Method of classification of noisy objects |
|
IPC classes for russian patent Method of classification of noisy objects (RU 2262121):
                        
 Adaptive seismic correlation object direction finder / 2248015
Direction finder can be used for taking azimuth relatively guarded objects at guarded areas, calculating number of objects in group target and classifying found objects. Direction finder has two seismic receivers, two processing channels with delay lines and correlators, maximal signal selector, correlator, testing module, commutator and calculator. To realize the direction finding function the method of passive diversity detection and ranging is used. The main information criterion for finding direction to object has to be the function of mutual signals correlation in two signal processing channels. Value of azimuth is judged from value of signal delay. Change in value of signal delay is equivalent to controlling directional diagram of seismic active aerial which allows classifying detected objects separately. Test influence is used for adaptation of speed of propagation of seismic wave which changes under influence of meteorological conditions. Current value of speed of propagation of seismic wave is judged from time of delay of test influence signal coming to second seismic receiver. Tuning of lines of delay is conducted correspondingly to those changes.
Adaptive seismic correlation object direction finder / 2248015
Direction finder can be used for taking azimuth relatively guarded objects at guarded areas, calculating number of objects in group target and classifying found objects. Direction finder has two seismic receivers, two processing channels with delay lines and correlators, maximal signal selector, correlator, testing module, commutator and calculator. To realize the direction finding function the method of passive diversity detection and ranging is used. The main information criterion for finding direction to object has to be the function of mutual signals correlation in two signal processing channels. Value of azimuth is judged from value of signal delay. Change in value of signal delay is equivalent to controlling directional diagram of seismic active aerial which allows classifying detected objects separately. Test influence is used for adaptation of speed of propagation of seismic wave which changes under influence of meteorological conditions. Current value of speed of propagation of seismic wave is judged from time of delay of test influence signal coming to second seismic receiver. Tuning of lines of delay is conducted correspondingly to those changes.
 Method of classification of noisy objects / 2262121
The method includes reception of the signal of noise radiation of the noisy object by the first receiving antenna and spectral analysis of the received signal of noise radiation of the noisy object, reception of the signal of noise radiation is also performed by the second receiving antenna, separated is the reciprocal spectrum of the signals of noise radiation received by the first and second receiving antennas, measured is the value of the carrier frequency of the autocorrelation function, and the decision on the class of the noisy object is taken at comparison of the measured carrier frequency of the autocorrelation function with threshold frequencies, each being determined as an average frequency of the initial noise radiation band of each standard object of a definite class.
 Method for detecting underwater objects and device for realization of said method / 2271551
Method includes determining, in the moment of temporary position of expanding spatial angles wave front, tracking belonging to acoustic beam (bearings) for each reflective element, positioned in wave packet of reflected signal (in space between frontal and back fronts of signal pulse, and limited body angle of direction characteristic of receiving antenna. Spatial receipt on basis of spatial-phase and spatial-correlative processing of reflected signal provides for detecting difference between spatial positions of reflecting objects within received signal wave front, which provides more information for object detection and, due to that, principally distinguishes the method from commonplace amplitude-temporal signals processing technology.
 Method for determining distance to sound source / 2276383
In the method, receipt of acoustic signals is performed by two linear groups of sound receivers. In first and second processing channels, electric signals are processed at frequency f, received by first and second linear groups of sound receivers, and in channel of frequency f1 - signals with frequency f1, received by first one of linear groups of sound receivers. Bearing to sound source is determined with utilization of relation of voltage amplitudes at outputs of first and second processing channels. Amplitude of signal voltage at output of first processing channel is connected, with supposition, that sound source is positioned on working axis of normalized characteristic of direction of first one of linear groups of sound receivers. Amplitude of sound pressure at input of first one of linear groups of sound receivers at frequency f is formed by dividing calculated value on proportionality coefficient, determined experimentally at frequency f. Level of sound pressure is calculated at input of first one of linear groups of sound receivers. Analogical calculations are performed for signal at frequency f1. Type of substrate surface is determined, and decrease of sound pressure level, caused by influence from obstructions, meteorological and atmospheric factors. Distance and topographic coordinates are calculated with consideration of influence of aforementioned factors.
Method for determining distance to sound source / 2276383
In the method, receipt of acoustic signals is performed by two linear groups of sound receivers. In first and second processing channels, electric signals are processed at frequency f, received by first and second linear groups of sound receivers, and in channel of frequency f1 - signals with frequency f1, received by first one of linear groups of sound receivers. Bearing to sound source is determined with utilization of relation of voltage amplitudes at outputs of first and second processing channels. Amplitude of signal voltage at output of first processing channel is connected, with supposition, that sound source is positioned on working axis of normalized characteristic of direction of first one of linear groups of sound receivers. Amplitude of sound pressure at input of first one of linear groups of sound receivers at frequency f is formed by dividing calculated value on proportionality coefficient, determined experimentally at frequency f. Level of sound pressure is calculated at input of first one of linear groups of sound receivers. Analogical calculations are performed for signal at frequency f1. Type of substrate surface is determined, and decrease of sound pressure level, caused by influence from obstructions, meteorological and atmospheric factors. Distance and topographic coordinates are calculated with consideration of influence of aforementioned factors.
 Device for determining direction towards a source of sound / 2276795
In device for determining direction to a source of sound, consisting of two photo-electric shadow devices and information processing systems, laser beams are directed at an angle of 90° to each other. In each photo-electric shadow device after focusing objective laser beam is split onto two laser beams, and these two laser beams go to two knives with mutually perpendicular edges. Edge of one of aforementioned knives in each photo-electric shadow device is parallel to plane, parallel to laser beams. Information, received from two photo-receivers, standing behind these knives, is utilized for maintaining similar sensitivity of both photo-electric shadow devices. Output signals from one of these photo-receivers and two other photo-receivers of photo-electric shadow devices are squared, amplified and added. Signal at output of adder is maintained constant due to loop of negative check connection from output of adder to inputs of amplifiers. On basis of signals at outputs of amplifiers with consideration of mutual phases of signal at outputs of photo-detectors by means of phase detectors and electronic computing machine, direction towards sound source is determined.
 Noise direction finder / 2284543
Noise direction finder comprises three vector receivers whose directional characteristics are oriented along the Cartesian co-ordinate system, amplifiers, band filters, three-channel unit for processing information, and computer.
 Method for determining position of a rifleman in an area / 2285272
In accordance to method, recording of sound signals is enabled in case of registration of impact waves from by-flying ultrasound bullet and barrel wave from expanding gases from barrel edge by sensitive elements, processing of these signals by means of processor, on basis of results of which position of sound source is determined. Method contains following innovations: sensitive elements are preliminarily fastened immovably relatively to optical axis of video recording device, synchronously with recording of sound signal by not less than 3 sensitive elements, recording of video image of possible position of sound source is performed by means of at least one video recording device, mounted with possible change of filming direction and position in space, during following processing of signals moment of arrival of barrel wave and frame from recorded video row, closest to aforementioned moment, are combined, and mark of rifleman position is placed on that frame. Recording of video image is performed in optical or infrasound or other range.
 Method for using navigational hydro-acoustic system by underwater apparatuses with determining of position by difference between distances to leading underwater device and response beacons / 2285273
Method for using navigational hydro-acoustic system by underwater devices includes determining position of leading underwater device relatively to responder beacons on basis of distances to responder beacons, determined by measuring expansion times of acoustic signal from underwater device to responder beacons and back. Position of each following underwater device is determined on basis of difference of total distances from leading underwater device to each responder beacon and from each responder beacon to following underwater device and distance from leading underwater device to following underwater device, determined by measuring onboard the following underwater device of differences between moments of receipt of acoustics signals of request of responder beacons by leading underwater device and responses of responder beacons, and distance to leading underwater device and direction towards it, known onboard the following autonomous underwater device.
 Mode of using by underwater vehicles of a hydro acoustic system with determination of the location by differences of distances to responder beacons / 2292057
Mode of using by underwater vehicles of a navigational hydro acoustic system is in simultaneous determination of the locations of all underwater vehicles of the group at inquiry by a hydro acoustic signal-command of one of the underwater vehicles of the group of (leading) responder beacons by one of the (driven) responder beacons. The location of each of underwater vehicles is determined by differences of distances to the leading responder beacon and to the drive responder beacon defined by measured intervals of time between reception of an acoustic signal of the request of the responder beacons by the leading responder beacon and acoustic signals of the response of the driven responder beacons. The location of the underwater vehicle is found as an intersection plot of hyperboloid of revolution whose number corresponds to the number of pairs of "leading-driven" responder beacons and focal points are located in installation plots of the corresponding responder beacons and the flatness passing through the center of the hydro acoustic antenna of the underwater vehicle transversely to the flatness of the true horizon.
|
FIELD: hydroacoustics, applicable for identification of objects according to their noise radiation. SUBSTANCE: the method includes reception of the signal of noise radiation of the noisy object by the first receiving antenna and spectral analysis of the received signal of noise radiation of the noisy object, reception of the signal of noise radiation is also performed by the second receiving antenna, separated is the reciprocal spectrum of the signals of noise radiation received by the first and second receiving antennas, measured is the value of the carrier frequency of the autocorrelation function, and the decision on the class of the noisy object is taken at comparison of the measured carrier frequency of the autocorrelation function with threshold frequencies, each being determined as an average frequency of the initial noise radiation band of each standard object of a definite class. EFFECT: enhanced efficiency of classification of noisy objects according to the spectral characteristics of their noise radiation. 2 dwg
The present invention relates to the field of hydro-acoustics, and is intended to recognize objects by their emissions. Known methods of classification of objects by analyzing the structure of the echo signals using spectral, correlation and kastelnik classification criteria (Anjuli, Gpoba "Pinger middle of the action. Leningrad: Sudostroenie, 1983, str). In systems using methods of detection and classification of targets by analyzing the noise sources, using the characteristics based on the characteristics of the spectral composition of the signal, the so-called "portrait". (Wesbury Analysis of hydroacoustic systems. Leningrad: Sudostroenie, 1988, str). More acoustic "portraits" is discussed in (Liu, Entandikwa "Automatic detection of sound images". L.: Energy, p.50). The closest analogue of the present invention is a method of classification described in (Wei and other Analysis information operator-hydroacoustical". Leningrad: Sudostroenie, 1989, str). The method comprises the following operations: reception of the signals from the noise sources noisy object receiving antenna; calculating evaluation of the integrated spectrum of the received signals from the noise sources; the analysis of the spectral composition; selection of discrete components; construction of scales; the adoption of the right of the noisy class of object features in the spectral content of the received signals from the noise sources. However, for modern objects is characterized by a reduction in the number of discrete components, resulting in discrete patterns spectra become uninformative, which makes classification by discrete components ineffective. Furthermore, the method of classification by discrete components of the spectrum will only work if all noisy objects of the same class have common discrete components, which for most modern noisy objects are not observed. (May, Entandikwa "Automatic detection of sound images" L.: Energy, p.50). The objective of the invention is to improve the efficiency of classification of noisy objects according to the spectral characteristics of the noise sources. To solve the problem in the way the classification of noisy objects, including the reception of signals from the noise sources noisy objects first receiving antenna, the spectral analysis of the received signals from the noise sources and the decision about the class noisy object according to the characteristics of the spectrum, introduced new features, namely: the signal from the noise sources produce a second receiving antenna, allocate mutual spectrum signals of the noise sources, the received first and second receiving antennas to produce the autocorrelation function of the reciprocal of the spectrum signals of the noise sources taken p is pout and second receiving antennas, measure the value of the carrier frequency of the autocorrelation function, and the decision of the noisy class object accept when comparing the measured carrier frequency of the autocorrelation function with a threshold frequency, each of which is defined as the average frequency of the original bands of the noise sources of the reference object of a certain class. The technical result of the proposed method is the possibility of classifying noisy object by the value of the average frequency spectrum of the noise sources. We show the possibility to achieve the technical result of the proposed method. If the signals from the noise sources of a certain object is subject to two receiving antennas, for temporary realizations of X1(t) and X2(t) signals the noise sources taken the first and second antennas, you can write
and where ωinthe upper cutoff frequency of the received signals from the noise sources; ωnthe lower cutoff frequency of the received signals from the noise sources;
what if the received mutual energy spectrum to expose once again the discrete Fourier transform, the result will be obtained autocorrelation function (secondary spectrum)
The function argument Direct measurement of this average frequency The invention is illustrated by drawings, where figure 1 presents the block diagram that implements this act is about; figure 2 provides explanations for the claimed method, which marked: fon- lower frequency limit of the bandwidth of the noise sources of the reference object of a class; fVAthe upper cutoff frequency in the bandwidth of the noise sources of the reference object of class A, ftime- off frequency of the reference object of a class; fNV- lower frequency limit of the bandwidth of the noise sources of the reference object class; fCCthe upper cutoff frequency in the bandwidth of the noise sources of the reference object class, fnopB- off frequency of the reference object class; fsism.- measured the average frequency of the frequency band signals, the noise sources of the 1st noisy object; fsism.- measured the average frequency of the frequency band signals, the noise sources of the 2nd noisy object, S(f) - dependence of the spectral density of the noise sources of the object from the frequency. A device that implements the method (figure 1), contains the first and second receiving antennas 1 and 2, the outputs of which are connected through the first block of the fast Fourier transform (FFT) 3 with the input of the second FFT block 4, and then through the block 5 frequency measurement unit 6 decision, the second input is connected to the input unit 7, the threshold frequency. With the help of the considered device proposed method is implemented as follows. The signals received from the two receiving the antennas 1 and 2, served on the FFT block 3 allocation mutual spectrum and the second FFT block 4. The output of block 4 is formed temporal autocorrelation function In(τ)defined by the bandwidth of the received signal to the noise sources. In block 5 is measured carrier frequency of the autocorrelation function, which is the average frequency of the frequency band signals of the noise sources, the received first and second receiving antennas. In the memory unit 7 has a frequency corresponding to the average frequency in the bandwidth of the noise sources of the reference objects, which are served on the second input unit of decision-making 6. At the first input unit 6 receives the measured carrier frequency autocorrelation function of the signals from the noise sources noisy object. Unit 6 compares the measured frequency with a threshold frequency and the decision on the noisy class object. The threshold frequency is recorded in the memory unit 7 in advance. The threshold frequency is determined for noisy objects of different classes, chosen as reference objects in field conditions at distances at which the dependence of the signals from the noise sources from the frequency when the spread in their environment practically no effect on the position of the middle frequency band signals of the noise sources. For the reference of the object class And (2) the threshold frequency, ftime
For the reference class objects (figure 2), the average threshold frequency fnopBequal Let the classification process were adopted signals the noise sources of the first and second noisy objects and the measured carrier frequency autocorrelation functions fCISMand fCISM. As can be seen from figure 2, fCISMlies near frequency ftimeand it allows you to classify the first noisy target as noisy object of class a, And fCISMlies near frequency fpornand it allows you to classify the second noisy target as noisy object of class C. the fact that fCISM<ftimeand fCISM<fporncan be explained by the peculiarities of propagation of the noise sources in the environment: the signals attenuate the noise sources in proportion to the distance, while high-frequency signals in the propagation attenuate faster than low frequency, this leads to a slight shift of the measured carrier frequencies of the autocorrelation functions of the first and second objects toward lower frequencies relative to the threshold frequency (Apptastic. The acoustics of the sea. Leningrad: Sudostroenie, 1966, p.78). Since all noisy objects have their own frequency ranges, with high reliability characterize the noisy class object, then the proposed method will work the ΓΌ for any noisy objects, the signals from the noise sources which lie in a specified band of frequencies. All the above can be the task of the invention is resolved. The classification method noisy object that includes the signal the noise sources noisy object first receiving antenna and the spectral analysis of the received signal the noise sources noisy object, characterized in that the reception signal of the noise sources also produce a second receiving antenna, allocate mutual spectrum signals of the noise sources, the received first and second receiving antennas to produce the autocorrelation function of the reciprocal of the spectrum signals of the noise sources, the received first and second receiving antennas, the measured value of the carrier frequency of the autocorrelation function, and the decision of the noisy class object accept when comparing the measured carrier frequency of the autocorrelation function with a threshold frequency, each of which is defined as the average frequency of the original bands of the noise sources of each from the reference objects of a certain class.
|