RU2521127C2 - Method for real-time determination of depth during bottom topography survey with side-scan sonar - Google Patents

Abstract

FIELD: physics, acoustics.

SUBSTANCE: invention relates to hydroacoustics. The method for real-time determination of depth during bottom topography survey with side-scan sonar, with subsequent reconstruction thereof, includes measuring the delay time of in-phase signals of bottom reverberation received by two antennae, spaced apart on the vertical by several wavelengths of elastic vibrations, and resolving measurement uncertainty, calculating depth, wherein to achieve the technical result, the instantaneous frequency of the signal in a lower channel is recorded each time phases interfering signals match, measuring the delay time of the appearance of a signal in the upper channel with the same instantaneous frequency value; multiplying the measured value of delay time with the value of the operational frequency of the interferometer; determining the sequence numbering of the series of measurements of the delay in arrival of in-phase signals in the period of each probing in real time; calculating depths which correspond to each interference band; and subsequent reconstruction of the bottom topography based on the depth measurements includes estimating representativeness (significance) critical points of the topography by presenting a smooth continuous surface of the bottom topography using a Kronrod-Rib tree. The side-scan sonar includes a probing pulse generator 1, a receiving-transmitting upper antenna 2, a time t_n measuring circuit 3, a receiving-transmitting lower antenna 4, an interferometer 5, a time Δt_n measuring circuit 6, a selector 7, a frequency detector 8, a reference generator 9, a storage device 10, an amplitude comparing circuit 11, a frequency detector 12, a computer 13, an antenna switch 14, a display and recording device 15.

EFFECT: high accuracy of determining depth using side-scan sonar and subsequent reconstruction of the bottom topography based on the depth measurements.

7 dwg, 2 tbl

Description

The invention relates to the field of hydroacoustics, and more particularly to methods for measuring depth with subsequent spatial interpolation of the restoration of the seabed topography with discrete depth measurements using hydroacoustic means, and can be used to measure the parameters of the aquatic environment in solving applied problems, including analysis of wind fields, radiological analysis and chemical pollution, topographic studies, etc.

There is a method of determining depths away from the vessel by means of a side-scan sonar (interferometer) by measuring the ranges and directions to the reflection elements of the bottom topography based on the use of the phenomenon of interference of bottom reverb signals received on two antennas spaced vertically by several wavelengths of elastic vibrations, excited by one of the antennas of the interferometer. As a result of isolating and fixing the arrival delay time at both receiving antennas of a series of common-mode signals, for example, on a recorder, a family of interference strips is formed containing information about the directions of arrival of a series of common-mode signals and their slant ranges to the bottom (On some features of processing interference echograms // Notes on hydrography. L .: GUNiO of the Ministry of Defense of the USSR, No. 203, 1979, pp. 11-16, [1]. Tarasyuk Yu.F. Remote measurement of depths // Shipbuilding Abroad, 1975, No. 1, pp. 104-105 , [2]. Naumov EA Use of the phenomenon of inter sound waves ference for determining the bottom inclination angle // Navigation. Collection of scientific works of Leningrad State University of Economics named after SO Makarov, 1974, issue 14, p. 252-255, [3]).

Depths Z _n and the corresponding horizontal distances Y _n from the antennas of the interferometer are calculated by the formulas [1]:

$$Z_n=D_n\sin {\Theta }_n=C{t}_n\cdot n/2N=D_n\cdot n/N,\left(\mathrm{one}\right)$$

$$Y_n=D_n\cos {\Theta }_n=D_n\cdot \sqrt{\mathrm{one}}-{\left(n/N\right)}^2\text{,}\left(2\right)$$

where Θ _n are the angles from the horizon to the direction of arrival of the in-phase signals;

D _n - the inclined range to the bottom in the direction of arrival of a number of in-phase signals;

C is the speed of sound propagation in water;

n is the serial number of the interference band;

N = b / λ is the maximum number of a number of recorded interference bands;

λ is the wavelength of acoustic vibrations;

b is the magnitude of the base vertical antenna spacing;

t _n - delay time of arrival of a number of common-mode signals.

It follows from formulas (1) and (2) that the angles Θ _{n are} functionally related to the numbers of interference fringes that are not directly determined, which makes it possible to only qualitatively assess the nature of changes in depth away from the vessel, and not to quantify the depths.

There is also a method for determining the numbers of interference fringes, based on visual decoding of the pre-calculated coincidences of fringes with given numbers from two families of interference fringes obtained, for example, when using two different but similar in magnitude antenna bases (On some features of processing interference echograms // Notes on hydrography. L .: GUNiO of the Ministry of Defense of the USSR, No. 203, 1979, pp. 11-16, [1]).

The main disadvantage of this method is the impossibility of determining depths directly during measurements, as well as the high probability of an error per unit in the digitization of a number of interference bands when visually decrypting the matching bands on a noisy echogram during cameral processing. An error in the digitization of the bands per unit leads to unacceptable errors in the determination of Z _n and Y _n . In addition, the low productivity of the method is obvious due to the need to obtain two families of interference bands and the high complexity of cameral processing and, as a result, the low reliability of the restoration of the bottom topography from the measured depths using a side-scan sonar.

The objective of the proposed technical solution is to increase the reliability of determining depths using a side-scan sonar and subsequent restoration of the bottom topography from the measured depths using a side-scan sonar.

The problem is solved due to the fact that in the method for determining depths in real time when examining the bottom topography with a side-scan sonar and then restoring it, including measuring the delay time of the common-mode bottom reverb signals received by two antennas spaced apart vertically by several wavelengths of elastic vibrations , and the resolution of the ambiguity of measurements, the calculation of depths, in which, unlike the prototype, at each coincidence of the phases of the interfering signals, they instantly the specified value of the signal frequency in the lower channel, measure the delay time of the appearance of the signal in the upper channel with the same instantaneous frequency, the measured value of the delay time is multiplied by the value of the operating frequency of the interferometer, determine the ordinal numbering of a number of measurements of the arrival delay of common-mode signals during each sounding in real scale time, depths are calculated corresponding to each interference band, and upon subsequent restoration of the bottom topography from the measured depths, an estimate the representativeness (significance) of critical relief points by representing a smooth, continuous bottom surface relief of the Kronrod-Riba tree.

New distinctive features, namely, that at each coincidence of the phases of the interfering signals, the instantaneous value of the signal frequency in the lower channel is recorded, the delay time of the appearance of the signal in the upper channel with the same instantaneous frequency is measured, the measured value of the delay time is multiplied by the value of the operating frequency of the interferometer, they determine the serial numbering of a series of measurements of the delay in-phase arrival of signals in the period of each sounding in real time, the depths are calculated, respectively which are associated with each interference band, and during subsequent restoration of the bottom topography from the measured depths, the representativeness (significance) of critical points of the topography is assessed by presenting a smooth continuous surface of the bottom topography by the Kronrod-Riba tree, and this allows achieving the technical result by performing additional measurements of the time of the relative delay of the arrival of signals from the same instantaneous frequencies to the separated antennas (Δt _n ), while the relative delay of the signal s will always be observed in the channel of the upper antenna.

To increase the reliability of measurements of Δt _n under interference conditions, they are performed within a strobe of duration τ = b / s, and the measurement results for each interference band are accumulated over several sensing cycles and averaged. Improving the accuracy of measurements of Δt _{n is} ensured by the fact that at the moment of isolation of the common-mode signals, the instantaneous signal frequency is fixed on the lower antenna at the time of formation of the nth band and the relative delay is determined by the delay time of arrival of the signal of the same frequency to the upper antenna.

The delay time is measured in each lane. The basis for this is the well-known theoretical position that the bottom reverb signal at the antenna input is a narrow-band Gaussian noise with the normal distribution of the instantaneous frequency within the signal spectrum. The spectral width of the bottom reverb signal is approximately equal to the spectral width of the emitted pulse. The probability density of the instantaneous frequency values of such a signal at the output of the receiver pre-amplifier with a high degree of accuracy also corresponds to the normal distribution law (Olshevsky VV Statistical methods in sonar. L .: Sudostroenie, 1983, p.100-130).

The width of the spectrum of the bottom reverb signals received by the interferometer depends on the duration of the emitted pulses and lies in the range of 1500 ÷ 500 Hz. The instantaneous signal frequency takes the value f _T ± (0 ÷ 750) · 10 ^-3 kHz, distributed according to the normal law. The dispersion of the process is determined by the signal to noise ratio.

Provided that the distance between the antennas is much less than the signal correlation interval (usually b <50λ), it can be argued that the difference in phase of the common-mode signals is proportional to the delay time of the appearance of the signal in the upper channel with an instantaneous frequency fixed in the lower channel at the time of gating. It's obvious that $$\Delta t_n=b/c\sin {\Theta }_n\text{.}\left(3\right)$$

We substitute in (3) the value sin Θ _n from formula (1), we obtain $$\Delta t_n=bn/cN.\left(\mathrm{four}\right)$$

From here $$n={\mathrm{<figure-callout\; id="0"\; label="F"\; filenames="00000008.png,00000010.png"\; state="\{\{state\}\}">F</figure-callout>}}_0\Delta t_n\text{,}\left(5\right)$$

where F ₀ is the operating frequency of the interferometer.

In the case of obtaining fractional values of n, due to the measurement error, the obtained numbers are rounded to the nearest integer.

The criterion for the reliability of digitization is the regularity of the received serial numbers. Individual outliers can be immediately corrected without repeated measurements.

The essence of the proposed technical solution is illustrated by drawings.

Figure 1. Block diagram of side-scan sonar. The side-scan sonar includes a probe pulse generator 1, an upper-transmitting and receiving antenna 2, a t _n time measuring circuit 3, a lower receiving and transmitting antenna 4, an interferometer 5, a Δt _n time measuring circuit 6, a selector 7, a frequency detector 8, and a reference oscillator 9 , a storage device 10, an amplitude comparison circuit 11, a frequency detector 12, a calculator 13, an antenna switch 14, a display and documentation device 15.

Figure 2. An example of obtaining persistence intervals for a one-dimensional curve. 16 - sectional levels, 17 - sectional sections, 18 - stability intervals.

Figure 3. Surface graph: a) - corresponding contours of heights and the Kronrod-Riba tree; b) - 18 critical points.

Figure 4. A vertical view of the surface of a Kronrod-Riba tree of FIG. 2a. The first digit in the rectangle is the critical point number in Fig. 26, the second is the height of the critical point.

Figure 5. The Kronrod-Reeb tree for representative points of the surface depicted in figa.

6. The Kronrod-Reeb tree for a noisy surface depicted in FIG. 2a. (88 critical points).

7. The Kronrod-Rib-tree for the surface depicted in Fig. 2a, and the Kronrod-Rib-tree for the noisy surface.

The proposed method is implemented by means of a side-scan sonar (Fig. 1). From the generator 1 of the probe pulses at time t _{0, the} pulse with the carrier frequency F ₀ through the antenna switch 14 is supplied to the transceiver antenna 2. At the same time, the leading edge of this pulse starts the circuit 3 for measuring time t _n . During the period of reception of bottom reverberation signals by antennas 2 and 4, at the moments of the coincidence of the phases of the signals from these antennas from the output of interferometer 5, a pulse is sent to circuit 3 to take the next time t _n , to start the time measurement circuit 6 Δt _n and to selector 7. B selector 7 at the time t _n sampled values of the signal voltage supplied from the frequency detector 8, the sample value proportional care instantaneous signal frequency from the frequency of the reference voltage generator 9. The signal sample value the instantaneous frequency f CSIRO in the memory 10 in the memory cell to each of the interference fringe and supplied to the comparison circuit 11 amplitudes. At the moment of appearance at the second input of the amplitude comparison circuit 11 of the same signal voltage value generated by the frequency detector 12, an output is sent to the circuit 6 from the output of the amplitude comparison circuit 11 to take the time Δt _n .

The measured values of t _n and the corresponding values of Δt _n are supplied to a computer 13 implemented on the basis of a microcomputer, where a predetermined series of values of D _n and n and the corresponding values of Z _n and Y _{n are} successively calculated. The values of Zn and Yn, linked to time, go to the display and documentation device 15.

According to the results of measurements by means of a display and documentation device, the relief of the seabed is built, which on the map is described by a smooth surface. The shape of this surface is substantially related to the presence of special points on the surface: points of local extrema (minima, maxima) and saddle points. The combination of such points, their location and height are important characteristics in displaying the shape of the surface of the relief, since they play the role of a discrete structure that is representative of a continuous surface.

The construction of the relief surface on a computer of the display and documentation device 15 according to the measurement data is always associated with the presence in the initial data of computational and measurement errors leading to distortions in the shape of the surface. Errors lead both to a distortion of the location of representative critical points present in the real surface of the relief, and to the appearance of unrepresentative false critical points. Therefore, the urgent task of identifying representative critical points in the calculated surface. The solution to this problem is a prerequisite for filtering noise and smoothing the calculated surface.

The solution to this problem should be based on some quantitative characteristic that determines the significance of the critical point for the particular surface under consideration. Let a certain acceptable level of significance be given, and the critical points are ordered in descending order of their significance values. Then those critical points that are more significant than the acceptable level will be representative. The algorithm for quantifying the significance of critical points is based on the following definitions.

Since the values of the absolute height (depth) of critical points do not allow us to judge which of them are more representative and which are not, with the exception of only the points with the largest minimum and maximum values. To determine the significance of the critical point of the surface as a suitable basis, we will use the concept of “topographic prominence” (Christopherson GL Using ARC / GRID to Calculate Topographic Prominence in an Archaeological Landscape. // Arc / INFO User Conference, widely used in foreign cartography, // Arc / INFO User Conference, 2003. - 15 pp. Podobnikar T. Method for Determination of the Mountain Peaks // 12th AGILE International Conference on Geographic Information Science, Leibniz Universitat Hannover, Germany, 2009, p.1-8).

Topographic significance is the difference in elevation between a peak and the highest saddle point that separates this peak from any higher peak. However, the direct use of this concept for our purposes is impossible. The fact is that, firstly, it considers only the points of local maxima and saddle points, but does not include the points of local minima of the relief, and, secondly, it does not rely on mathematical concepts, which does not guarantee the absence of logical and algorithmic mistakes. The latter is manifested, for example, in that the determination of topographic significance does not allow an unambiguous choice of a saddle point, relative to which the height of the vertex is measured.

At the same time, in Russian geography there is a concept similar in meaning - "relative height", however, it does not coincide with the concept of "topographic significance". Relative height is the topographic excess of a point on the earth’s surface relative to another point, measured vertically, equal to the difference in the absolute heights of these points (for example, the height of a mountain peak above the bottom level of a nearby valley); the vertical distance from the specified initial level to the level, point or object taken as a point. Therefore, the concept of relative height has a different meaning depending on the context, which does not allow them to use. This problem can be solved if we generalize the concept of topographic significance by including local relief minima in it by constructing a unique method for identifying for each saddle point the corresponding minimum or maximum point. The main problem along this path is the definition of an algorithm for identifying the correspondence between saddle points and extreme points. An effective method for assessing significance can be obtained by using tools for describing smooth functions, which include a cartographic representation of a relief surface (Zhukov Yu.N. Mathematical tools for describing a cartographic representation of a relief of the Earth // Navigation and Oceanography. 2011, No. 32, p. 60-69) .

The cartographic representation of the relief is an analogue of a mathematical object — the non-degenerate Morse function. Such a function has only simple critical points:

saddles, minima and maxima, its topological properties are described, for example, in (Milnor J. Theory of Morse. - M.: Publishing House of LCI, 2011. - 184 p.). For functions of this type, methods have been developed for identifying and organizing critical points of the surface using topological characteristics. In computational topology, this method is called topological persistence (Bauer U. Persistence in discrete Morse theory. Dissertation zur Eriangung des mathematisch-naturwissenschaftlichen Doktorgrades Doctor rerum naturalium der Georg-August-Universitat GOttingen, 2011 .-- 109 p.).

It is natural to use an analog of this method to solve our problem of finding a correspondence between saddle points and surface extrema.

Without loading the text with mathematical details and details, we give an informal description of the method adapted to our task, using the example of obtaining persistence intervals for a one-dimensional curve (Fig. 2). Let a smooth one-dimensional curve be given. Take a horizontal line cutting a given curve. At some level, the cross section of the curve is a set of disconnected horizontal intervals (segments). Roughly speaking, an integer number of intervals determines the topological type of section at a given level. Let us consider how the topological type of cross sections of a curve changes at each level as the secant line moves from the minimum level of the curve to the highest. As the cross-sectional level increases, the number of intervals changes only at the moment the secant passes through a point corresponding to either a maximum or a minimum. In the level range between two consecutive extrema, the number of section intervals does not change. Therefore, the dynamics of the number of intervals of the section in the process of increasing the level of the section changes in accordance with the following simple rule: if a minimum is found, then the number of intervals increases by one, if a maximum is found, then the number of intervals decreases by one. In this process, there is always a minimum that creates an interval, and there is some maximum that this interval destroys. More specifically: the current high is a pair of the last previous low seen. The successive maximum and minimum are a conjugate pair of extrema. The pairing of a pair of critical points is due to the topological properties of the curve under consideration.

This procedure allows you to uniquely obtain a set of pairs of minima-maxima determined by the topology of the curve. The magnitude of the absolute height difference of the critical points Δ = | h _min -h _max | constituting the conjugate pair determines the quantitative assessment of the stability interval and the corresponding conjugate points. The stability intervals obtained by the indicated method are the desired objects for the one-dimensional case.

The given example for the one-dimensional case gives reason to introduce the concept of “significance” of the critical point of the relief surface as a measure of the topological stability of the conjugate pair of critical points into which it enters. A quantitative criterion for the significance of ε is defined as follows. To the set of pairs of critical points P, we associate the set of differences in the heights of the critical points P = {Δ _i } (i = 1, ..., N, N is the number of pairs). The significance of a particular pair of critical points in a given set of pairs P is defined as the ratio ε _i = Δ _i / Δ _max - the maximum value among {Δ _i }. Thus, each point in a pair has the same significance value. The value of ε is always normalized and lies in the range 0–1.

The implementation of a similar procedure for a two-dimensional surface is much more complicated. Among the critical points of a two-dimensional surface, saddle points appear, which are not in the one-dimensional case. These points significantly complicate the matter, since the topological type of the horizontal section of the surface also changes at the saddle points, as well as at the points of extrema. In addition, saddle points on the vertical axis always lie between the largest maximum and minimum. When lifting the sectional plane, the saddle points can occur in any sequence with respect to each other and minor extrema.

In the two-dimensional case, a pair of critical points forming a stability interval always consists of either a saddle and a minimum, or a saddle and a maximum. To each such pair of critical points we associate the magnitude of the absolute height difference of the critical points Δ = | h _s -h _e | making up the pair. Here h _s is the height of the saddle point, h _e is the height of the point of extremum, minimum, or maximum.

To calculate the set of pairs of critical points for two-dimensional smooth surfaces, an auxiliary mathematical apparatus is usually used - the Kronrod-Reeb tree (DKR). A cartographic description of the relief can be represented by a mathematical object - the Morse function, for which the DKR is a discrete analogue that uniquely describes the position, height of the saddle points and local extremes, and most importantly, the DKR describes the relationship between them, which is determined by the topology of the surface under consideration. Each surface uniquely corresponds to a certain DKR. To date, algorithms have been developed for calculating the DKR for all types of surface representations in computers (Doraiswamy H., Natarajan V. Efficient Algorithms for Computing Reeb Graphs // Computational Geometry: Theory and Applications, Volume 42, Issue 6-7, August, 2009, p .606-616. Kunii T. L. Constructing a reeb graph automatically from cross sections. // IEEE Comput. Graph. Appl. 11.6 (1991), 44-51. Pascucci V. Loops in Reeb graphs of 2-manifolds . // Discrete and Computational Geometry 32, 2 (2004), 231-244), and therefore we will not dwell on the procedure for calculating DKR. We note an important practical circumstance: the vertices of the DKR are naturally equipped with the coordinates of the location and the height of the corresponding critical points.

An example of a surface and its corresponding DKR are presented in FIGS. 3 and 4. Table 1 presents the intervals of stability and significance values calculated from the DKR. Note that in the process of calculating the DKR, its vertices are equipped with spatial and altitudinal coordinates of the corresponding critical points, which makes it possible to display the DKR in the plane of the studied function and in the high-altitude image (Fig. 4).

The algorithm for calculating the pairs of critical points by DKR is obvious, but cumbersome enough to bring it completely. The algorithm uses the idea that the DKR points corresponding to the saddles are points of triple branching, and the points corresponding to the minima and maxima are points with one adjacent edge. We give only the basic scheme of the algorithm. The key to the algorithm is to use a dynamically changing DKR in accordance with the sequence of detected pairs of critical points. A list of critical points ordered by height is scanned from bottom to top, starting from the second from the bottom. If the point is a minimum or maximum, then the saddle closest to it in the DKR is searched. They form a saddle-extreme pair. The vertices of the extrema, the saddles, and the rib connecting them are removed from the DKR, and the bonds broken by removal of the bond are restored in accordance with the order of the vertices before removal. If the vertex is a saddle and has two underlying children, then the minimum minimum height is selected, and they form a pair of critical points. Corresponding vertices and edges are removed from the DKR with subsequent restoration of bonds. As a result of this process, the final DKR will have only two vertices corresponding to the largest minimum and maximum, and one edge connecting them. This pair of critical points forms the most significant pair with Δ _max .

Note that the above method for assessing the significance of relief forms includes the concept of topographic significance, but unlike the latter, it gives a unique algorithm for obtaining pairs of saddle-extremum points.

For the surface shown in figure 3, in accordance with its DKR calculated pairs of critical points and their significance. The results are presented in table 1.

Table 1

The table of significance of critical points for the surface depicted in figure 3.

The point number corresponds to the number of critical points in figb.

The sequence number of the appearance in the list of pairs Point type Point number Point number Point type The value of the stability interval The significance of the pair The sequence number of the significance of the pair one Saddle eighteen 12 Max. 1.28 0.08796 four 2 Saddle 8 four Max. 2.78 0,19086 3 3 Min one 3 Saddle 0,0007 0.00003 7 four Min 17 16 Saddle 0,03 0,00218 6 5 Min 6 5 Saddle 0.0001 0.000008 8 6 Saddle fourteen 13 Max. 0.0000 0.0000003 9 7 Min fifteen eleven Saddle 0.34 0,02355 5 8 Min 7 2 Saddle 3.03 0.20785 2 9 Min 10 9 Max. 14.57 one one

Let us demonstrate the application of the introduced notion of significance on the simplest example of the influence of noise in the initial data on the calculated surface shape.

For example, from table 1 it follows that for some critical points their significance is extremely small, less than a percent. This is a sign that the corresponding critical points are the result of computational noise when calculating the surface. These unrepresentative critical points can be removed from the DKR, leaving only significant ones. Figure 5 shows the DKR after removal.

We introduce a random noise uniform over the interval [-0.5 0.5] into the altitudinal coordinates of the function depicted in Fig. 3a. The results of constructing the isolines of such a surface and the DKR calculated for it are presented in Fig.6.

From the last example it follows that the proposed method is sensitive to the noise component. Therefore, it can be an effective tool in the algorithm for preliminary smoothing of the bottom topography obtained from the measured depths. In addition, the above examples show that the similarity of two surfaces of the same relief site can be established to a first approximation using significant critical points. If the significance and spatial coordinates have rather close (from the point of view of the problem being solved) values, then the corresponding surfaces are similar. Other, more subtle, comparison methods should be applied after checking this correspondence. To illustrate this statement, FIG. 6 shows two DCDs for representative critical points of the surfaces depicted in FIG. 3 and FIG. 6, and Table 2 shows the significance values for representative points of these surfaces. The latter changes slightly under the influence of noise (table 2).

table 2

The values of the interval of stability and significance for representative critical points Point number The value of the stability interval of the critical point The value of the stability interval of the critical point Criticality
point Criticality
point No noise With noise No noise With noise one 14.57 14.64 one one 2 14.57 14.64 one one 3 3.03 3.03 0.20786 0,20691 four 3.03 3.03 0.20786 0,20691 5 2.78 2.67 0,19086 0.18200 6 2.78 2.67 0,19086 0.18200 7 1.28 1.30 0.08796 0,08897 8 1,280 1.30 0.08796 0,08897

The determination of representative critical points is of practical value from the point of view of detecting noise disturbances in a surface and comparing two surfaces with each other. The Kronrod-Riba tree is the main tool and structure for representing the topological and geometric features of representative critical points of the relief surface. The problem of identifying the representativeness (significance) of critical relief points using a computer is solved by using the representation of a smooth continuous surface of the bottom relief of the Kronrod-Riba tree. Such surface parametrization makes it possible to identify significant critical points of the surface, taking into account its topological structure. This method allows you to identify noise disturbances in the surface data and to compare the topological properties of two or more surfaces obtained from sonar measurements. The combination of common with the prototype and distinctive features provides the appearance of new properties of the proposed method, namely:

- increasing the reliability of determining the ordinal numbering of a series of sloping distances corresponding to interference maxima;

- simplified the process of resolving the ambiguity of measurements;

- provides automation of determining depths and their relative coordinates in real time without visualizing interference fringes and using an auxiliary antenna base;

- determination of representative critical points of the relief reveals noise disturbances in the surface data and compares the topological properties of two or more surfaces obtained from hydroacoustic measurements. The technical implementation of the method is carried out by means of a side-scan sonar, which has industrial applicability and on well-developed mathematical software. Information sources.

1. On some features of the processing of interference echograms // Notes on hydrography. L .: GUNiO of the Ministry of Defense of the USSR, No. 203.1979, pp. 11-16.

2. Tarasyuk Yu.F. Remote measurement of the depths // Shipbuilding abroad, 1975, No. 1, pp. 104-105.

3. Naumov EA Using the phenomenon of interference of sound waves to determine the angle of inclination of the bottom // Navigation. Collection of scientific papers LVIMU them. S.O. Makarova, 1974, issue 14, p. 252-255.

Claims (1)

A method for determining depths in real time when examining the bottom topography with a side-scan sonar and then recovering it, including measuring the delay time of common-mode bottom reverb signals received by two antennas spaced apart vertically by several wavelengths of elastic vibrations, and resolving the ambiguity of measurements, calculating depths, characterized in that at each phase matching of the interfering signals, the instantaneous value of the signal frequency is recorded in the lower channel, the time is measured I delay the appearance of the signal in the upper channel with the same instantaneous frequency value, the measured value of the delay time is multiplied by the value of the operating frequency of the interferometer, determine the serial numbering of a number of measurements of the arrival delay of common-mode signals during each sounding in real time, the depths are calculated corresponding to each interference band and during the subsequent restoration of the bottom topography from the measured depths, the representativeness (significance) of critical points of the topography is assessed and by presenting a smooth, continuous bottom topography with the Kronrod-Riba tree.

Images