GB2355527A

GB2355527A - Method for detection and classification of sound sources, especially of vehicles

Info

Publication number: GB2355527A
Application number: GB9303313A
Authority: GB
Inventors: Gunnar Becker; Alwin Guedesen; Karl-Emil Hansen; Juergen Klemp; Guenter Tummoscheit
Original assignee: Atlas Elektronik GmbH
Current assignee: Atlas Elektronik GmbH
Priority date: 1992-04-10
Filing date: 1993-02-19
Publication date: 2001-04-25
Anticipated expiration: 2013-02-19
Also published as: ITMI930694A1; FR2749403B1; DE4212072C2; GB9303313D0; GB2355527B; NL9300436A; DE4212072A1; NL194516B; FR2749403A1; NL194516C; IT1276036B1; ITMI930694A0

Abstract

In a method for the detection and classification of sound sources. especially of vehicles, using geophone 11-13 installed in the ground for the reception of sound signals coupled into the ground, a reception level 19 which is picked off at the geophone 11 output is compared 20 with at least one reference level, preferably to determine whether it exceeds the level. For optimum adaptation of the reference level to the special characteristic of the ground at the location of the geophones the seismic propagation speed is determined 14 from the transit time from 11 to 13 and the wave type is also determined 17 form the phase difference between outputs of orthogonally orientated geophones 11 and 12. An adaptation level for the measured seismic propagation speed is extracted from stored characteristics of the reception level 18 for the various wave type in standard ground, and the reference level is hence corrected 22.

Description

I 2355527

DESCRIPTION

Method for detection and classification of sound sources, especially of vehicles The invention relates to a method for detection and classification of sound sources, especially of vehicles, of the type stated in the precharacterising clause of Claim 1.

In the case of such a known method (DE 38 40 732 Al), which is used in an alarm device which responds to tracked vehicles in order to activate armourpiercing, stationary landmines, the tracked vehicle is detected in that the reception level of the geophone exceeds a preset reference level. Since the reference level is selected such that it cannot be exceeded by the reception level produced by relatively light wheeled vehicles, classification of the detected vehicle as a tracked vehicle is linked thereto. As a result of the dependency of the reception level of the geophone on the characteristics of the ground at the location of use, the reference level must be adapted, that is to say matched, to the characteristics of the ground. A measure f or the respective characteristics of the ground is the seismic propagation speed measured at the location of use. This can be determined as a cross-correlation function of the output signals of the geophone and of a second geophone arranged at a known distance therefrom. However, it has been found that, despite this reference-level adaptation by means of the seismic propagation speed of the ground I I sound waves measured at the location of use, the detection probability and the f alse alarm speed which can be achieved are still unsatisfactory, particularly when it is desired to extend the method to detection and classification of different vehicles, such as wheeled and tracked vehicles of different types.

The invention is based on the object, in the case of a method of the type stated initially, of improving the adaptation of the reference level to the properties of the geophone location of use such that the reliability and accuracy of detection and classification are considerably improved.

In the case of a method for detection and classification of sound sources of the generic type defined in the precharacterising clause of Claim 1, the object is achieved according to the invention by means of the features in the characterising part of Claim 1.

The method according to the invention makes use of the knowledge that the reception level of the geophone is influenced considerably more than has been assumed in the past by the wave type of the ground sound waves which preferably forms at the location of formation (sic] as a result of the characteristics of the ground. As is known, ground sound waves propagate as longitudinal waves (P-waves), transverse waves (S-waves) or surface waves (Rayleigh-waves). Ac cording to the method according to the invention, the wave type is now determined, and the reference level which is adapted on the basis of the measured seismic propagation speed is modified using this parameter.

In this case, according to a preferred refinement of the method according to the invention for wave types which produce a significant reception level, characteristics are measured and stored as seismic propagation models which reproduce the dependency of the geophone output level, related to a defined speed of the 3 geophone, for example 1 cm/s, f rom the seismic propagation speed. A level-adaptation value is now extracted from that propagation model which is associated with the determined wave type, using the measured seismic propagation speed at the location of use of the geophone, by means of which level-adaptation value the predetermined reference level is varied in order to match the detection and classification threshold to the characteristics of the ground at the location of use, in an optimum manner. The P-wave and the Rayleigh-wave have been found to be significant wave types, while in contrast the S-wave has relatively little significance in this context.

Further advantageous refinements of the method according to the invention result from the further claims.

In order to determine the wave type of the ground sound which is predominant at the location of propagation (sic], two geophones are aligned at the location of use, according to a preferred embodiment of the method according to the invention, such that the reception direction of the one geophone is approximately vertical and that of the second geophone is approximately at right angles thereto. The phase difference existing between the output signals of the two geophones provides information on the wave type of the received ground sound wave. If both output signals are in phase or in antiphase, then the wave is a P-wave, while if there is a phase difference of approximately 90 or 270 between the output signals, then the ground sound wave is a Rayleigh-wave.

If one wishes to detect sound sources which are moving on a predetermined route, then in order to improve the reception level it is advantageous additionally to align the second geophone such that its reception direction is parallel to the route. In addition, the location of use for both geophones should be as close as possible to the route.

Using the method according to the invention, wheeled vehicles and tracked vehicles can be reliably separated, in terms of detection and classification, if as according to an advantageous embodiment of the invention - an upper reference level and a lower reference level are predetermined, the upper reference level being set such that it is exceeded exclusively by the reception level triggered by tracked vehicles. The lower reference level is raised so far that it is not exceeded by reception levels which are triggered by other noise sources which are weaker than wheeled vehicles, for example pedestrians. Light and heavy wheeled vehicles and light and heavy tracked vehicles can be identified by the provision of intermediate reference levels.

The method according to the invention is described in more detail in the following text with reference to a block diagram which is shown in the drawing, in which:

Fig. 1 shows a block diagram of functional modules for carrying out the method for detection and classification of sound sources, Fig. 2 shows a diagram of the relative geophone output level as a function of the seismic propagation speed for P-wave and Rayleigh-waves.

In the block diagram which is shown in Fig. 1 in order to explain the method for detection and classification of vehicles, as an example of sound sources, individual functional modules are specified, for the sake of better understanding, which individual modules are allocated to the individual method steps from the reception of the ground sound waves triggered by the vehicle to display of the classification result.

In detail, there are two geophones 11 and 12 at the location of propagation [sic] indicated by 10 in Fig.

1, the direction (sic] of maximum sensitivity (called the reception direction for short in the following text) of which are aligned at right angles with respect to one another. The geophones 11 and 12 are buried in the ground close to a movement path, f or example a road, which is intended to be monitored, such that the reception direction of the geophone 11 is aligned vertically and that of the geophone 12 is aligned horizontally and parallel to the movement path (z-axis and x-axis in Fig. 1). A third geophone 13, with a vertical reception direction, is likewise buried in the ground at a predetermined distance d from the first geophone 11. The burying depth of geophones 11 to 13 is shallow, so that they are close to the ground surface. The geophones 11 and 12 can be combined in a container which is then inserted into the ground at the location of use 10, suitably aligned. The output signals of the three geophones 11, 12 and 13 are subjected to signal processing in accordance with the following signal-proces sing and evaluation method:

In the case of the output signals of the first and third geophones 11 and 13, their time offset with respect to one another is determined, and the seismic propagation speed c of the ground sound at the location of propagation (sic) 10 is determined (module 14 f or determining the speed of sound) from the known distance d between these geophones 11 and 13 and the determined time offset. The time offset can be obtained, for example, by forming the cross -correlation function of the two output signals, as is described in DE 38 40 732 Al. The speed value c then results by dividing the distance d between the geophones 11 and 13 by the determined time offset. In order to obtain an improved estimated value of the seismic propagation speed, a plurality of measurements are carried out and the measured values averaged (module 15 for forming the average of the speed 6 values) In the case of the two output signals from the first and second geophones 11 and 12, the phase difference between these output signals is determined (module 16 for phase-difference determination) and the wave type of the ground sound prevailing at the location of propagation (sic] 10 is determined from the phase difference, which wave type essentially influences the reception level at the geophones 11 and 12 (module 17 for determining the type of sound wave). If the output signals of the geophones 11 and 12 are in phase or in antiphase (phase difference approximately 0 or 180), then the wave type is concluded to be a longitudinal-wave or P-wave. If the phase difference is approximately 90 or 270, then the ground sound wave prevailing at the location of formation (sic] 10 is determined to be a surface-wave or Rayleigh-wave.

A characteristic which indicates the dependency of the geophone output level relative to a reference speed of the geophone from the seismic propagation speed is in each case stored in a memory 18 ("seismic propagation model" module) for the "P-wave" wave type or mode and for the "Rayleigh-wave" wave mode. These characteristics, determined by a large number of measurements, f or the two said wave types are represented in Fig. 2. It can clearly be seen that the relative output level of the geophone is very much larger in the case of the sound waves triggered by the sound source propagating as Rayleigh-waves than in the case of the ground sound waves of the same sound source propagating as P-waves. Using the wave type which has been determined as described above and which preferably forms at the location of use 10, the relevant characteristic is sought in the seismic propagation model in Fig. 2, and a so-called adaptation level P.d.pt is extracted from this characteristic with the aid of the seismic speed of sound c, averaged as 7 described above.

The peak value of the reception level measured by the geophone is detected (module 19 for determining the reception level P...) from the output signal of one of the geophones 11 or 12, in this case the geophone 11. The reception levels Pmo. is compared with two reference levels Prof (module 20 for classification) for detection and classification of wheeled and tracked vehicles, which reference levels Prof are defined such that the upper reference level Prof is exceeded exclusively by those reception levels P.,,,, which are triggered by tracked vehicles, and the lower reference level Prof is also exceeded by those reception levels P. which are caused by wheeled vehicles.

Standard reference levels for these vehicles are stored in a memory module 21 ("class-specific reference levels" module) for predetermined standard characteristics of the ground. In order to obtain the aforementioned reference levels Prof, which are matched to the particular characteristics of the ground of the location of use 10 in an optimum manner, each of the standard reference levels is corrected (module 22 for correction) using the adaptation level Padapt derived from the seismic propagation model (module 18 and Fig. 2). If the reception level P.o. of the geophone 11 exceeds the upper and lower reference levels Pro. thus obtained, then a tracked vehicle is identified. If the reception level Pmax exceeds only the lower reference level P,ef, then a wheeled vehicle is identified. This classification result is displayed (module 23 "display").

A refined classification can be carried out if two further reference levels Prof are also introduced which are characteristic for differentiating between light and heavy tracked vehicles or wheeled vehicles, so that a total of four staggered reference levels Prof are provided, with which the reception level P,. o f the I I geophone 11 is compared. These four reference levels P,,,,f are also the correction result of four corresponding standard reference levels, which are stored in the memory module 21 and are corrected using the adaptation level P adapt - 9

Claims

PATENT CLAIMS

1. Method for detection and classification of sound sources, especially of vehicles, using geophones installed in the ground for the reception of the sound waves which are triggered by a sound source and are coupled into the ground, in the case of which method a reception level which is measured using the geophone is compared with at least one reference level, to determine whether it exceeds said level, which reference level is adapted to the characteristics of the ground at the location of use of the geophone and," for adaptation, the seismic propagation speed at the location of use of the geophone is measured, as a measure of the characteristics of the ground, characterised in that the wave type (Pwave or Rayleigh-wave) of the ground sound wave received by the geophones (11, 12) is determined and is used as a further parameter in the adaptation of the reference level (Pr.f) -

2. Method according to Claim 1, characterised in that in each case one characteristic for the dependency of the geophone output level relative to a geophone reference speed is determined from the seismic propagation speed of the ground sound waves for significant wave types (P-wave and Rayleigh-wave) of the ground sound waves and is stored as a seismic propagation model, and in that a corresponding adaptation level (P,,d.pt) is extracted from the associated propagation model, using the measured seismic propagation speed and the determined wave type, and the reference level (P..e) is hence corrected.

3. Method according to Claim 1 or 2, characterised in that a plurality of reference levels are provided such that the upper reference level is exceeded only by those reference levels (P,a,) at the geophone which are caused by tracked vehicles, and in that the lower reference level is exceeded even by reference levels (P...) which are caused by wheeled vehicles.

4. Method according to Claim 3, characterised in that two further reference levels (Pret), located between the upper and lower reference levels, are provided in order to separate light and heavy wheeled and tracked vehicles.

5. Method according to one of Claims 1 to 4, characterised in that a first geophone (11) having a vertical reception direction and a second geophone (12) having a reception direction at right angles thereto are arranged at the location of use (10), and in that a conclusion is drawn on the wave type of the received sound waves, from the phase difference between the two geophone output signals.

6. Method according to Claim 5, characterised in that the predominant propagation type of the ground sound waves at the location of use (10) is concluded to be a first wave type (Rayleigh-wave) in the case of a phase difference of approximately 90 or 270, and is concluded to be a second wave type (P-wave) in the case of a phase difference of approximately 0 or 180.

7. Method according to Claim 5 or 6 for sound waves which move on a predetermined path, characterised in that the location of use (10) of the geophones (11,, 12) is located as close as possible to the path and the reception direction of the second geophone (12) runs parallel to the path at the location of use (10).

8. Method according to one of Claims 5 to 7, characterised in that a third geophone (13) is installed in the ground [lacuna] predetermined distance (d) from the location of use (10) of the geophone and with a vertical reception direction, and in that the seismic propagation speed (c) at the location of use (10) is determined from the time offset between the output signals of the first and third geophones (11, 13) and the known geophone separation distance (d).

VL1 Amendments to the claims have been filed as follows Method for detection and classification of sound sources, especially of vehicles, using geophones in stalled in the ground for the reception of the sound waves which are triggered by a sound source and are coupled into the ground, in which method a reception level which is measured using a geophone is compared with at least one reference level, to determina whether it exceeds said level, which reference level is adanted to the characteristics of the ground at the location of use of the geophone and, for adaptation, the seismic propagation speed at the location of use of the geophone is measured, as a measure of the characteristics of the ground, characterised in that the wave type (P-wave or Rayleigh-wave) of the ground sound wave received by two geophones (11, 12) is determined and is used as a further parameter in the adaptation of the reference level (P,.f).

2. Method according to Claim 1, characterised in that one characteristic for the dependency of the geophone output level from the seismic propagation speed of the ground sound waves for / ac-nif s ign icant wave type (P-wave and Rayleigh-wave) of the ground sound waves is stored as a seismic propagation model, and in that a corresponding adaptation level (P.d.pt) is extracted from the associated propagation model, using the measured seismic propagation speed and the determined wave type, and the reference level (P,.f) is corrected.

3. Method according to Claim 1 or 2, characterised in that two reference levels (P,,f)are provided such that the upper reference level is exceeded only by those reception levels (P.) at the geophone which are caused by tracked vehicles, and in that the lower reference level is exceeded even by reception levels (P.) which are caused by wheeled vehicles.

4., Method according to Claim 3, characterised in that two further reference levels (P....), located between the upper and lower reference levels, are provided in order to separate light and heavy wheeled and tracked vehicles.

S. Method according to one of Claims 1 to 4, characterised in that a first geophone (11) having a vertical reception direction and a second geophone (12) having a reception direction at right angles thereto are arranged at the location of use (10). and in that a conclusion is drawn on the wave type of the received sound waves, from the phase difference between the two geophone output signals.

6. Method according to Claim 5r characterised in that the predominant propagation type of the ground sound waves at the location of use (10) is concluded to be a f-,,rst wave type (Rayleigh-wave) in the case of a phase difference of approximately 90 or 270, and is concluded to be a second wave type (P-wave) in the case of a phase difference of approximately 0 or 180.

7. Method according to Claim 5 or 6 for sound waves which move on a predetermined path, characterised in that the location of use (10) of the geophones (llf 12) is located as close as possible to the path and the reception direction of the second geophone (12) runs parallel to the path at the location of use (10).

8. Method according to one of Claims 5 to 7, characterised in.: that a third geophone (13) 46s installed in the ground at a predetermined distance (d) from the location of use (10) of the qeophones and with a vertical reception direction, dnd in that the seismic propagation 14 speed (c) at the location ot use (10) is determined from the time offset between the output signals of the first and third geophones (11, 13) and the known distance (d) between the first and third geophones (11,13).

9. Method for detection and classification of sound sources, especially of vehicles, substantially as hereinbefore described with reference to the accompanying drawings.