signal inputs of the computing unit of the square module of the coefficients of the discrete Fourier transform, the output of which is connected to the signal input of the switch, the first output of which is connected to the signal input of the memory unit, and the second to the first signal input of the division unit, the second signal input of which is connected to the output of the memory unit, and the output - with the signal input of the calculation and subtraction of the average, the output of which is connected to the first signal input of the multiplication unit, the second signal input of which is connected to the output A weighting function, and an output with the input of the inverse fast Fourier transform unit, a digital-to-analog converter, the output of which is connected to the input of the ufiKf 4L synchronous generator, the first output of which is connected to the control input of the analog-digital converter and the first controlling the input of the fast Fourier transform unit, the second output with the second control input of the fast Fourier transform unit, the third output with the third control input of the fast Fourier transform unit and the control input of the calculator the slope of the square of the module of the discrete Fourier transform coefficients, the fourth OUTPUT - with the first control input of the memory unit, the fifth output - with the control stroke of the switch, the sixth output, with the second control input of the memory unit, the control input of the de uttnt / flecoll at nilinHi - HHilllil tliliJ II III .J III --- IIIIM s o 5 0 5 15 0 and the first control input of the calculator and subtraction of the average, the seventh and eighth outputs - with the second and third control inputs of the calculator and subtraction of the average, the ninth output - with the fourth control input of the calculator and subtraction of the average, with control inputs of the generator the weight function and the block .. multiply, and the first control input of the inverse fast Fourier transform block, the second and third control inputs of which are connected to the tenth and eleventh respectively. sync generator outputs, so that, in order to increase the likelihood of recognizing single and grouped pulse signals, the first and second comparison units and the display unit of single and group pulses are inputted into it, are connected to the outputs of the comparison units, and the output is connected to the input of a digital-to-analog converter, the signal input of the first comparison unit is connected to the output of the inverse fast Fourier transform unit, and the first control input is connected to the twelfth clock output, thirteen fifth output is connected to a first control input of the second comparator CIM block, and the fourteenth clock output connected to second inputs of the gate of the first and second comparing units, the signal input of which is connected to the output of analog-to-digital converter. I11 (1П1 --- 11 (1И1Ш1 -f .J  t t I M / 2 i- 1} INP1 / L6SOD Illll-jiij IMMI - HIIII n uHnymcDl iiniinn iiiiniinti + Illtl llllll I T. lilliti llllll i J. () ufmyattel 1IA Mill J uhngatcol db jrrr Compiled by S. Lebedev Editor N. Shvyshchka V. Kadar Proofreader E. Sirohman Order 4119/44 Circulation 728 Subscription VNIIPI USSR State Committee for inventions and discoveries 113035, Moscow, Zh-35, Raushsk nab., 4/5 Production and printing company, Uzhgorod, st. Project, 4 r 5ЛФ Hllllllllll Illllllfiri . ll№ (H11 1 inpushvv1 inni inin Hlll.t -fc /) t t -t msei-j