RU83344U1 - UNDERWATER OBJECT DETECTION DEVICE FOR EVALUATING A MEASUREMENT RANDOM OF A HYDROLOCATOR ECHO SIGNAL - Google Patents

Abstract

A device for detecting underwater objects containing a receiver unit (hydroacoustic antenna), the function of which is to send a location signal and receive an echo signal, an analog-to-digital converter (ADC) unit, an amplifier and bandpass filter unit, whose function is to amplify the signal at the output of the sonar antenna up to levels necessary for the operation of the ADC, a block of matched filters, the function of which is to optimally receive the reflected echo signal against the background of noise, the signal generator links, power amplifier, detector, characterized in that a fractal dimension meter is connected between the matched filter block and the detector, which is capable of evaluating a measure of the randomness of the amplitudes of the echo signal at the output of the matched filter and expressing this measure as a number (fractal dimension), The detector function is to compare the received number with the threshold of signal and interference energy and make a decision: whether the desired location object is detected or not.

Description

Application area

The utility model is an electronic device and relates to the field of sonar and sonar. More specifically, the utility model can be used to search for and detect artificial underwater objects, such as sunken ships, equipment, underwater vehicles and other artificial underwater structures.

State of the art

The prior art Phase parametric sonar (patent RU 2097785), containing a serially connected synchronizer, a radio pulse generator, a power amplifier and a radiator, a phase shift meter and an indicator connected in series, two selective amplifiers, the outputs of each of which are connected to the corresponding inputs of the phase shift meter, characterized in that the driver of the control signal and the series-connected low-frequency antenna and the amplifier of the difference signals are introduced into it s frequency, which output is connected to the inputs of selective amplifiers, the first control input of which is connected to the first output of the control radio pulse generator, and the second control input to the output driver control signal, two inputs of which are connected respectively with two control outputs radio pulse generator.

The ACTIVE HYDROLOCATOR (patent RU 75061U) is known in the art. The utility model relates to sonar technology, and more particularly to the field of active sonar, including active sonar, designed to detect targets, measure coordinates and motion parameters of detected targets. The technical result of the utility model is to provide the ability to measure the depth of the target in the directional or weakly directed in the vertical plane of the reception of echo signals. To achieve the specified technical result in

an active sonar containing emitting and receiving acoustic antennas, a generator, a synchronization unit, a signal processing unit, a target echo detection unit, a target echo delay measurement unit relative to a target relative to a probe radiation time, a sound velocity vertical distribution unit and a target depth measuring unit , a unit for determining the possible target depths corresponding to the measured delay times of the echo signals from the target and the possible values of the angles of arrival of the echo signal is introduced s in the vertical plane, a unit for determining possible distances to the target corresponding to possible target depths and possible angles of arrival of echo signals in the vertical plane, a unit for determining t probable target locations (N m. Dm), the target depth being determined in the target depth measurement unit from the set depths H m.

The prior art sonar detection and classification of underwater and surface targets for surface ships (patent RU 20389U), containing a multi-element acoustic antenna, multi-channel radiation and reception paths connected to an acoustic antenna, a digital computer system containing a master oscillator connected to the radiation path, a device for digital processing of echo signals connected to the output of the receiving path and a display and control panel, characterized in that the composition of the digital processing device with latter is present forming apparatus further includes a classification probing signal connected to the input radiation path parallel master oscillator, and a digital processing device and classifying echo sounding signal classification, an input coupled to the output of the spatio-temporal processing apparatus processing digital echo signals.

The closest analogue is the METHOD FOR DETECTING THE HYDROLOCATOR ECHO SIGNAL (patent RU 2293358). The invention relates to the field of sonar and radio engineering and can be used to build sonar signal detection systems. The technical result is to increase noise immunity when detecting an echo signal in the presence of noise and interference. A method for detecting an echo signal of a sonar contains radiation of a probing signal, receiving an echo signal with a hydroacoustic antenna mixed with noise interference, sampling an electrical signal at the output of a sonar antenna, a set of sampled samples

an electrical signal of duration T, obtaining sequentially, at regular intervals over the entire time of detection of sets of discretized samples of an electrical signal and fast Fourier transform of the obtained sets of discretized samples, while for each set of discretized samples of an electrical signal, a complex spectrum is determined, the sets are shifted for a time 1 / 8T <T <1/2 T, remember each previous and each subsequent complex spectra, determine the mutual spectrum between each previous the next and each subsequent set, select the set with the maximum energy spectrum, which decide on the detection of the signal. All the solutions known above, the prototype and other known existing echo-signal detectors in the sonar compare the level of the noise and signal aggregate with the noise level, and when the specified threshold is exceeded, it is decided that the desired object has been detected [4]. To increase the detection range, a consistent technique is used, which allows obtaining a higher SIR at the output of the matched filter. The disadvantage of this approach is the fact that there is a theoretical limit to the signal-to-noise ratio (SIR) at which such detection is possible with a given probability of false alarm. This leads to a limit on the range of the detector at a given probability of false alarm.

Effect: increasing the range of the detector with a given probability of false alarm.

Brief Description of the Drawings

Figure 1 shows a block diagram of a utility model, where 1 is a receiver-emitter unit (sonar antenna) that sends a location signal and receives an echo signal, 2 - a block of amplifiers and bandpass filters that amplifies the signal at the output of the sonar antenna to levels necessary for the operation of the ADC, 3 - a block of analog-to-digital converters (ADC), 4 - a block of matched filters that provide optimal energy reception of the reflected echo signal against a background of noise, 5 - a signal generator, 6 - a power amplifier, 7 - zmeritel fractal dimension, 8 - detector.

Figure 2 shows an example of an electrical circuit of an amplifier unit (U / PF unit).

Figure 3 shows an example of an electrical circuit of a bandpass filter (block U / PF).

Figure 4 shows an example of an electrical circuit of a power amplifier in accordance with a push-pull circuit (UM unit).

Utility Model Implementation

The claimed technical result is achieved due to the fact that (see Figure 1) a device for detecting underwater objects containing a receiver-emitter unit (sonar antenna (1)), the function of which is to send a location signal and receive an echo signal, an analog-digital unit converters (ADC) (3), a block of amplifiers and band-pass filters (2), the function of which is to amplify the signal at the output of the hydroacoustic antenna to the levels necessary for the operation of the ADC, a block of matched filters (4), whose function is to optimal reception of the reflected echo signal against the background of interferences, a generator (5) of the sending signal, a power amplifier (6), a detector (8), characterized in that a fractal dimension meter (7) is connected between the matched filter block (4) and the detector (8) ), made with the possibility of evaluating the measure of randomness of the amplitudes of the echo signal at the output of the matched filter (4) and expressing this measure as a number (fractal dimension), and the function of the detector (8) is to compare the received number with the threshold of signal energy and noise and accept R Solutions: the desired location object was detected or not.

The detection device can be made in the form of a separate printed circuit board with an external transceiver antenna, in the form of several boards, or as part of a sonar complex.

The transceiver antenna (1) is located separately from the remaining units of the device to ensure direct contact with the aquatic environment. Any hydroacoustic antenna, both piezoacoustic and induction, an antenna array or a single receiving-emitting element, can act as an antenna.

Amplifier and bandpass filter (U / PF) blocks (2) and the PA power amplifier (6) are analog and should be located as close to the antenna as possible to reduce electrical noise in the detector. The ADC block (3) is the interface between the analog and digital parts of the board, and is structurally located on the digital part.

The amplifier and the band-pass filter (2) can be implemented on operational amplifiers, consist of one or more stages, and the main requirement for this unit is

receiving at the input of the ADC block a signal in a given frequency band and with a given gain. An example of such an amplifier is shown in FIG. 2.

The ADC block (3) uses standard analog-to-digital conversion chips with the required bit depth, sampling frequency, and interface. An example of such chips is the AD7814, AD9271, and so on.

The blocks of the generator (5), the matched filter (4), the fractal dimension meter (7) and the detector (8) belong to the digital part of the board and can be made in the form of a processor program or firmware of an FPGA chip.

The generator unit (5) is digital, located inside the processor and consists of the function of generating a probing signal at given times. For reliable operation of the detector, such signals can be either simple short signals or complex signals of long duration. An example of such a generator is a pseudo-random phase-manipulated signal generator according to the law of the M-sequence with a code length of 31 [6].

A power amplifier (6) is necessary to ensure maximum detection distance by providing maximum signal energy [4]. An example of the construction of a power amplifier can be the circuit in Fig.3.

The matched filter (4) is calculated to detect a signal of a previously known shape generated by the generator unit, the output signal of the filter does not match the shape of either the input or the signal for which the filter is designed (except for a pulse with an envelope of a Gaussian shape) [5] . The signal with which the filter is matched, however, if it is present in the input signal with noise, allows you to get the maximum amplitude of the filter output signal, that is, this filter maximizes the signal-to-noise ratio for a known signal. Analytically, the functioning of a matched filter can be expressed by the equation:

Where yi is the i-th sample at the filter output, xi is the data at the filter input, hi is the filter transfer characteristic, which is a reverse copy of the send signal. The solution to the problem is based on the fact that a randomness estimation block (7) of the amplitude distribution at the output of the matched filter (4) is added, and further comparison is made not with the specific signal and noise power with a threshold based on the interference energy estimate, but with the obtained randomness measure with a predetermined the threshold.

The technical result is achieved due to the fact that any random process, such as noise or an echo signal with a random phase, can be represented as a fractal object for further processing [1]. The main characteristic of a fractal object is its dimension [1, 2, 3]. Fractal dimension, as a rule, is a non-negative non-integer number, reflecting, in some way, the geometric complexity of the object, that is, it is a measure of the randomness of the process. The appearance of a certain echo signal from an artificial object in the sonar signal will change the value of the fractal dimension of the signal as a whole. This fact allows us to use the magnitude of the fractal dimension of the sonar signal to detect it. The most effective use of the fractal dimension in the detection of low-contrast extended objects against a background of strong interference, since the magnitude of the fractal dimension of the signal is practically not affected by the amplitude characteristics of the signals that form it [3].

The novelty of the claimed utility model is due to the fact that it uses a non-energy method of processing the echo signal, which allows to detect the presence of an echo signal from artificial objects on a highly noisy signal. None of the known underwater object detection devices have the described functionality.

Device essence

The detection device can be made in the form of a separate printed circuit board with an external transceiver antenna, in the form of several boards, or as part of a sonar complex.

The transceiver antenna is located separately from the remaining units of the device to ensure direct contact with the aquatic environment. Any hydroacoustic antenna, both piezoacoustic and induction, an antenna array or a single receiving-emitting element, can act as an antenna.

The U / PF and UM blocks are analog and should be located as close as possible to the antenna to reduce electrical noise in the detector. The ADC block is the interface between the analog and digital parts of the board, and is structurally located on the digital part.

An amplifier and a band-pass filter can be implemented on operational amplifiers, consist of one or more stages, and the main requirement for this unit is

receiving at the input of the ADC block a signal in a given frequency band and with a given gain. An example of such an amplifier is shown in FIG. 2.

The amplifier is built according to the scheme of the instrumental amplifier, provides amplification of the differential signal from the hydroacoustic antenna to levels suitable for further processing.

This amplifier has an input signal amplitude limiter built on capacitors C10, C31 and diodes D2, D3, D7, D8. Capacitors limit the current at a given frequency, allowing you to use the same antenna for both reception and transmission, and then the high voltage from the output of the transmitter power amplifier will not be destructive for the receiver even without the use of mechanical switches.

Diodes D2, D3, D7, D8 short-circuit any input voltage exceeding the opening voltage of these diodes to ground, which is a limiter of the input signal amplitude.

Resistors R6, R17 bind the signal to the "zero" level. Operational amplifiers U3, U6, U8 constitute a tool amplifier in which U3, U8 provide a high input impedance and a wide frequency range, and U6 is the required gain specified by resistors R3, R4, R18.

The signal from the output of the amplifier through the capacitor C20 and the resistor R13 falls on another limiter, consisting of counter-active zener diodes D6, D9, and then transmitted to the bandpass filter shown in Fig.3.

Capacitors C8, C13, C17, C23, C26, C33 are filtering capacitors for power to each leg of the microcircuit.

The band-pass filter in this example is built of two stages on operational amplifiers U16, U17, separated through a capacitor C63, R49 and a limiter on two counter-active diodes D20.

Capacitors C60, C66, C61, C67-filtering capacitors for power to each leg of the chip.

The gain and frequency characteristics of each stage are set by two capacitors and three resistors (C55, C59 and R46, R47, R51 for the first stage and C57, C64, R48, R50, R52 for the second).

The signal at the output of the bandpass filter is decoupled by a constant component using a capacitor C65, it is limited to the flow in front of the limiter by a resistor R53 and is limited to two counter-active diodes D21.

Thus, this circuit provides the operation of an active strip amplifier.

The ADC unit uses standard analog-to-digital conversion chips with the required bit depth, sampling rate and interface. An example of such chips is the AD7814, AD9271, and so on.

The blocks of the generator, matched filter, fractal dimension meter and detector belong to the digital part of the board and can be made in the form of a processor program or firmware of an FPGA chip.

The generator block is digital, located inside the processor and consists of the function of generating a probing signal at given times. For reliable operation of the detector, such signals can be either simple short signals or complex signals of long duration. An example of such a generator is a pseudo-random phase-manipulated signal generator according to the law of the M-sequence with a code length of 31 [6].

A power amplifier is needed to ensure maximum detection distance by providing maximum signal energy to the package. An example of the construction of a power amplifier can be the circuit in Fig.4.

The power amplifier shown in Fig. 4 can be built according to the push-pull scheme (two mirror power transistors operate in turn, providing bi-directional magnetic field generation on the primary winding of the step-up transformer from a constant voltage source to increase the efficiency of this transformer).

The control signal from the generator unit is fed to two inputs - ln +. In - (direct and inverse signal); diodes D7, D8 provide signal transmission to power IGBT transistors only with a given polarity and set the deadband voltage (the voltage of any input signal below 0.6 V locks the transistor); resistors R15 and R28 limit the control current of transistors, resistors R24, R25 provide a blocking voltage in the absence of a signal, resistor R26 is power and serves to control the quality factor of the filter composed of the primary winding of the transformer and capacitor C8. An excessively high Q factor distorts the amplitude of the signal at the beginning and end of the transmission.

Resistor C6 and inductance L2 serve as a filter for supplying voltage Vbat, which feeds the transformer primary winding through resistor R27. Capacitor C8 is selected in such a way as to ensure the resonant frequency of the filter together with each arm of the primary winding of the transformer at the sending frequency.

The transformer secondary is connected directly to the sonar antenna.

Thus, this circuit is an example of a power amplifier that can amplify the control signal both in voltage (due to the use of IGBT transistors in key mode, high DC voltage Vbat and the presence of step-up transformer T3), and in current (transistors in key mode provide high efficiency of the amplifier, the presence of an LC filter from the primary winding and capacitor C8 increases the transformation efficiency), and then the main energy consumer in such a system is a hydroacoustic antenna. The matched filter is calculated to detect a signal of a previously known shape generated by the generator unit, while the output signal of the filter does not coincide in shape with either the input or the signal for which the filter is designed (except for a pulse with an envelope of a Gaussian shape). The signal with which the filter is matched, however, if it is present in the input signal with noise, allows you to get the maximum amplitude of the filter output signal, that is, this filter maximizes the signal-to-noise ratio for a known signal. Analytically, the functioning of a matched filter can be expressed by the equation:

Where yi is the i-th sample at the filter output, xi is the data at the filter input, hi is the filter transfer characteristic, which is a reverse copy of the send signal. The implementation of the matched filter can be in the form of a function in the “C” language for a digital processor or in the form of firmware for a chip with programmable logic, for example, in the “C” language:

float sf (* x, * h, len)

{

float sum = 0;

for (int i = 0; i <len; i ++)

{

Sum + = x [i] * h [i];

}

return Sum;

}

Or in VHDL:

Filter: process (CLK, RESET)

Signal Sum: integer range 0 to 65535;

Begin

If RESET = '1' then

Sum <= 0;

Elsif rising_edge (CLK) then

for INDEX in len downto 0 loop

Sum <= Sum + x (INDEX) * h (INDEX);

end loop;

End;

The detector block is also digital and implements only one comparison of input data with a threshold:

1 - object detected, 0 - not detected.

The principal difference of the proposed device is the use of a fractal dimension measuring unit instead of directly connecting the detector to a matched filter. This unit evaluates the measure of randomness of the amplitudes of the echo signal and expresses this measure in the form of a number, which becomes possible when using the concept of fractal dimension:

Where z is the amplitude, N (z) is the density distribution of the amplitudes in the echo signal, calculated over the analysis window (i = l..N) as follows:

where N is the length of the analysis window, xi are the instantaneous values of the amplitude of the echo signal, ε is the allowable error level of the analysis of the amplitude (resolution of the histogram). That is, for a digital discrete signal, N (z) is a histogram of amplitudes. This unit is implemented as a procedure on any DSP (digital signal processing processor), for example ADSP2185, ADSP21363, TMS320C6xxx, TMS320C5xxx, or simply a CPU (digital processor), for example Intel IXP420, which has sufficient speed and implements calculations using the above formula, or on a chip with programmable logic, for example Altera EP2C8C144, EP1C12Q244 or any other that has the required number of cells and speed. Then the calculation of the histogram in the language "C" will look like this:

Histogram (* histogram, * x, Zsteps, N)

{

for (int i = 0; i <Zsteps; i ++)

histogram [o] = 0;

for (int i = 0; i <N; i ++)

{

int index = floor (x [i] * Zsteps / MAX_Z);

histogram [index] ++;

}

}

A calculation of fractal dimension as an example may look like this:

FractalD (* histogram, Zsteps, e)

{

float D = 0;

for (int i = 0; i <Zsteps-8; i ++)

{

floatdNl.dN2.dN4;

dNI = histogram [i + l] -histogram [i];

dN2 = histogram [i + 2] -histogram [i];

dN4 = histogram [i + 8] -histogram [i];

if ((dN2 / dN1 <e) & (dN4 / dN2 <e))

D = atan (log (abs ((histogram [i + 8] -histogram [i]))) / 8);

}

Return D;

}

The principle of operation of the module

Measurement of the fractal dimension of various, natural and artificial objects is an ambiguously solved problem, since there is no exact definition of the concept of fractal dimension [1] and, as a rule, there is no necessary amount of data (signal with infinite sampling frequency and infinite length). Therefore, the fractal dimension of any formation is measured indirectly - by the slope of the dependence S = F (d), where S is the measured value, and d is the scale. Due to the fact that the sonar signals for coherent processing are narrow-band in order to exclude the scattering of different frequency components along different propagation paths, the most informative point of view of processing is the amplitude of the signal at the output of the matched filter. In this device, the generator sends a pulse signal through a power amplifier and a receiving-emitting antenna to the marine environment. A copy of the send signal goes to the matched filter in order to provide the maximum SIR at the filter output. The signal reflected from the location object hits the antenna, amplifies, passes through a band-pass filter to reduce the level of interference, gets to the ADC unit and then to the matched filter.

The fractal dimension measuring unit evaluates the statistical distribution of signal amplitudes at the output of the SF and expresses this estimate as a number, that is, a fractal dimension.

The detector compares the received number with the threshold and makes a decision whether the desired location object is detected or not.

INFORMATION SOURCES:

1. Payten H.O., Richter P.H. The beauty of fractals. Images of complex dynamic systems: Per. from English - M., World, 1993

2. Mandelbrot B.B. The Fractal Geometry of Nature. - N.Y .: Freerman, 1982

3. Cronover P.M. Fractals and chaos in dynamical systems: Per. from English - Moscow, Postmarket, 2000

4. Evtyutov A.P., Mitko V.B. Engineering calculations in sonar - Leningrad, Shipbuilding, 1988

5. Barton D., Ward G. Handbook of radar measurements: trans. from English - M., Soviet Radio, 1976

6. Kazarinov Yu.M. Radio engineering systems. - M., High School, 1990

Claims (1)

A device for detecting underwater objects containing a receiver unit (hydroacoustic antenna), the function of which is to send a location signal and receive an echo signal, an analog-to-digital converter (ADC) unit, an amplifier and bandpass filter unit, whose function is to amplify the signal at the output of the sonar antenna up to levels necessary for the operation of the ADC, a block of matched filters, the function of which is to optimally receive the reflected echo signal against the background of noise, the signal generator links, power amplifier, detector, characterized in that a fractal dimension meter is connected between the matched filter block and the detector, which is capable of evaluating a measure of the randomness of the amplitudes of the echo signal at the output of the matched filter and expressing this measure as a number (fractal dimension), The detector function is to compare the received number with the threshold of signal and interference energy and make a decision: whether the desired location object is detected or not.