RU2816481C1 - Method for determining depth of immersion of object noisy in sea - Google Patents

Abstract

FIELD: hydroacoustics.

SUBSTANCE: used in solving problems of processing noise signals in hydroacoustic systems, and serves to determine the depth of immersion of an object noisy in the sea. The result is achieved by proposing a method for determining the immersion depth of a noisy object in the sea, based on receiving a noise signal with a hydroacoustic antenna, measuring the angle of arrival of the signal in the vertical plane, calculating the autocorrelation function of the signal, followed by determining the time delay between paired beams, and determining the immersion depth of the noisy object based on an analysis of the delay and angle of arrival of the signal, and when implementing the method, the receiving antenna of the noise direction finder is lowered to a depth that obviously exceeds the maximum possible immersion depth of the noisy object.

EFFECT: fast determining the immersion depth of a noisy object.

1 cl, 1 dwg

Description

The invention relates to the field of hydroacoustics, can be used in solving problems of processing noise signals in hydroacoustic systems, and is intended to determine the depth of immersion of an object noisy in the sea.

Known methods [Koretskaya A.S., Melkanovich B.C. RF Patent No. 2650830 dated 04/17/2018. A device for obtaining information about an object making noise in the sea. IPC G01S 3/80, Zelenkova I.D., Afanasyev A.N., Koretskaya A.S. RF Patent No. 2740169 dated January 12, 2021. A method for determining the coordinates of a sea noise target. IPC G01S 15/00], which are designed to determine the range and depth of immersion of an object noisy in the sea. The methods are based on comparing the parameters of the received noise signal with similar parameters calculated for a set of hypotheses about the possible location of an object in grid nodes by distance and depth for current hydrological and acoustic conditions. In the first method, signal-to-interference ratios in several frequency ranges are used as compared parameters, and in the second method, time delays between the moments of arrival of two beams and the total intensities of these beams are used. The decision about the target location is made based on the grid node for which the calculated and predicted signal parameters coincide with the observed parameters according to a given similarity criterion. To obtain the calculated and predicted parameters, a predictive calculation of the beam structure of the signal is carried out.

A common disadvantage of these methods is the large computational costs required, both at the first stage for the formation of the beam structure of the signal for a set of possible positions of the noisy object on the distance-depth grid, and at the second stage for comparing the parameters of the received signal with the calculated and predicted parameters according to the same “distance-depth” grid.

Taking into account the fact that simple methods are known for determining the distance to a noisy object, for example [Volkova A.A., Konson A.D. Potential capabilities of the two-frequency distance estimation method // Hydroacoustics. - 2009. - No. 9. - P. 43-51], it would be useful to have a method for determining only depth, implemented at low computational costs.

The closest analogue to the problem solved and the procedures performed to the proposed invention is the method [Koretskaya A.S., Zelenkova I.D. RF Patent No. 2788341 dated January 17, 2023. A method for localizing an object noisy in the sea in space. IPC G01S 3/80].

In this method, the following basic operations are performed:

measure the speed of sound in water depending on depth,

form a table of the dependence of delays on the depth of immersion of a noisy object in the sea under current hydrological-acoustic conditions, for which a ray structure of the signal at the antenna input is formed for a set of possible positions of the noisy object on the “distance-depth” grid, calculated for each possible position of the noisy object from this set the calculated delay in the travel time of the pair of the most intense rays arriving at the antenna input at close angles, and combine the calculated delays obtained for a set of distances at the same depth into a single table cell by forming an average value,

detect the broadband signal of a noisy object in the sea at the output of the noise direction finder,

measure the autocorrelation function of the signal,

detect the most intense correlation maximum in it, excluding the global maximum at the origin, and find its abscissa, which is the measured time delay of the pair of rays,

determine the depth of immersion of an object noisy in the sea by selecting that cell of the table by depth that corresponds to the cell of the calculated delay that is closest to the measured delay.

The method is based on the physical dependence of the delay between the travel times of the rays on the immersion depth of the noisy object, and determines the immersion depth of the object based on the analysis of this dependence. In the method, at the first stage, a beam structure of the signal is formed for a set of possible positions of the noisy object along the distance-depth grid, and then the calculated delays obtained for a set of distances at the same depth are combined into a single table cell. This makes it possible to reduce computational costs at the second stage by enumerating only hypotheses about depth, without enumerating hypotheses about distance. However, the large computational costs of the first stage associated with the formation of the beam structure of the signal remain. It should be noted that these computational costs are indeed large, since in order to form the ray structure of the signal, it is first necessary for each ray, and there can be more than 500 of them, emerging from each point of the “distance-depth” grid, to carry out the so-called broaching, when on the basis the current coordinates of the ray position are calculated and its subsequent coordinates are saved, in other words, in a programmatic way, for each ray its trajectory is drawn.

The objective of the invention is to eliminate the computational costs associated with the formation of the beam structure of the signal.

To solve the problem, a method for determining the depth of immersion of an object noisy in the sea, in which

measure the speed of sound in water depending on the depth C(x),

form a set of hypotheses about the immersion depth of the object _Hi ,

detect the broadband signal of a noisy object in the sea at the output of the noise direction finder,

measure the autocorrelation function of the signal, find the most intense correlation maximum in it, excluding the global maximum at the origin, and find its abscissa, which is the measured delay T in the travel time of the pair of rays,

determine the depth of immersion of an object noisy in the sea,

new features have been introduced, namely:

immerse the receiving antenna of the noise direction finder to a depth _h0 , which obviously exceeds the maximum possible immersion depth of the noisy object,

measure the angle of arrival of the signal in the vertical plane θ ₀ ,

for each hypothesis about the immersion depth of the object H _i, the compliance criterion is calculated using the formula where T is the time delay of a pair of beams, C(h ₀ ) is the speed of sound at the depth of the direction finder, is the beam exit angle corresponding to each hypothesis about the depth of the object, C(H _i ) is the speed of sound at the depth of each hypothesis about the depth of the object, θ ₀ is the angle of arrival of the signal,

and the immersion depth of a noisy object H in the sea is determined by choosing the hypothesis about the depth for which the compliance criterion takes the smallest value

The technical result of the invention is a significant reduction in the computational costs required to determine the immersion depth of a noisy object.

The technical result is achieved due to the fact that in the method, instead of complex procedures for forming the beam structure of the signal at the antenna input for a large set of possible positions of the noisy object along the distance-depth grid, simple analytical dependencies are used that connect the signal parameters at the receiving point with the parameters at the emission point . Let's show it.

Let's consider the propagation of sound in the marine environment from the signal source (noisy object) to the receiver (noise direction finder antenna). In the horizontal plane, the sound signal propagates in a straight line, and in the vertical plane it undergoes refraction, that is, its path is curved. In the marine environment, this is due to the variability in the speed of sound propagation at different depths. Within the framework of ray theory [Brekhovskikh L.M., Lysanov Yu.P. Theoretical foundations of ocean acoustics. M.: Science. - 2007. - 370 pp.] this is explained through the concept of a ray tube (or briefly a beam). A beam is understood as an audio signal emerging from a source and passing at some distance from it through an arbitrarily small circuit. Within the framework of the ray theory, it is believed that sound energy “flows” through a set of ray tubes (rays) without intersecting their walls. An infinite number of rays emerge from the signal source towards the receiver in the vertical plane at angles from +90° up to -90° down relative to the horizontal plane. Each of these rays travels along its own unique path, undergoing total internal reflections and/or bouncing off the bottom or surface.

Despite the fact that an infinite number of rays emerge from the source, not all of them reach the receiver, which is located at a fixed point in depth and range. Many of the rays lose a large proportion of their intensity when reflected from the surface and bottom, while others have a trajectory that, for a fixed distance to the receiver, passes below or above the depth of its immersion. Thus, a single source and receiver, as fixed points, connect a limited number of beams.

Each beam is characterized by its own unique trajectory and time of sound propagation along the beam. The beams do not reach the antenna simultaneously, but with a certain delay equal to the difference between the times of sound propagation along the individual beams. The fastest beam reaches the antenna first, then the second, third, etc. It is not possible to determine the total time of sound propagation along the beam by analyzing the received signal. However, analysis of the autocorrelation function of the signal makes it possible to determine the time delays between the moments of arrival of each two rays.

In the work [Volkova A.A., Konson A.D., Koretskaya A.S. Spatial localization of a broadband signal source by immersion depth under conditions of an underwater sound channel // Hydroacoustics. - 2022. - No. 2. - P. 14-25] it is shown that if the receiver is lowered to a depth exceeding the immersion depth of the source, then the source and receiver are connected between a set of pairs of rays that have the following three properties:

1) The rays of the pair emerge from the source point at angles that are close in absolute value, but opposite in sign.

2) The rays of the pair approach the receiver point with angles that are close both in absolute value and sign.

3) For each pair of rays the following relation is satisfied:

where T is the travel time delay between the beams in a pair,

C(h ₀ ) - speed of sound at the depth of the receiver,

H - source depth,

A is the absolute value of the angle of exit of the rays from the source, which can be considered the same for both rays according to the first property.

Then, on the left side of the above equation we have two quantities that can be measured at the reception point (delay and speed of sound), and on the right side there are two quantities that are unknown (the depth of the source and the angle of the beam at the source). Due to the fact that the equality contains two unknown quantities, the authors of the work [Spatial localization of the source...] were unable to obtain an analytical expression for estimating the depth and propose a method for determining it.

However, knowing the dependence of the speed of sound on depth, and having measured the angle of arrival of the signal in the vertical plane, which, according to the second property for pairs of rays, is the angle of arrival of the mixture of signals of both beams in the pair, it is possible to draw up a tabular relationship between the absolute value of the angle of exit of these rays from source and the immersion depth of the source. That is, a set of hypotheses about the immersion depth of an object H _i can be associated with a set of beam exit angles A _i . The correspondence of pairs of hypotheses H _i ↔A _i means that if the observed signal source was located at a depth _Hi , then the absolute value of the exit angle of the pair of rays from this source would be equal to A _i .

To obtain pairs of hypotheses H _i ↔A _i it is necessary to use Snell’s law [Burdik BC Analysis of hydroacoustic systems. L.: Shipbuilding. 1988], according to which for any two points along the ray trajectory the following relation is preserved:

Where - beam angle and sound speed at a certain reference depth I,

θ _J and C _J are the beam angle and the speed of sound at some other depth J.

It is important that this relationship is valid for any distance between the source and the receiver, and calculations using this relationship are simple and do not require ray tracing procedures and the formation of the beam structure of the signal. It should be noted that the dependence of the speed of sound on depth can change along the signal propagation path, however, these changes are significantly less significant than daily or, especially, seasonal changes in the speed of sound [Komlyakov V.A. Shipborne means of measuring the speed of sound and modeling of acoustic fields in the ocean. SPb.: Science. - 2003]. Therefore, measuring the speed of sound at the installation site of the direction finder is considered sufficient for the operation of all similar methods, which include analogue methods and the prototype method.

Taking the reference depth I as the immersion depth of the receiver (the direction finder antenna) h ₀ , and knowing the dependence of the speed of sound on the depth C(x), from expression (2) we can easily obtain pairs of interrelated quantities: hypothesis about the immersion depth of the object H _i ↔ angle output of the beam A _i at this depth, where to obtain

each L; can be written:

where C(H _i ) is the speed of sound at the depth of the hypothesis about the depth _Hi ,

θ ₀ - measured angle of arrival of the signal,

C(h ₀ ) is the speed of sound at the depth of the direction finder antenna.

Then, by substituting interconnected pairs of hypotheses H _i ↔A _i obtained on the basis of relation (3) into the right side of expression (1) with fixed measured parameters T and C(h ₀ ) on the left side, it is necessary to achieve equality of the left and right sides of the expression ( 1). The depth of immersion of an object noisy in the sea will be determined by selecting a hypothesis about the depth from the pair for which the best match in equality will be obtained:

In the limiting case, when one of the hypotheses exactly coincides with the immersion depth of the object, equality (4) must be satisfied exactly. However, in a real situation, a set of hypotheses about the immersion depth of an object is formed with a certain step (sufficient to obtain the necessary accuracy in determining the depth), so there may not be an exact match in the equality. Let us define the criterion for matching the hypothesis and the depth of immersion of the object in the form:

Then the immersion depth of the object can be defined as that of the hypotheses H _i for which the compliance criterion takes the smallest value.

The basis of the proposed method is the use of simple analytical relations (5) and (3), which make it possible to determine the immersion depth of an object noisy in the sea without forming a beam structure of the signal at the antenna input, which can significantly reduce computational costs. To make this possible, the method includes a procedure for measuring the angle of arrival of the signal in the vertical plane, which is included in expression (3), and a procedure for immersing the receiving antenna of the noise direction finder to a depth that obviously exceeds the maximum possible immersion depth of the noisy object, to create conditions in which the relation is satisfied (5). The physical basis for determining the depth, associated with the presence of a dependence of the delay between the travel times of the rays on the immersion depth of the noisy object, corresponds to the physical basis of the prototype method.

The essence of the invention is illustrated in Fig. 1, which shows an enlarged block diagram of a device that implements the proposed method. The block diagram includes series-connected blocks: Antenna 1, block 2 for forming the directional characteristic of the FCN, block 3 for calculating the autocorrelation function of the signal (ACF), block 4 for determining the depth of the object's immersion (DLUB). The output of block 5 for measuring the vertical distribution of sound speed (VRSD) is connected to the second input of block 4.

Using equipment (Fig. 1), the method is implemented as follows.

Antenna 1 is lowered to a depth h ₀ , obviously exceeding the maximum possible immersion depth of the noisy object. In this case, conditions are created for the propagation of sound rays between the source (object) and the receiver (noise direction finder antenna) in pairs necessary for the implementation of the method. After antenna 1 is lowered to the desired depth, it receives a noise signal, which is transmitted to block 2.

In block 2 of the FCN, procedures for generating directional characteristics in the horizontal and vertical planes are carried out. From block 2 to block 3 of the ACF, a noise signal is received from the measured horizontal and vertical 0 _O directions. The design of the antenna and the procedure for generating directional characteristics are known, for example, from [Evtyutov A.P., Kolesnikov A.E., Korepin E.A. Handbook on hydroacoustics // Leningrad: Shipbuilding. - 1988. - 552 p.].

In block 3 of the ACF, the autocorrelation function of the noise signal is calculated [Rabiner L., Gould B. Theory and application of digital signal processing. - Ripol Classic, 1978] and the detection of the most intense local maximum, excluding the global maximum at the origin, which is formed due to other physical processes, and in this case, is not of interest. Detection of a local maximum can be implemented based on an algorithm for detecting anomalous outliers of a random process [Taylor D. Introduction to the theory of errors: Trans. from English - Mir, 1985]. Next, the abscissa of the local maximum is recorded, which represents the measured delay T over the travel time of the pair of rays.

In block 6 of the VRSZ, the speed of sound in water is measured depending on the depth C(x). Hydroacoustic sound speed meters in water are well-known devices; they are mass-produced and installed together with hydroacoustic equipment [Komlyakov V.A. Shipborne means of measuring the speed of sound and modeling of acoustic fields in the ocean. SPb.: Science. - 2003].

In block 4 TAB, which receives the measured angle of arrival of the signal in the vertical plane θ ₀ from block 2 and the dependence of the speed of sound on depth C(h) from block 6, a table of hypotheses about the depth of the object's immersion _Hi is formed depending on the beam exit angle A _i . To do this, using formula (2), for each hypothesis about the depth of the object’s immersion, the corresponding beam exit angle is calculated: The number of hypotheses and their values are chosen based on the required accuracy in determining the immersion depth of the object.

In block 4 DEPTH, which receives: antenna depth h ₀ from block 1, measured delay T from block 3, dependence of the speed of sound on depth C(x) from block 5, determine the immersion depth of the object. For this:

- form a set of hypotheses about the immersion depth of the object H _i

- for each hypothesis about the immersion depth of the object H _i, the compliance criterion is calculated using the formula where T is the time delay of a pair of beams, C(h ₀ ) is the speed of sound at the depth of the direction finder, is the beam exit angle corresponding to each hypothesis about the depth of the object, C(H _i ) is the speed of sound at the depth of each hypothesis about the depth of the object, θ ₀ is the angle of arrival of the signal.

- as the depth of the object, select the hypothesis for which the correspondence criterion takes the smallest value

All formulas that are used in the method collectively use no more than ten simple arithmetic and standard trigonometric operations and require minimal computational effort. The required accuracy of depth estimation can be achieved by forming a set of hypotheses about the depth of the object's immersion, and if the formation is carried out similarly to the prototype method, it will be obtained no lower.

All of the above allows us to consider the problem of the invention solved. A method is proposed that can be used to solve problems of processing noise signals in hydroacoustic systems, and is intended to determine the depth of immersion of a noisy object in the sea.

Claims (1)

A method for determining the immersion depth of an object noisy in the sea, in which the speed of sound in water is measured depending on the depth C(x), a set of hypotheses about the immersion depth of the object _Hi is formed, a broadband signal of an object noisy in the sea is detected at the output of the noise direction finder, and the autocorrelation function is measured signal, detect the most intense correlation maximum in it, excluding the global maximum at the origin, and find its abscissa, which is the measured delay T in the travel time of a pair of rays, determine the depth of immersion of an object noisy in the sea, characterized in that the receiving antenna of the direction finder is immersed in depth h ₀ , which obviously exceeds the maximum possible immersion depth of the noisy object, the angle of arrival of the signal in the vertical plane θ ₀ is measured, for each hypothesis about the immersion depth of the object _Hi, the compliance criterion is calculated using the formula where T is the time delay of a pair of beams, C(h ₀ ) is the speed of sound at the depth of the direction finder, is the beam exit angle corresponding to each hypothesis about the depth of the object, C(H _i ) is the speed of sound at the depth of each hypothesis about the depth of the object’s immersion, θ ₀ is the angle of arrival of the signal, and the immersion depth of the noisy object in the sea H is determined by choosing that hypothesis about depth for which the compliance criterion takes the smallest value

Images