RU2568142C1 - Method of detecting moving ground objects from seismic signal - Google Patents

Abstract

FIELD: physics.

SUBSTANCE: disclosed is a method of detecting moving ground objects from a seismic signal, which includes, every second in an input seismic signal after pre-amplification during a sliding time window, finding the weighted average spectral frequency in the low-frequency transmission band, which corresponds to the effective bandwidth of the useful signal, and a high-frequency component generated by counting the number of positive and negative extrema of the input seismic signal over a defined period of time; calculating a tracking threshold, which is defined as the sum of the adaptive and constant components. The adaptive component is determined in two steps When the ratio of the weighted average spectral frequency in the low-frequency transmission band to the high-frequency component of the signal exceeds the tracking threshold, an object is detected, and the increase or reduction of the ratio of said characteristics determine whether the object is approaching or moving away.

EFFECT: high accuracy and information value of obtained data.

4 dwg

Description

The proposed method relates to the field of alarm systems and technical means of protection and can be used to protect sections of the terrain and approaches to objects, in reconnaissance and alarm systems and in devices for controlling the undermining of engineering ammunition.

A known method of detecting moving objects at a guarded boundary [1], which consists in registering a seismic signal in several predetermined frequency bands, determining the spectra of recorded seismic signals, analyzing the high-frequency and low-frequency components of the spectrum of the seismic signal, by the ratio between which they judge the approach or removal of a moving object.

The disadvantage is low noise immunity, since detection occurs when the signal level simply exceeds a predetermined pore value.

Of the known closest in technical essence is a method for detecting moving objects by a seismic signal [2], according to which a seismic signal of an object is recorded, a weighted average frequency of the spectrum is formed in the low-frequency band of the spectrum corresponding to the effective frequency band of the useful signal, and the high-frequency component of the signal, the ratio of these characteristics (informative feature) is compared with a threshold value.

The informative part of the spectrum of signals of a moving person and equipment (effective frequency band of a useful signal) lies in the range of 5 ... 25 Hz, i.e. is relatively low-frequency, and the spectrum of a number of interference (rain, thunder, etc.) is relatively high-frequency and lies in the range of 100 ... 150 Hz.

On the low-frequency band of the spectrum of the signal received at the output of the band-pass filter with a passband of 5 ... 25 Hz, a weighted average frequency of the spectrum is formed.

The high-frequency component of the signal is calculated from the initial signal as the ratio of the number of positive and negative extrema of the input seismic signal over a fixed time interval to the duration of this interval.

A decision to detect an object is made based on the results of comparing an informative feature with a threshold value. The threshold value is determined in advance by the seismic signal of the seismic background recorded in adverse conditions (rain, strong wind, thunderstorm, etc.) to increase the noise immunity of the device. The nature of the change in the analyzed informative feature determines the approximation (with increasing ratio) or distance (with decreasing ratio) of the object from the seismic receiver.

The disadvantage of this method is the low detection range caused by the constancy and relatively high level of the threshold, which is determined to increase the noise immunity of the seismic signal of the seismic background in adverse conditions.

It is proposed to use an adaptive (tracking) threshold, the value of which depends on the specific operating conditions of the system. The tracking threshold is made up of two terms - adaptive and constant components.

The threshold value is proposed to be formed according to the following rule:

1) TSA _n = TSA _n- 1 + Δ ASP, if PI _n > PI _n-1

or

TSA _n = TSA _n-1 , if PI _n ≤ SI _n-1 ;

2) TSA _n = PI _n , if PI _n <TSA _n ,

3) SP _n = TSA _n + SRP,

where TSA _n , TSA _n-1 are the values of the adaptive component of the tracking threshold in the n-th and (n-1) -th analysis time windows, respectively,

ΔASP - increment of the adaptive component of the threshold of the tracking threshold,

PI _n , PI _n-1 - values of the informative feature in the n-th and (n-1) -th time window of the analysis, respectively,

SP _n - the value of the tracking threshold in the n-th time window of the analysis,

PSP is a constant component of the tracking threshold.

The introduction of a tracking threshold is equivalent to the use of another informative feature - the rate of change of the ratio of the weighted average frequency of the spectrum in the effective frequency band of the useful signal and the high-frequency component of the signal. For useful signals, due to a change in the amplitude and frequency composition of the signal when an object is approaching or moving away, the rate of change of the informative feature is noticeably higher than for a seismic background signal. The constant component of the tracking threshold is introduced to ensure high noise immunity and is selected from the range of 0.03 ... 0.1 V. Smaller values are taken when a person-group of people is detected, large values - when a technique is detected.

Using the tracking threshold allows you to increase the range and probability of the correct detection of moving ground objects while maintaining high noise immunity of the device.

The proposed method for detecting moving ground objects by seismic signal is illustrated in the drawings.

In FIG. 1a, the seismic signal of a walking person moving for several tens of seconds with a beam of 50 m relative to the seismic receiver is shown.

In FIG. 1, b and c show the weighted average frequency of the spectrum in the effective passband (5 ... 25 Hz) and the high-frequency component of the signal, respectively.

In FIG. 1, d shows the ratio of the weighted average frequency of the signal in the frequency band 5 ... 25 Hz and the high-frequency component of the signal and the tracking threshold.

In FIG. 2 and FIG. Figure 3 shows similar graphs for seismic signals moving with a beam 100 m relative to the seismic receiver of the vehicle and the seismic background, respectively.

In FIG. 4 shows a diagram of the implementation of the device according to the proposed method.

A device that implements a method for detecting moving ground objects by a seismic signal contains (Fig. 4): 1 - a seismic receiver, 2 - a preliminary amplifier, 3 - a band-pass filter, 4 - a shaper of a time window, 5 - a shaper of the spectrum, 6 - a shaper of the number of extrema 7 - shaper of the average frequency of the spectrum, 8 - shaper of the high-frequency component of the signal, 9 - divider, 10 - shaper of the initial value of the tracking threshold, 11 - shaper of the tracking threshold, 12 - resolver.

The output of the seismic receiver 1 is connected to the input of the pre-amplifier 2, the output of which is connected to the input of the bandpass filter 3 and to the first input of the shaper of the number of extrema 6, the output of the bandpass filter 3 is connected to the first input of the spectrum shaper 5, the second input of which is connected to the output of the temporary shaper window 4, the output of which is also connected to the second input of the shaper of the number of extrema 6 and the second input of the shaper of the high-frequency component of the signal 8, the first input of which is connected to the output of the shaper of the extremum number 6, the output of the spectrum shaper 5 is connected to the input of the weighted average frequency shaper of spectrum 7, the output of which is connected to the first input of the divider 9, the second input of which is connected to the output of the shaper of the high-frequency component of the signal 8, and the output is connected to the second input of the shaper of the tracking threshold 11, the first input of which is connected to the output of the shaper of the initial value of the tracking threshold 10, the output of the shaper of the tracking threshold 11 is connected to the first input of the resolver 12, the second input of which is connected nen yield divider 9.

The method is as follows. To find the weighted average frequency of the spectrum in the low-frequency region of the spectrum, the seismic signal after preliminary amplification is passed through filter 3 with a passband of 5 ... 25 Hz. Further, during a sliding rectangular window, the Fourier transform is performed in the spectrum shaper 5 and the amplitude modulus of each harmonic is found. The parameters of the temporary sliding window are set in the former of the window 4. The length of the window is 3 s and its offset is 1 s. In the shaper of the weighted average frequency of the spectrum 7, the calculation of the weighted average frequency in the effective frequency band of the useful signal is performed. To determine the high-frequency component, the number of extrema in the positive and negative regions in the input seismic signal is initially calculated in the shaper of the number of extrema 6 after its preliminary amplification over a sliding time window. Next, in the shaper of the high-frequency component of the signal 8, the number of extrema is divided by the length of the window. In the divider 9 is the ratio of the weighted average frequency of the spectrum in the low-frequency bandwidth and the high-frequency component of the signal. In the shaper of the tracking threshold 11, the value of the tracking threshold is determined taking into account the ratio of the weighted average frequency of the spectrum in the low-frequency passband and the high-frequency component of the signal during the analyzed and previous time window. The initial value of the tracking threshold is set by the imager of the initial value of the tracking threshold 10. The calculated value of the tracking threshold is compared with the ratio of the weighted average frequency of the spectrum in the low-frequency passband and the high-frequency component of the signal; when the threshold is exceeded, a decision is made to detect the object. Additionally, by the nature of the change in the ratio of the indicated characteristics, the approximation (with an increase in the ratio) or the removal (with a decrease in the ratio) of the object is determined.

The proposed method for detecting moving ground objects by seismic signal allows to increase the range and probability of correct detection of seismic detection means while maintaining their high noise immunity.

Information sources

1. Patent RU 2165629, G01V 1/00. Kryukov I.N., Ivanov V.A., Dyugovanets A.P., Shualov A.G. A method for detecting moving objects at a guarded line.

2. Patent RU 2365945, G01V 1/00. Dudkin V.A., Myasnikova N.V., Mitrokhin M.A., Firsov M.S., Volskov A.A. A method for detecting moving objects by a seismic signal.

Claims (1)

A method for detecting moving ground objects by a seismic signal, including recording the seismic signal of the object, generating a weighted average frequency of the spectrum in the low-frequency bandwidth corresponding to the effective frequency band of the useful signal, and the high-frequency component of the signal by dividing the number of positive and negative extrema of the input seismic signal generated during the moving time time window, for the duration of this window, calculating the ratio of the weighted average often s of the spectrum in the low-frequency bandwidth to the high-frequency component of the signal and comparing the ratio of these characteristics with the detection threshold, determining by analyzing the change in time of a certain ratio of approaching and removing the detected object, characterized by using a tracking threshold, defined as the sum of the adaptive and constant components, while the adaptive component is determined in two stages: at the first stage, the value of the adaptive component increases by a given constant value pr increasing the ratio of the weighted average frequency of the spectrum in the low-frequency passband to the high-frequency component of the signal or does not change otherwise, at the second stage the adaptive component is taken equal to the ratio of the weighted average frequency of the spectrum in the low-frequency passband to the high-frequency component of the signal, provided that the value of this ratio is greater than the adaptive component, calculated in the first stage, or not changed otherwise.

Images