RU2739478C1 - Method for processing a pseudo-noise signal in sonar - Google Patents

Abstract

FIELD: hydro acoustics.

SUBSTANCE: invention relates to hydroacoustics and can be used for constructing systems for automatic detection of echo signals received by a sonar on a background of noise and reverberation interference, and measurement of object parameters when using pseudo-noise signals. Disclosed method of processing a pseudo-noise signal in sonar comprises emitting a probing pseudo-noise signal, receiving echo mutually correlation processing, the determination of the maximum correlation response, determining speed and distance of located object, determine the mean value of the frequency emitted F_em probing signal, signal reception is carried out with a static fan of directional characteristics crossing at a level of not less than 0.7 of the maximum, a set of temporal realizations of the echo signal with duration T is carried out simultaneously on all characteristics of directivity, correlation coefficient between time implementations of successive adjacent spatial channels is determined, pairs of adjacent spatial channels are selected, between which correlation coefficient exceeds 0.5, selecting a pair of channels with a maximum correlation coefficient, determining a cross-correlation function between channels of the selected pair, determining a maximum of the cross-correlation function, determining the time position of the selected maximum and its amplitude, determining the area of positive readings of the maximum of the correlation function, determining the zone of negative readings before the beginning of the zone of positive readings and the zone of negative readings after the zone of positive readings, boundary readings having different polarity are determined, the period of the carrying autocorrelation function is measured and the carrier frequency of the received echo signal is determined, after which the radial velocity is determined.

EFFECT: proposed method allows to reduce processing time of a pseudo-noise signal and to increase its efficiency when used in sonar for detecting objects and measuring parameters.

1 cl, 2 dwg

Description

The invention relates to the field of hydroacoustics and can be used to build systems for automatic detection of echo signals received by the sonar against the background of noise and reverberation interference and measurement of object parameters using pseudo-noise signals.

A known method for detecting an echo signal, considered in the book L. Rabiner, B. Gould "Theory and application of digital signal processing", Mir, Moscow, 1978 The method contains the emission of a probe signal with duration T at a known frequency; echo reception; sampling of the input signal, a set of input sampled samples of duration T; determination of the energy spectrum using the fast Fourier transform (FFT), shifting the input signal set in time, repeating the procedure for a set of time-shifted input sampled samples of duration T and determining the energy spectrum, selecting the set with the maximum signal energy, deciding on the detection by set with maximum signal energy.

The disadvantage of this method is that it uses a long tone signal and at the output the spectrum of the echo signal is determined, which can be used to determine the approach speed, but it is impossible to determine the length of the echo signal, which is one of the main classification features.

Known methods of detecting and measuring parameters of echo signals from objects using the so-called complex signals, the processing of which at the output is formed the correlation function of the echo signal and the emitted probe signal. These methods have found application in radar and hydroacoustics (R. Benjamin Analysis of radio and sonar signals Voenizdat M. 1969). The main properties of these signals are determined by the type of internal modulation, which determines the form of the uncertainty function. For classification problems in sonar, the most interesting are pseudo-noise signals, which have a push-button uncertainty function, which provide good resolution in time and speed. (V.A. Zaraisky, A.M. Tyurin Theory of sonar location published by VMAOLU, L. 1975 p. 242). As a rule, such signals are processed using multichannel correlators (ibid., P. 255) or multichannel matched filters (ibid., P. 333). If the sonar is stationary and the target is stationary, then a correlation is made between the emitted signal and the received echo signal. However, such a situation practically never occurs, and due to the own movement and the movement of the target, the spectrum of the reflected signal is shifted in accordance with the Doppler effect (ibid., P. 200), as a result of which the spectra do not coincide and a folded correlation function is not formed. For this reason, multichannel processing is used in reception, where each channel corresponds to a certain target speed.

The closest in terms of the number of common features with the proposed invention is the method of processing a complex signal given in the book (V.A. Zaraisky, A.M. Tyurin "Theory of sonar" ed. VMAOLU, L. 1975 p. 255)

The method for processing a complex signal contains the following operations: emission of a probing complex signal, generation of M-reference signals, the center frequency of which is shifted in frequency relative to the emitted signal by the value K, reception of an echo signal, determination of M correlation functions between the echo signal and each of the M-reference signals, measurement the amplitudes of the correlation functions, the choice of the correlation function with the maximum amplitude, the determination of the temporal position of the maximum of the correlation function to determine the distance, the determination of the reference signal number to determine the speed, the display of the result on the indicator.

The disadvantage of this processing method is that to obtain the potential properties of the pseudo-noise signal, cross-correlation processing of the received signal with a reference signal is required, which must be an exact copy of the received signal. For this purpose, a bank of reference signals is used, each of which is shifted relative to each other by an amount equal to half of the frequency correlation interval. Therefore, it is necessary to carry out cross-correlation processing of the input dial-up time realization with all the reference signals available in the bank of copies, which requires a lot of time to obtain the final result.

The object of the invention is to reduce the processing time of a pseudo-noise signal in sonar tasks.

The technical result of the claimed invention is to reduce the processing time and increase the efficiency of the obtained result of processing a pseudo-noise signal when using it in sonar to detect objects and measure parameters.

The specified technical result is achieved by the fact that in the known method of processing a pseudo-noise signal in sonar, containing the emission of a probing pseudo-noise signal, receiving an echo signal, cross-correlation processing, determining the maximum correlation response, determining the speed of the target object and the distance, new features are introduced, namely, the average value the frequencies of the emitted probing signal F _rad , the signal is received by a static fan of directional characteristics intersecting at a level of at least 0.7 of the maximum, a set of time realizations of an echo signal of duration T is performed simultaneously for all directivity characteristics, the correlation coefficient between time realizations of successive adjacent spatial channels , pairs of adjacent spatial channels are selected, between which the correlation coefficient exceeds 0.5, of which a pair of channels with the maximum value of the correlation coefficient is selected , determine the cross-correlation function between the channels of the selected pair, determine the maximum of the cross-correlation function, determine the time position of the selected maximum and its amplitude, determine the zone of positive readings of the maximum of the correlation function, determine the zone of negative readings before the beginning of the zone of positive readings and the zone of negative readings after the zone of positive readings , determine the boundary samples having different polarity, measure the amplitude A _{1 of the} sample having negative polarity, measure the time value of this sample t ₁ , measure the amplitude of the A2 sample having a positive polarity following the negative sample, measure the time value belonging to this sample t ₂ , calculate the time position of the zero amplitude of the signal of the carrier autocorrelation function according to the formula T ₁ = A ₂ D / (A ₁ + A ₂ ), where D is the sampling interval, the amplitude A _{3 of the} sample is measured, having a negative polarity from the zone following the positive zone, measure the time value of this count t ₃ , measure the amplitude of the count A ₄ having a positive polarity preceding the negative count, measure the time value belonging to this count t ₄ , calculate the time position of the zero amplitude of the signal of the carrier correlation function according to the formula T ₂ = A ₃ D / (A ₃ + A ₄ ), calculate the half-period of the carrier frequency of the correlation function by the formula P ₁ = T ₂ -T ₁ , determine the carrier frequency of the signal by the formula F _nes = 1 / 2P ₁ , determine the speed of the object by the formula V _rad = (F _nes -F _rad ) / 0.69F _from , where 0.69 has the dimension of Hz / knot. KHz. (F _from in KHz, F _notc and F _emit in Hz, V _rad in _knot .)

To improve the accuracy of the carrier measurement, you can increase the sampling rate of the input signal, which will decrease the interval between samples and increase the accuracy of determining the carrier frequency.

The essence of the proposed technical solution is as follows. In the prototype, to determine the echo signal, cross-correlation processing of the received signal with a set of reference signal banks is used, which requires significant computing costs. Significantly less time is needed to calculate the correlation coefficient between two time-domain realizations. The correlation coefficient is determined according to the standard procedure. If the time samples of the first spatial channel X _i and the time samples of the second spatial channel Y _i , then the correlation coefficient QC is determined by the following known relationship.

S_x S_y - соответствующие средние квадратичные отклонения, Х_ср и Y_cp - среднее по временным отсчетам.where X _n are the samples of the first spatial channel of the Sl (n, 1) array, Y _n are the samples of the second spatial channel of the Sl (n, 2) array,

S _x S _y - the corresponding standard deviations, X _cf. and Y _cp - the average over time readings.

Then the maximum value of the correlation coefficient is determined. The determination of the correlation coefficient does not require multi-channel cross-correlation processing and is a standard procedure for all computing devices. Almost all of these procedures can be implemented on modern computers and laptops, in which the computing programs Matlab, Matsard and others are implemented (AB Sergienko Digital signal processing SPb. "BHV - Petersburg" 2011).

The presence of the maximum correlation coefficient between adjacent spatial channels uniquely determines the temporal implementation with an echo signal. This eliminates the need for multi-channel cross-correlation processing. Now the question arises in determining the radial speed and distance.

Since the spatial channels intersect at a level of 0.7, the received echo signal in these channels has the same duration, the same frequency and the same structure. Therefore, it is possible to define one cross-correlation function between adjacent spatial channels. The signals in adjacent spatial channels are practically the same, and therefore the cross-correlation function is practically the autocorrelation function (ACF) of the received echo signal, which will simplify the consideration.

The correlation function is a short-term function, the duration of which is determined by the bandwidth of the spectrum of the studied probe signal. It is known (J. Bendat, A. Pirsol Applications of correlation and spectral analysis. Transl. From English. Moscow, "Mir", 1983, p. 71) that the signal parameters are related by the ratio: the width of the main maximum of the correlation function of the received signal (the first zero of the envelope ) is inversely proportional to the width of the spectrum of the received signal ΔT = 1 / ΔF, and the correlation function itself, adopted by the adjacent spatial channels X ₁ and X _2, is determined by the expression:

где W_{x1, x2} взаимные спектры. После несложных преобразований получим корреляционную функцию принятого сигнала

where W _{x1, x2 are} mutual spectra. After simple transformations, we obtain the correlation function of the received signal

где ω_в и, соответственно, f_в - верхняя граничная частота спектра эхосигнала, ω_н, и, соответственно, f_н - нижняя граничная частота спектра эхосигнала. Эти частоты определяют полосу принятого эхосигнала. Аргумент функции

определяет среднюю несущую частоту f_cp принятого эхосигнала объекта.

where ω _in and, accordingly, f _in is the upper boundary frequency of the echo signal spectrum, ω _n , and, accordingly, f _n is the lower boundary frequency of the echo signal spectrum. These frequencies determine the bandwidth of the received echo. Function argument

determines the average carrier frequency f _{cp of the} received object echo.

, а другая - средней частотой спектра сигнала

Составляющая, определяемая полосой сигнала, является огибающей функции В(τ), а составляющая, определяемая средней частотой, является несущей частотой функции. При движении цели ее частота изменяется в зависимости от скорости собственного движения и от скорости цели, что приведет к смещению спектра отраженного сигнала относительно спектра излученного сигнала F_изл.Thus, the correlation function contains two components, one of which is determined by the signal spectrum band

, and the other is the average frequency of the signal spectrum

The bandwidth component is the envelope of the function B (τ), and the center frequency component is the carrier frequency of the function. When the target moves, its frequency changes depending on the speed of its own movement and on the speed of the target, which will lead to a shift in the spectrum of the reflected signal relative to the spectrum of the emitted signal F _rad .

Разрешающая способность процедуры БПФ определяется длительностью входной временной реализации процесса. Поскольку длительность корреляционной функции мала, то и спектральные отсчеты на выходе БПФ будут иметь плохое разрешение. Это не позволит измерить радиальную скорость. Поэтому предлагается измерить не саму несущую частоту корреляционной функции, а длительность полупериода

корреляционной функции принятого эхосигнала, который однозначно связан с несущей частотой принятого эхосигнала, и по нему рассчитывать радиальную скорость.It is very difficult to measure the value of the average frequency of the correlation function using the FFT procedure, since the duration of the correlation function is determined by the signal bandwidth

The resolution of the FFT procedure is determined by the duration of the input time implementation of the process. Since the duration of the correlation function is short, the spectral samples at the FFT output will also have poor resolution. This will prevent radial velocity measurements. Therefore, it is proposed to measure not the carrier frequency of the correlation function itself, but the duration of the half-period

the correlation function of the received echo signal, which is uniquely associated with the carrier frequency of the received echo signal, and from it calculate the radial velocity.

The result of determining the correlation function at the output of the processing system is a sequence of time samples (Fig. 1). By the location of these samples and by measuring the amplitudes of these samples, the half-period of the carrier frequency of the correlation function of the received echo signal is determined. The correlation function is an aperiodic function that has a global maximum, relative to which the positive half-cycle of the carrier frequency is located. Determine the maximum amplitude in the zone of positive samples of the carrier frequency half-period. After that, you can determine the temporary position of the transition points from a positive value to a negative value, which corresponds to the temporary position of the half-period. For this, the zone of negative readings is determined before the beginning of the zone of positive readings and the zone of negative readings after the zone of positive readings. The boundaries of the half-period should be within these limits. To find these boundaries (Fig. 1) determine the amplitudes A ₁ and A ₂ , corresponding to the readings t ₁ and t ₂ , and A ₃ and A ₄ , corresponding to the readings t ₃ and t ₄ , which makes it possible to determine the time position of the transitions of the correlation function through zero T ₁ and T ₂ and the half-period of the correlation function P, which determines F _carried . Ultimately, V _rad , determined by the formula V _rad = (F _nes -F _rad ) / 0.69F _from , where 0.69 is a coefficient that takes into account the Doppler effect; 0.69 has a dimension of Hz / knot, KHz. (F _from in KHz, F _carried and F _erad in Hz, V _rad in knot.) (J. Warren Horton. "Basics of sonar". Sudpromgiz. 1961, p. 452).

The essence of the invention is illustrated in FIG. 1 and FIG. 2, with FIG. 1 shows a sequence of measurements of the half-period of the carrier autocorrelation function, FIG. 2 block diagram of a device implementing the method,

Antenna 1 (Fig. 2) with a static fan of directivity characteristics is connected in series with a receiver 2 of echo signals and a special processor 3, which includes a series-connected unit 4 for determining the correlation coefficients, a unit 5 for selecting spatial channels with the highest correlation coefficient, a unit 6 for determining the correlation function between the selected channels, block 7 for measuring the zone of positive samples, block 9 for measuring the carrier frequency of the correlation function, block 10 for determining the speed and distance. The device (Fig. 2) also includes a control and display unit 11, the input of which is connected to unit 10, the first output through unit 12, the PShS generator is connected to antenna 1, and the second output to unit 10. The second output of unit 6 is connected to unit 8 with the second input of block 9, and the third output of block 6 is connected to the second input of block 10.

Principles of digital transformation and processing are given in sufficient detail in the work ("Application of digital signal processing" p / p Oppenheim M. Mir 1980, pp. 389-436.). At present, almost all hydroacoustic equipment is performed on special processors that convert the acoustic signal into digital form and digitally generate directional characteristics, multichannel processing and signal detection, as well as correlation processing and procedures for its analysis. The implementation of special processors is discussed in sufficient detail in the book by Yu.A. Koryakin, S.A. Smirnov, G.V. Yakovlev "Ship sonar equipment" St. Petersburg "Science" 2004 p. 281.

It is expedient to demonstrate the implementation of the proposed method using the example of the device operation (Fig. 2).

From the control unit 11 and the display signal is supplied to the generator 12 ECP which generates a probe signal and pseudonoise emits its transceiver radiating antenna 1. Simultaneously, the control unit 10 determining the distance and speed transmitted average value of radiation frequency in Hz F _rad. The reflected echo signal from the object is received by the static characteristics of the antenna 1 and is transmitted to the receiver 2 of echo signals, where the analog signal received by the antenna 1 is converted into digital form, after which the temporal realizations of the spatial channels are transmitted in successive time portions to the special processor 3. In block 4 of the special processor 3 determining the correlation coefficient (CC) between successive temporal realizations of adjacent spatial channels. In block 5, spatial channels are determined for which the CC has exceeded the threshold value of 0.5, and the spatial channels with the maximum value of the correlation coefficient are selected among them. Since the spatial channels intersect at a level of 0.7, the amplitude values of the echo signals will differ slightly from each other, and their structure will be identical. Therefore, their cross-correlation function will be practically the autocorrelation function of the ACF of the received echo signal. The obtained correlation function is transmitted to the unit 6 for measuring the zone of positive samples, where the maximum positive samples are selected, the extreme positive samples are determined, their temporal position and amplitude are measured, and transmitted to the unit 9 for measuring the carrier frequency. At the same time, the same correlation function is transmitted to the block 8 for determining the zone of negative samples, where the amplitude values and time positions of the extreme negative samples adjacent to the zone of positive samples are determined and transmitted to the second input of the block 9 for measuring the carrier frequency of the correlation function. The procedure for measuring the positive half-period of the carrier frequency of the autocorrelation function is shown in FIG. 2. Based on the measured values A ₁ , t ₁ , A ₂ , t ₂ , A ₃ , t ₃ , A4, t ₄ , the time positions T ₁ and T _{2 are} calculated and the positive half-period of the bearing correlation function is determined. The resulting half-cycle value is used to calculate the carrier frequency F _carried in hertz. This value is passed to block 10 to calculate the radial velocity of the formula V _rad = (F _{rad carried} -F) / 0,69F _of which has a dimension of 0.69 Hz / kt. kHz. (F _out in kHz, F and F _{rad carried} in Hz, V _rad in kt.) (J. Warren Horton Fundamentals sonar Sudpromgiz L. 1961, pp. 452), and the distance is determined by the temporal position of the maximum. Typically, the sampling frequency in block 2 of the echo receiver is selected 2 times higher than the upper frequency of the received signal, taking into account the speed of the target. To improve the accuracy of measuring the carrier frequency, you can increase the sampling frequency in block 2, which will reduce the interval between samples in the zone of positive samples and in the zone of negative samples and increase the accuracy of determining the carrier frequency. In this case, the very width of the zone of positive readings and the zone of negative readings will not change.

Thus, it is possible to significantly reduce the amount of computational operations using sequential determination of the correlation coefficient between temporal realizations of adjacent spatial channels, which allows finding a temporal implementation with an echo signal from a target, determining the correlation function of the received echo signal, and determining the average frequency of the received echo signal, and then measuring the speed and distance ...

Claims (1)

A method for processing a pseudo-noise signal in sonar, containing the emission of a probing pseudo-noise signal, reception of an echo signal, cross-correlation processing, determination of the maximum correlation response, determination of the speed of the positioned object and the distance, characterized in that the average value of the frequency of the emitted probe signal F _rad , signal reception a static fan of directivity characteristics intersecting at a level of at least 0.7 of the maximum is carried out, a set of temporal realizations of an echo signal of duration T is carried out simultaneously for all directivity characteristics, a correlation coefficient is determined between temporal realizations of successive adjacent spatial channels, pairs of adjacent spatial channels are selected, between which the correlation coefficient exceeds 0.5, of which a pair of channels with the maximum value of the correlation coefficient is selected, the cross-correlation function between the channels of the selected pair is determined , determine the maximum of the cross-correlation function, determine the time position of the selected maximum and its amplitude, determine the zone of positive readings of the maximum of the correlation function, determine the zone of negative readings before the beginning of the zone of positive readings and the zone of negative readings after the zone of positive readings, determine the boundary readings having different polarity, measure the amplitude A ₁ reference frame having a negative polarity, measured time value of the reference t _1, measuring the amplitude of the reference frame A ₂ having a positive polarity following the negative count, measured time value belonging to this count t _2, calculate the time position of the zero amplitude carrier signal correlation function according to the formula T ₁ = A ₂ D / (A ₁ + A ₂ ), where D is the sampling interval, measure the amplitude A _{3 of the} sample having negative polarity from the zone following the positive zone, measure the time value of this sample t ₃ , measure amplitudes at the count A ₄ having a positive polarity preceding the negative count, measure the time value belonging to this count t ₄ , calculate the time position of the zero amplitude of the signal of the carrier correlation function according to the formula T ₂ = A ₃ D / (A ₃ + A ₄ ), calculate half cycle of the carrier frequency correlation function from the formula P ₁ = T ₂ -T _1, the carrier frequency signal is determined by the formula F _nec = 1 / 2n _1, velocity of the object is determined by the formula V _rad = (F _nec -F _rad) / 0,69F _of where 0.69 is Hz / kt. kHz (F _from in kHz, F _nec and F _emit in Hz, V _rad in knots).

Images