On the possibility of using an adaptive filter in an interference-type security system when detecting a moving underwater object.

On the possibility of using an adaptive filter in an interference-type security system when detecting a moving underwater object.

KOLMOGOROV Vladimir Stepanovich,
Candidate of Technical Sciences, Associate Professor

SENCHENKO Aleksey Georgievich,
candidate of technical sciences

On the possibility of using an adaptive filter in an interference-type security system when detecting a moving underwater object

Source: magazine «Special Equipment» No. 1 2008

As shown in [1], one of the problems that currently requires a solution is the problem of protecting underwater objects. Such objects include oil production facilities, hydronautics, aquaculture facilities, etc. The protection of objects adjacent to water areas also requires illumination of the underwater environment. There are devices that use the principle of detecting a moving object by disrupting the established interference pattern created by emitting acoustic or electromagnetic waves in closed spaces, such as a car interior, a guarded construction site, etc.

Such protected objects have stationary, immobile boundaries [2]. The use of this principle in marine conditions is difficult, since in the aquatic environment there are only two boundaries — the surface and the seabed, while the sea surface has a very variable shape over time and is statistically heterogeneous. Therefore, the use of this principle in marine conditions is impossible without the use of adaptive filtering principles. In order to assess the possibility of using adaptive filters when implementing an interference detector with an increased antenna aperture due to the appearance of imaginary sound sources when locating the sea surface, experimental studies of a device for detecting a moving object [3] were carried out in a hydroacoustic pool.

Fig. 1. Calculation of the acoustic field up to a distance of 2 km (a),
its horizontal section at a depth of 50 m (b)
and a vertical section at a distance of 500 m (c)
from a non-directional sound source

Fig. 2. Scheme of the experimental setup in the hydroacoustic pool:
1 — model of a moving body;
2 — receiving transducer;
3 — emitter

These studies have shown the possibility of using an adaptive non-recursive filter of the LMS (Least Mean Square) type, based on minimizing the gradient of the instantaneous error value between the reference and input signals using the least squares method [4]. To analyze the acoustic field in the marine environment, calculations were made for various areas of the Sea of ​​Japan using a wave program for calculating the acoustic field based on pseudo-differential parabolic equations [5]. Fig. 1a shows one of the calculations of the acoustic field up to a distance of 2 km, and Fig. 1 6, c — its horizontal and vertical sections.

Analysis of the acoustic field in marine conditions shows that the interference pattern can be observed for up to tens of kilometers, which is confirmed by a number of experimental works, for example in [6], while in a number of areas of the marine environment, a periodic structure of the acoustic field is observed both in amplitude and in phase. If a moving object is placed in this established interference structure, amplitude-phase fluctuations of the signal will be observed at the receiver output due to a change in the established interference pattern, which is adequate to the amplitude-phase modulation of the signal.

Fig. 3. Results of experimental studies when pulling a model of a moving body in a hydroacoustic pool under calm water conditions

Fig. 4. The amplitude envelope of the signal when pulling a flat plate
in a hydroacoustic pool in the horizontal (a) and vertical (b) planes

To confirm this phenomenon, experiments were conducted in a hydroacoustic pool. The experimental setup for conducting studies of the vertical plane in a hydroacoustic pool is shown in Fig. 2.

During the experiments, body 1 was set in motion by an electric motor and a pulling mechanism. In this case, transducer 3 emitted a high-frequency hydroacoustic signal, which after re-reflection from the water surface was received by receiver 2, after which it was detected by the amplitude detector and recorded on the PC. The interference structure of the acoustic field is formed by coherent addition of the emitted and reflected signals from the water surface and the walls of the pool.

The change in the level of the detected signal at the output of the amplitude detector during the experimental studies conducted in the hydroacoustic pool is shown in Fig. 3.

When conducting experiments in a hydroacoustic pool, a thin plate measuring 10×10 cm was also used to eliminate the influence of hydrodynamic phenomena from a moving body. The plate moved in both horizontal and vertical planes. The results of the experiment are shown in Fig. 4.

As can be seen from Fig. 4, when the body moves, fluctuations in the signal amplitude occur over time at the receiver output. In these experiments, the model of the moving body did not cross the lines of the receiving and emitting base. That is, fluctuations in the signal amplitude envelope are caused by the model's movement in the interference field of the receiving and emitting system. Due to the presence of a large number of signal reflections from the walls of the pool, non-periodicity of the interference maxima in the signal amplitude envelope was observed on the recorder (Fig. 4).

The plate was moved vertically at a distance of 1.5 m from the receiving and emitting base of the converters, while in the first 6 seconds a background signal was recorded from the amplitude detector, then the plate began to move.

Fig. 5. Results of experimental studies
when pulling a model of a moving body under conditions of a rough surface:
a) background recording with a rough surface without pulling the model;
b) signal from the output of the amplitude detector when pulling
the model against the background of a rough surface of the pool;
c) signal from the output of the amplitude detector when pulling the model
against the background of the agitated water surface of the pool
when using an adaptive filter

As can be seen in Fig. 3, 4, a signal appears at the output of the amplitude detector, characterizing the presence of a moving object when approaching the protected area. But when creating “agitation” of the water surface of the pool, this signal was not observed — it was “noised” by the appearance of modulation from the agitated surface, as shown in Fig. 5a.

Fig. 5 shows a comparative analysis of the results of experimental studies with a moving model in the presence of a disturbed water surface of the pool without using an adaptive filter (Fig. 56) and using an adaptive filter of the LMS type (Fig. 56) from the DSP block of the Simulink extension package of the MATLAB system. As can be seen in Fig. 56, the use of an adaptive filter allows us to isolate the signal caused by a change in the interference pattern of the acoustic field by a moving object, in the presence of a disturbed surface of the water area.

Thus, the use of adaptive filtering when creating an interference-type security system will allow «tuning out» from background interference and detecting an underwater moving object in the controlled area. In conclusion, it should be noted that the obtained results can also be interpreted for interference systems using electromagnetic radiation, as discussed in a number of works, for example, in [2], in the presence of a rough water surface.

Literature

1. Zvezhinsky S.S. Capabilities of new magnetometric detection means for protecting civilian and military facilities/Special equipment, 2004, No. 6.
2. Patent of the Russian Federation for invention No. 2130646 dated 05/20/1999. Method of detecting objects in a controlled area (authors: Trefilov N.A. et al.).
3. Patent of the Russian Federation for utility model (positive decision dated 08/30/07). Device for detecting a moving object (authors: Kalashnikov I.I., Kolmogorov V.S., Senchenko A.G., Yurchenko E.N.).
4. Sergienko A.B. Digital signal processing. 2nd ed. SPb.: Piter, 2006, 752 p.
5. Avilov KV Software package RPZEMS, Scientific and technical center «Module»/Collection of programs «Software for calculating the characteristics of the sound propagation channel in the marine environment», compiled by Zinyakov Yu.N. — M.: RTU VMF, 2003.
6. Orlov EF, Sharonov GA Interference of sound waves in the ocean. Vladivostok: Dalnauka, 1998, 195 p.