DE19539068A1 - Acoustic monitoring system - Google Patents

Abstract

The present invention relates to a system for locating and tracking aircraft by detecting the acoustic signals radiated from the target at multiple locations and performing various processing of the acoustic signals to derive the desired information. A second aspect of the present invention enables the classification and identification of helicopters.

Description

The The present invention relates to a system for automatic acoustic Localization, tracking and classification of in the air flying goals.

monitoring systems can be divided into two main classes, namely in active and in passive systems.

The The most important active systems are the radar systems. They offer one Localization and tracking tracking on large Reaches and under most prevailing conditions, but how in all active systems, they are also easier to detect than passive systems and are thus more vulnerable.

• 1) Infrarotabbildungen,
• 2) optische Überwachung durch Satelliten und ferngesteuerte Fortbewegungsmittel (RPVs),
• 3) elektromagnetische Kennsignalerkennung und
• 4) akustische und seismische Überwachungssysteme.

Passive systems include:

• 1) infrared pictures,
• 2) optical monitoring by satellites and remote-controlled means of transport (RPVs),
• 3) electromagnetic identification signal recognition and
• 4) acoustic and seismic monitoring systems.

The largest associated with infrared and optical systems Disadvantages are that they have visual contact between the sensor and the destination (that is, they can not be objects capture obscured by obstacles); their respective performance is caused by adverse circumstances such. B. reduced precipitation. In addition, they are expensive.

passive Electromagnetic identification is a useful Technology for detecting and detecting a small number of aircraft. If however, a large number of low-flying targets occur together with interfering signals, the classification becomes the goals difficult. This technique also requires visual contact between the target and the recipient.

The Use of phased arrays in, for example, radar technology is well known. Typically, an array of antennas is used to send and receive a scanning beam and to pass through the Change the phase of the signal at each antenna To control the direction of the transmitted beam.

acoustic and seismic systems are useful for detection, classification and localization of targets up to a distance of 20 km, in particular low flying planes - and they are complementary to other sensor types. They provide a good all-round detection, use relatively simple sensors and processors and are effective, even if the goal is through the topography, trees, smoke or Precipitation o. The like. Is covered.

• (i) Mittel zum Empfangen akustischer Signale von einem Ziel an mehreren Stellen;
• (ii) Mittel zum sequentiellen Abtasten des empfangenen Signals an jeder Stelle;
• (iii) Mittel zum Umwandeln von Abtastsignalen in digitale Form;
• (iv) Mittel zur Verarbeitung der digitalen Daten, um eine Angabe über den Bereich, den Azimut und die Höhe eines jeden erfaßten Ziels zu liefern und
• (v) Mittel zum Übertragen der verarbeiteten Daten.

According to the invention, a system for automatic acoustic localization and target tracking of a target comprises:

• (i) means for receiving acoustic signals from a destination at multiple locations;
• (ii) means for sequentially sampling the received signal at each location;
• (iii) means for converting sample signals to digital form;
• (iv) means for processing the digital data to provide an indication of the range, the azimuth and the height of each detected target, and
• (v) means for transmitting the processed data.

According to one preferred embodiment comprises the system further means for further processing the digital data, thereby the target is classified, as to whether it is a helicopter and about such a classified helicopter identify.

According to one Another preferred embodiment, the means for receiving acoustic signals a rectangular tetrahedral arrangement of microphones.

According to one another preferred embodiment, the system has a Variety of subsystems, each subsystem one above described system and the subsystems with each other and are connected to a central transmitter.

The Initial transmission of the processing system is the one to transparent relay, an intermediate receiver / transmitter represents from where the transfer takes place.

Typically, in a detection system, a signal threshold is given and the detection depends on whether this threshold is exceeded. If this threshold is set too high, real goals will not be detected. However, if the threshold is set too low, noise signals will mimic targets that are not present (false alarm). The threshold is set with reference to Receiver-Operating Characteristic (ROC) curves (see, for example, Robert J. Ulrick "Principles of Underwater Sound", 2nd edition, 1975, McGraw-Hill ). These curves show the relationship between the probability that a true target will be detected and a false alarm will be triggered at different thresholds.

The microphone-detected acoustic signal is superimposed by local noise due to wind. Wind has a frequency characteristic tik, which falls by 6 dB per octave, ie, that the frequency power spectrum derived from this signal, which is received at each microphone, has a baseline that is graded by that amount.

This Problem can be reduced by using a compensation filter become. Such a device has a response behavior which increases with frequency, reducing the effect of local wind noise is compensated.

The Attenuation of an acoustic signal passing through the air is transported, grows with the frequency and thus become for a signal of several components, the high frequency components more damped than the low frequency components. Thus, at the time the signal becomes a receiver system reached, which is far from the source, distorted, so that the high frequency components have a lower relative Have intensity as the signal at the place of the source.

This Phenomenon can be evaluated by the obtained spectrum of an identified target is compared to the spectrum which is on the Location of the noise source is expected and by an area estimate is derived from the degree of observed distortion.

in the The following is the invention, for example, and with reference to the following figures explained in more detail. Show:

1 a flow chart illustrating the processing of signals according to a preferred embodiment of the invention,

2 a sensor used in a preferred embodiment of the invention,

3 a block diagram of the data processing hardware used in an embodiment;

4a and 4b the effect of a split window normalization of a frequency power spectrum;

5a to 5c the method used to determine the azimuth and the height of the target according to a preferred embodiment of the invention;

6 a typical data set used to estimate the range in which a target aircraft resides and having a series of received spectra from a typical helicopter as it approaches and flies over the receiving location;
and

7 a plurality of subsystems connected to each other and to a central transmitter.

According to 1 the main processing can be started (step 1) if the integral noise pressure level (OASPL) in a selected frequency range exceeds a predetermined threshold. For most helicopters, the range is selected to include the fundamental and fourth harmonics of the rotor blade frequencies of the main and tail rotor, which range is typically between 5 and 520 Hz.

To the startup will be a prefiltered signal from each acoustic channel entered (step 2) and these analog data are digitized (step 3) before a complex fast Fourier transform with these Data is executed (step 4) to a corresponding Frequency power spectrum to generate, which is an indication of the component frequencies is that the time signal with the associated intensities represent.

After this a "split window normalization" of the frequency power spectrum (Step 5), the classification depends of the target as a helicopter (step 6) from capturing the fundamental rotor blade frequency lines of the main and the tail rotor together with their harmonic series in the frequency domain.

It becomes a threshold for a line of the frequency range chosen to perform the classification and one certain probability of detection at a given false alarm rate specify. This is done with reference to receiver operator curves.

once is detected the helicopter is identified (step 7), by the fundamental frequency of the main rotor blade frequency with the of the tail rotor is compared. This ratio varies for a given helicopter type very little (a 1% Variation is typical) and is not affected by the Doppler effect. The ratio is used to a helicopter to characterize by dealing with circumstances of others certain helicopters in a lookup table is compared. Typically, the rotor blade frequencies are within a range between 10 and 20 Hz for the main rotor and between 50 up to 130 Hz for the tail rotor. Relationships that do not change over time are considered as of Considering the same goal.

The determination of the azimuth and elevation angles of a target (step 8) depends on the derivation of the relative phase angles of each microphone the complex fast Fourier transform data. The time delay between each of the microphones can be determined from the differential phase angles at each microphone, and this information, along with knowledge of the velocity of the advancing acoustic wavefronts, is used to derive only unambiguous azimuth and elevation angles. By determining the main and tail rotor frequencies with the same azimuth and elevation angles, a second method of identifying the signals from the same target is available.

Of the The area of the target that has been identified can be estimated be (step 9) by successive frequency spectra of the Target is observed when this approaches and by this with the records of this target type about known ranges is compared.

The processed data is then transmitted (step 10).

2 shows the acoustic sensor used in the present embodiment, the four microphones 1 has, which are arranged in a rectangular tetrahedral arrangement. The microphones are either externally polarized condenser microphones or electret microphones, each having an integrated preamplifier (not shown).

3 shows the hardware used which is used to perform the processing and which has a modified personal computer. Analog signals from each of the four microphones pass through low frequency response amplifiers 2 , Each amplifier has a smoothing filter (not shown) at each input along with a filter that cuts low frequencies (not shown). The amplified signals are sequentially passed through a multiplexer 3 which is controlled by an asynchronous clock (not shown) running at a frequency of 100 kHz. From there, the signals pass through a 12-bit (72 dB) analog-to-digital converter 4 and are used as intermediate signals in a kilo point 16 bit FIFO buffer 5 saved. If this FIFO contains 512 words, that is, if it is half full, a command is issued by a "Data Available Flag", whereafter the data is sent over the 8 bit data bus 7 the PC memory 6 via four single 2k buffers (one for each input). This time signal can be displayed if necessary. The data then becomes the 1kx16 FIFO 8th sent, which is actually a 1k × 16 data memory. A "Data-ready flag" is set to the 1k point data blocks from the input of the FIFO 8th to the digital signal processor 9 which performs the complex fast Fourier transform that supplies the corresponding data in the frequency domain. An example of a processor capable of performing the necessary Fourier transform is, for example, a Data Axquisition Processor No. DAP 2400e / 4 from Amplicon Liveline Ltd, Centenary Industrial Estate, Hollingdean Road, Brighton, UK ,

The frequency domain data is sent to the output of the 16-bit FIFO 10 when an output-empty-flag is set and the data reaches the PC-bus 7 when an output data available flag is busy.

The Passes output signal in the form of a frequency power spectrum a split-window normalization before a frequency-line threshold determined by the PC. The output data stream then has the Form of successive areas in the frequency domain and contains Ones and zeros. Alternative states show the presence of frequency lines or lack of presence.

the It is known to those skilled in the art that the data for identification of a helicopter obtained by using only a microphone can be.

4a shows a frequency power spectrum before (A) and after (B) the split-window normalization.

This is a procedure during which a window through the Frequency power spectrum runs and in which the middle Power level detected on both sides of this window is subtracted from all values within this window becomes.

4b Figure 5 shows that for a window of given width n, the average of all data points over a similar width is subtracted before the window and a similar width after the window from all data points within the window. The value n is chosen to be larger than a typical peak width but small enough to contain a plurality of peaks.

The Number of data points within a window of a given Width in the frequency power spectrum depends on the required Resolution and this determines the sampling interval, which required in the original time domain data collection is. A useful resolution does not require one acceptable long sampling interval (eg requires a resolution 1/4 Hz, a sampling interval of 4 s). This problem is overcome by accepting that the sampling intervals that each Correspond to data point, may overlap.

5a shows a representation of a rectangular tetrahedral arrangement of microphones on a set of Cartesian axes. The right-angled vertex of the tetrahedron corresponds to the zero point and the microphone arranged there carries the designation M ₀ . The remaining microphones are designated M _x , M _y and M _z corresponding to the Cartesian axes on which they are located. The distance between M ₀ and the other microphones is denoted by d.

Of the Elevation angle α of the target T is defined as the Angle between the line OT and the XY plane and the azimuth θ is defined as the angle between the Y-axis and the line OS, where S is the projection of the target T on the XY plane.

5b shows the XY plane containing the microphones M ₀ , M _x and M _y . The target T which is above this level is also shown (with the distances not to scale).

The acoustic signal emerging from the target T has the form of wavefronts P which are planar at the location of the sensor. The time t _x that elapses between a wavefront P that reaches the microphone M _x and until the wavefront reaches the microphone M ₀ is given by t x = (φ x - φ 0 ) / ω (1), where φ _x , etc. mean the phase angles of the signal at the microphone M _x , etc., and ω is the angular frequency of the signal.

t _x can also be expressed in terms of the actual distance traveled by the wavefront P and the rate at which the wavefront progresses (C = the speed of sound in air): t x = dsinθcosa / C (2).

Combining the two expressions for t _x given by equations (1) and (2) yields sinθ = {C (ψ x - φ 0 )} / (ωdcosα) (3).

By a similar approach, the phase difference between M _y and M _D can be used to derive the expression: cosθ = {C (φ u - φ 0 )} / (ωdcosα) (4).

Equations (3) and (4) are then combined to unambiguously determine θ: tanθ = sinθ / cosθ = (φ x - φ 0 ) / (Φ y - φ 0 ) (5).

5c is a representation of the plane containing M ₀ , M _z and the target T.

The time t _{z ,} which elapses until a plane wave front reaches PM _z and M ₀ , is given by t z = (φ z - φ 0 ) / ω (6).

The time t _z is also given by t z = (dsinα) / C (7).

A combination of equations (6) and (7) yields sinα = c (φ z -φ 0 ) / ωd (8).

There 0 ° ≤ α ≤ 90 ° leads the equation (8) for an unambiguous determination of α.

6 Figure 4 shows the change of the obtained spectrum with the distance of the target in the case of a helicopter flying over the record position for a distance greater than 5 km. By comparing the successive frequency spectra obtained from the target (with the target identified) with this data, the distance of the target can be estimated. The overall intensity can also provide a first-order estimate of the distance.

7 shows several subsystems 11 that communicate with each other and with a central receiver / transmitter 13 are coupled. coupling agent 12 may be a copper transmission line, an optical fiber cable, a radio link or a combination thereof. Each subsystem has an in 2 shown microphone assembly together with in 3 shown data processing hardware. The processed data from each subsystem may each be associated with digital codes that identify the subsystem.

Of the One skilled in the art will make other configurations of the connections of the subsystems have a presence. For example, a single could Data processing device can be centrally located and for each Sensor perform required data processing.

QUOTES INCLUDE IN THE DESCRIPTION

This list The documents listed by the applicant have been automated generated and is solely for better information recorded by the reader. The list is not part of the German Patent or utility model application. The DPMA takes over no liability for any errors or omissions.

Cited non-patent literature

• - Robert J. Ulrick "Principles of Underwater Sound", 2nd edition, 1975, McGraw-Hill [0014]

Claims (4)

System for automatic localization and tracking a goal with: (i) means for receiving acoustic signals the destination in several places; (ii) means for sequential Sampling the signals received at each point; (iii) funds for converting the sampled signals into digital form; (Iv) Means for processing the digital data to a distance, Azimuth and a height of the target recorded and (V) Means for transmitting the processed data. System according to claim 1, characterized by means for further processing of the digital data such that a Target is classified according to whether it is a helicopter or not and, in the case of a helicopter, the Helicopter type to identify. System according to claim 2, characterized in that the Means for receiving acoustic signals at least one rectangular Contains tetrahedral arrangement of microphones. System for automatic localization and tracking of targets with a variety of subsystems, each subsystem a system according to claim 1 or 2 and wherein the subsystems connected to each other and to a central transmitter.