GB2186687A - Passive determination of target data of a vehicle - Google Patents

Abstract

Without self-betrayal, the position, speed and course of a vehicle are determined by evaluation of its radiated wave energy eg sound. Dispersion properties of the transmission medium such as a shallow water channel are utilised in the measurement of the interference field produced by the vehicle's own waves propagating in the transmission medium and measured by at least two transducers. Received signals are subjected to a frequency analysis and their intensities are stored as intensity patterns in dependence upon measurement time and frequency. The pitch or time-spacing of interference lines within the intensity pattern, which consist of adjacent intensities of equal power, and the time shift of the intensity patterns in relation to one another are linked together with the bearing which is determined by measurement of the transit time shift of the reception signals, for ascertaining the target data. The use of this measurement method is especially effective in the art of water-borne sound. <IMAGE>

Description

SPECIFICATION Method for the passive determination of target data of a vehicle The invention relates to a method for the passive determination of target data of a vehicle from a measurement location in the manner as stated in the opening statement of Claim 1.

Wherever vehicles are to be observed, monitored, pursued or combatted, measurement methods are necessary for the detection of position, vehicle speed and course, working without self-betrayal. By way of example in coastal protection passing marine craft should not be able to ascertain a monitoring of a coastal region by on-board radar or sonar installations, so that in the case of an invasion defence measures may be introduced with correct target. The determination of target data in another measurement region, for example an open sea area, serves in another military application case for the judgement of a combat situation and estimation of the effectiveness of tactical measures.

For this purpose in the art of water-borne sound by way of example the wave energy generated by the vehicle itself, namely the travelling noise, which is received at the measurement location, can be exploited for the determination of the target data. Such a method is known from German Patent Specification No. 887,926 in which the course of a marine craft is determined from three bearings. If in addition for example the vehicle speed of the marine craft is estimated on the basis of its propeller rotation rate, the range and course can also be calculated.

On the other hand if the range is known the then unknown vehicle speed is determined. In the initial phase of evaluation of sound direction-findings, a target path thus obtained is still dependent to a large extent upon the accuracy of the initial estimated values, namely range or vehicle speed. Only when after an own manoeuvre at least three further bearings have been measured can the unknown target data be calculated independently of the estimated values. All additionally ascertained bearings effect a compensation of the measurement errors and in the case of a resolution method by drawing on the plotting table also a compensation of the drawing inaccuracies in the determination of the course by the evaluator.In the case of an automatic evaluation of the bearing and calculation of the target path by regression methods in fact the calculated target path approaches the actual course evermore accurately, but the result of the calculation, taking consideration of a direction-finding affected by measuring errors, can be falsified more greatly than if the bearing affected by measurement error were to remain out of consideration.

It is also known from this Patent Specification to superimpose a bearing angle time curve on a predetermined family of curves in order to determine the ratio of vehicle speed and vehicle range. Such an evaluation is especially time-consuming and to a great extent dependent upon the judgement of the evaluator, so that inaccurate target data easily result. Furthermore the number of measured values to be taken into consideration is greatly limited by the manual evaluation.

The invention is based upon the problem of indicating a passive method for the determination of target data of a vehicle radiating self-generated wave energy, of the initially stated kind, which permits an indication of the target data from a stationary measurement location, automatically and without estimation of initial conditions as for example range or vehicle speed, within minimum time.

This problem is solved in accordance with the invention by the features indicated in the characterising part of Claim 1.

The invention is here based upon the physical laws of propagation of wave energy in a transmission medium with dispersion properties. As a rule such a transmission medium consists of individual strata with different transmission properties for the wave energy radiated by the vehicle. In one of the strata at least two transducers are installed as measuring arrangement which convert the wave energy radiated by the vehicle into electric reception signals.

If the method according to the invention is to be used in aviation for the passive measurement of the target data of aircraft or on land for the measurement of land vehicles, for example tanks, then as transducers microphones are inserted into stratifications of the atmosphere or geophones into ground strata and convert the acoustic energy radiated by reason of the noise of travel in the transmission stratum at the measurement location into electric reception signals.

The method according to the invention can likewise be used if the vehicle is radiating electromagnetic waves, for example light, which penetrate into a transmission stratum with dispersion properties, for example ice strata, and are propagated there.

When the method according to the invention is used in the art of water-borne sound for the passive determination of the target data of marine craft, two hydrophones are arranged as transducers in a water stratum as transmission stratum. In the simplest case this transmission stratum with dispersion properties is a shallow-water sound transmission channel, briefly shallow-water channel, in which the water stratum is limited by parallel air and ground strata as limit strata and the properties of the transmission medium, such as speed of propagation, are nearly constant. Likewise however the method according to the invention can be used if several stratifications with different transmission properties are to be noted in the water.

According to an article by C.L. Pekeris, "Theory of Propagation of Explosive Sound in Shallow Water", The Geological Society of America, Memoir 27, 1948 and a book by J. Tolstoy and C.S. Clay "Ocean Acoustics: Theory and Experiment in Underwater Sound", McGraw-Hill Book Company, New York, 1966, it is known that the propagation of sound of a noise source situated in shallow water at low frequencies can be described by a superimposition of own waves or modes. Such a physical model of the propagation of sound can be visualised to mean that the sound in the shallow-water channel is totally reflected on the water surface and partially reflected on the bottom, so that a zig-zag propagation of plane wave fronts over the distance is established.Above a critical limit frequency, which is equal to the speed of sound in water divided by about four times the height, own waves or what are called modes are developed.

The number of own waves is dependent upon the frequency of the radiated acoustic energy.

Each time an odd-number multiple of the critical limit frequency is exceeded a further own wave is added. The angle at which the wave front is reflected on the water surface and on the bottom increases with the ordinal number of the own waves. The wave fronts then travel a longer distance and strike more frequently upon the limit strata and then experience higher damping.

The own waves or modes constitute solutions of a partial wave equation for the shallowwater channel. Stated more precisely they are the own functions of the shallow-water channel in horizontal direction. The own waves are cylindrical waves which move away concentrically from the sound source. They have in the propagation direction a period which is the shorter the higher is the frequency of the propagating sound wave. The phase speed of the own wave is dependent upon the frequency of the radiated sound and at higher frequencies it decreasingly approaches the speed of propagation in water. The sound pressure course in the vertical direction is dependent upon the ordinal number of the own wave. At the water surface the sound pressure is equal to zero, at the bottom it always has a finite value, the number of zero positions in between is one less than the ordinal number.

By superimposition of several own waves an interference field is produced in the shallowwater channel. This interference field builds up about the sound source. Three-dimensional amplitude fluctuations are to be noted in the radial direction of the sound source. The distance between equal extreme values is called interference wave length. This interference wave length is solely dependent upon the properties of the shallow-water channel and the frequency of the radiated sound, it becomes greater towards higher frequencies.

When a marine craft is travelling sound is radiated in a wide frequency range and by reason of the developing own waves an interference field occurs in the shallow-water channel. This interference field is connected with the marine craft as sound source.

In an article by Weston et al, ''Interference of Wide-Band Sound in Shallow Water", Admiralty Research Laboratory, Teddington, Middlesex, 1971, reproduced by National Technical Information Service, a method is described with which transmission properties of a shallow-water channel are examined. A wide-band noise of a sound source is received by a hydrophone of fixed location. The sound source here moves with a rectilinear course initially towards the hydrophone and subsequently away from it. From the noise spectrograms are calculated successively per unit of time. The intensities of these spectrograms are represented as a function of the frequency by columns in grey sound script. A spectrogram is entered into each column which is allocated to the distance in each case between hydrophone and sound source.An intensity pattern results which proceeds in fan form towards the location of the hydrophone. This grey sound recording reflects the interference field which the sound waves of the radiated noise cause by reason of the propagation of own waves or modes.

In the method according to the invention for the determination of target data of a vehicle again spectrograms are produced from the time course of the reception signals of each transducer for frequency analysis and spectral outputs of the reception signals of each spectrogram are stored for example as intensity recording over the frequency. The individual intensity recordings are allocated to their measurement time point. As intensity recording a grey tone image can be produced. The stored spectrograms form a two-dimensional intensity pattern within a frequencytime co-ordinate system of which one axis is allocated to the frequency and the other axis is allocated to a time base and divided for example into time units.

From this intensity pattern in accordance with the invention a section is selected within a predeterminable frequency range, which section extends over a time interval of a predeterminable number of time units. Within the section adjacent intensities of equal power are sought which in the frequency-time co-ordinate system form continuous interference lines. In the case of a vehicle course leading over the measurement location, that is on over-running of the measurement location, these interference lines form nearly straight lines which proceed in fan form through the section. The origin of the fan is to be allocated to the measurement location. In the case of a passing course in which the course of the vehicle is at a transverse distance from the measurement location, a structure of hyperbola type may be recognised. The apex points of the hyperbolae characterise the greatest approach to the measurement location. If the vehicle is stationary, the transducers receive for each frequency a specific level and a strip pattern of interference lines along the individual frequency tracks occurs in the section of the intensity pattern. The pitch of the interference lines is infinitely great. (The pitch is here measured in relation to the frequency axis.) If the vehicle is moving the levels received per frequency vary over the time. The interference lines in the intensity pattern are bent and their pitch assumes finite values. The pitch of the interference lines is dependent upon the speed of approach of the vehicle to the measurement location or the radial speed component of the vehicle speed in relation to the measurement location.If the vehicle is approaching the measurement location at a great speed of approach, the pitches of the interference lines are less than if the vehicle were to travel to the measurement location from the same distance at a lower speed of approach. The associated tangential speed component of the vehicle speed contributes nothing to the formation of the intensity patterns. If a vehicle is travelling in a circle with constant vehicle speed around a transducer, a pattern of the stored intensities occurs which possesses no intensity differences along the frequency tracks. In place of the intensity pattern of fan form a pattern of parallel strips occurs which extend along the frequency tracks, as if the vehicle were to be stationary.

Only an additional radial speed component leads to the structuring of the intensity pattern being of fan form. This can also be imagined to mean that the interference field is characterised by concentric circles about the vehicle which characterise the minima and maxima of the interference waves at the distance of the interference wave lengths. In the case of a circular travel the transducer detects in each case one and the same intensity of the interference field. Only by a radial speed component are the minimum and maximum of the intensities ascertainable in alternation on the transducer.

It can be said that the interference field is coupled with the vehicle and is drawn with the speed of approach or the radial speed component of the vehicle speed over each transducer. If the vehicle is travelling on a course along the prolongation of the line of connection between the two transducers, every momentary value of the interference field will be received first by the one transducer and a little later by the other transducer. The time shift between the scanned interference fields is directly dependent upon the speed of approach, it is in inverse proportion thereto, namely the greater as the speed of approach is less or the radial speed component of the vehicle speed is less. This time shift is ascertained for the method according to the invention with the aid of the intensity patterns.The intensity patterns of the two sections are here shifted in relation to one another in the time direction until they coincide. The time shift necessary for this purpose constitutes the sought value.

Furthermore for the ascertainment of the target data in one of the sections the pitch of at least one of the interference lines, preferably the interference line passing through the middle of the section, is measured.

From these measurement data-the pitch of the interference lines and the time shift of the intensity patterns from both sections-the target data of the vehicle are calculated, taking consideration of the bearing angle and a time variation, in accordance with Claims 1 and 2.

When used in the art of water-borne sound the bearing angle can have been ascertained by any desired other sonar installation, but it is especially advantageous in accordance with Claim 3 to use the two transducers as direction-finding installation for the ascertainment of the bearing angle. A transit time difference of the reception signals on the transducers is measured, multiplied by the speed of propagation of the wave energy, divided by the interval of the transducers and the inverse sine is formed which delivers the bearing angle.

If from the measurement location only vehicles on a prescribed traffic route are to be observed, the two transducers should be arranged either in the direction of travel or parallel to the direction of travel. Direction-finding is then superfluous since the traffic route or course of the vehicle is known. The radial speed component is then calculated at any time correctly from the quotient of interval of the transducers and time shift, the range is calculated from the pitch ascertained in each case multiplied by the radial speed component.

In the case of any desired course of the vehicle in relation to the measurement location, with a determination of the bearing angle according to Claim 3, the radial speed component is determined from the quotient of transit time difference and time shift multiplied by the speed of propagation of the wave energy in the medium. The distance between vehicle and measurement location is ascertained in that the pitch is multiplied with the transit time difference and the speed of propagation of the wave energy and divided by the time shift. The tangential speed component is obtained by multiplication of the distance with the time variation of the bearing angle.

The advantages of the method according to the invention consist in that immediately after the detection of the wave energy generated and radiated by the vehicle, the target data can be ascertained continuously. It appears from the intensity pattern whether only ambient noise is received by the transducers or a vehicle has driven into the measurement range, since in the latter case then a structuring of the intensity pattern of random appearance takes place immediately and interference lines develop. As soon as interference lines can be seen it is possible to measure the pitch and time shift. The pitch of an interference line can be determined most simply by approximation of a straight line and the time shift between the interference patterns of the two sections, with the aid of the art of correlation.A further advantage consists in that during a movement action of the vehicle the determination of the target data is possible from the stationary measurement location without self-betrayal, namely without radiation of its own transmission energy or its own manoeuvring, so that the vehicle cannot perceive the monitoring by on-board measurement installations. Survey work for the installation of the measuring installation becomes superfluous if the target data of the vehicle in relation to the measurement location are of interest. The dimensions of the measuring arrangement at the measurement location are advantageously substantially smaller than the measurement range which can be monitored with the method according to the invention.Where the method according to the invention is used in the art of water-borne sound the measuring arrangement with its hydrophones is installed for example on a stationary ship or a submarine as observation station, or on several buoys or a linkage laid out on the sea bottom.

It is of quite special advantage that the accuracy of the determination of range and vehicle speed is independent of the distance between measurement location and vehicle, and the first measurement can be carried out with detectability. Moreover the determination of the target data is independent of the course of the vehicle. They can be determined in the same way in the case of an over-run, in which the course leads over the measurement location, as in the case of a passing course where the course leads at a transverse distance past the measurement location. It is also advantageous that manoeuvres of the vehicle do not influence the determination of the target data if the radial speed component within the time interval varies only inappreciably.The determination of range and vehicle speed is moreover advantageously completely independent of the movement behaviour of the vehicle in preceding time intervals and in succeeding time intervals, that is to say prior history or future travelling behaviour does not enter the measurement. Using the method according to the invention one is in a position constantly to ascertain the momentary target data of a vehicle even if the vehicle is travelling any desired courses with changing vehicle speeds. Naturally the vehicle speed can be stated only if it was nearly constant within the time interval.

According to an advantageous further development of the method according to the invention in accordance with Claims 4 and 5 a third transducer is set up at the measurement location in order to obtain unambiguous direction-finding results. The transducers are used by pairs for the ascertainment of the transit time differences. From the transit time differences, angles in relation to the central perpendicular to the interval of each transducer pair are calculated and these angles are converted into angle values in relation to a common reference direction. The direction-finding angle is determined from the transit time differences pertaining to angle values of equal magnitude. Thus what is called a mirror direction-finding is precluded.

According to an advantageous further development of the method according to the invention in accordance with Claim 6 the ascertained transit time intervals are compared with one another and the pair of transducers the reception signals of which display the greatest transit time difference is sought out. The reception signals of this transducer pair are subjected to frequency analysis for the ascertainment of the time shift. The time shift ascertained from the intensity patterns of the reception signals of this transducer pair and the transit time difference of their reception signals are combined with one another for the determination of the radial speed component and the range.From the intensity patterns of these reception signals either the pitch of the interference line in the middle of one of the two sections is obtained or the arithmetic mean value of the pitches of the interference lines extending in both sections through the middle of the section is determined.

Likewise it is possible in place of the transit time differences of the reception signals of each transducer pair to compare the time shift of the intensity patterns and to evaluate the reception signals of that transducer pair, the intensity patterns of which display the maximum time shift from one another, for the bearing angle and time shift calculation.

By the selection method according to Claim 6, intensity patterns of the reception signals of that transducer pair the connection line of which best coincides with the connection between measurement location and vehicle are evaluated. The radial speed component of the vehicle speed which causes the formation of the interference lines and the time shift of the intensity patterns in the sections also points in the same direction from the vehicle to the measurement location. The advantage of the method according to the invention in accordance with Claim 6, consists in that the reception signals of that transducer pair which guarantees the maximum accuracy for the determination of range and vehicle speed are evaluated, since the time shift to be measured between the two is at the greatest. This time shift in the case of screening of the time axis of the frequency time coordinate system in time units comprises the largest number of time units and guarantees that the relative error is smallest. Furthermore it is advantageous that determination of the target data is possible even if the vehicle is set on a course along a central perpendicular to the connecting line of one of the transducer pairs. In the case of this course the reception signals of this transducer pair in fact supply a structured intensity pattern, a comparison of the two intensity patterns for the determination of the time shift shows that the intensity patterns are of identical formation and display no time shift in relation to one another, because the two transducers simultaneously explore the same interference field.By the provision of three transducers and the evaluation of their reception signals by pairs an unambiguous determination of all target data is always guaranteed, since one of the free transducer pairs always has such an orientation that unambiguous determination of the target data is guaranteed.

According to a further development of the method according to the invention in accordance with Claim 7 the bearing angle is ascertained from the reception signals of that pair of transducers which displays the minimum transit time difference. This pair of transducers provides the maximum accuracy of measurement in comparison with the others.

With the method according to the invention it is advantageously possible to determine the course of the vehicle which is ascertained on the one hand by the bearing angle between measurement location and vehicle and on the other from a speed angle, as stated in Claim 8.

The speed angle lies between the radial speed component, the direction of which in relation to the reference direction includes the bearing angle, and the vehicle speed, pointing in the direction of the course. The course is ascertained from the sum of bearing angle and speed angle. The speed angle is either calculated from the ratio of the tangential and radial speed components of the vehicle speed or, according to an advantageous further development of the method according to the invention in accordance with Claim 9, by the inverse tangent of the product of pitch and time variation of the bearing angle taking account of a factor.The advantage of the method according to Claim 9 consists in that it is not first necessary to determine the speed components themselves, but the course can be calculated directly from the measured values, namely the pitch of an interference line in the section of the intensity pattern and the time variation of the bearing angle.

According to an advantageous further development of the method in accordance with the invention according to the features of Claim 10, for the frequency analysis only wave energy in a frequency range around a mean frequency is evaluated, which is propagated in the form of modes and causes interferences within the transmission stratum. This frequency range is ascertained in that along each frequency track a kind of modulation of the intensities is ascertained over the time and thence a measure of modulation is determined. This modulation measure would be the degree of modulation known in literature in the case of the presence of sinusoidal and non-stochastic processes. The modulation measure indicates how distinctly the own waves propagate in the transmission stratum and their interference is detectable.The frequency range lies in the lower part of the frequency spectrum of the reception signals, since on account of the damping in the transmission stratum only own waves of lower frequency are measured over great distances and on account of the small interference wave length in this frequency range the intensity pattern is finely structured.

The modulation measurement is for example determined in that the variance of the intensities on each frequency track is ascertained and the variance is related to the squared mean value of all intensities stored there and reduced by the number 1. The root of the difference then supplies the modulation amount.

The modulation amount along a frequency track is great only if the reception signal, transmitted by own waves, lies above the ambient noise level. Then intensity extremes result on the frequency tracks at the interval of half the interference wave length on the frequency track.

Due to disturbances in the propagation of the own waves however at some frequencies the modulation amount can recede greatly so that no continuous interference line or no equally structured intensity patterns of the reception signals of the two transducers can be found.

Therefore a cohesive range of adjacent frequency tracks will advantageously be selected as frequency range, for which the ascertained course, preferably smoothed over the frequency, of the modulation amount lies above a predeterminable threshold so that with the maximum possible certainty the pitch of interference lines and the time shift of the intensity patterns can be determined in the two sections.

According to an advantageous further development of the method according to the invention in accordance with the features of Claim 11 the reception signals are evaluated in a frequency interval placed higher than the frequency range as regards their transit time difference, in order to ascertain the bearing angle therefrom. Own waves in this frequency interval cannot falsify the direction-finding, since their phase speeds are by approximation equal to the propagation speed.

As is seen, the transmission properties of the transmission stratum desired for the determination of the pitch and time shift, which ensure a propagation of own waves and their interference, have a troublesome effect for the direction-finding. Due to the selection in accordance with the invention of frequency range and frequency interval an optimum adaptation of the measurement to the transmission properties has been achieved.

The further developments according to the invention in accordance with Claims 12, 13 and 14 state the calculation instructions in accordance with which the range and vehicle speed can be obtained from the measured values, and the factor listed there either is calculated according to Claim 15 from the interference wave length of two own waves which is established from the radiated wave energy at the mean frequency of the frequency range and from its frequency derivation, or is fixed in accordance with the features of Claim 16 as equal to 1.1 times the value of the mean frequency of the frequency range. This factor is typical for the propagation properties of the transmission stratum and can already be ascertained or fixed before the commencement of measuring.Numerous experiments have shown that the exact knowledge of the mechanism of the transmission stratum is not necessary at all to determine this factor, but that the approximation by 1.1 times the value of the mean frequency already supplies good measurement results.

If a structure of fan form of the intensity pattern has become recognisable, it is a sure sign that a detectable sound source has entered the measurement area. Of course an immediate measurement of the target data until the vehicle approaches the measurement location is of interest. However the pitch of a recognisable interference line at a point of the frequency-time co-ordinate system of the interference pattern can be determined only if a part of the interference line is clearly distinguished.

The earliest time moment for determining the pitch of the interference line occurs when the time interval is selected according to Claim 1 7 so that at least two intensity maxima are to be noted on the frequency trace of the mean frequency. With this dimensioning the object is achieved that in a section defined by frequency range and time interval a distinct intensity pattern is to be noted which is also sufficiently well structured for a comparison of the sections as regards their time shift. Of course it is also possible with smaller or greater time intervals to achieve measurement results. However with too small a time interval one runs the risk of obtaining no sufficiently finely structured intensity pattern in the upper range of the frequency range, because no intensity maximum and minimum is detected there any more.If the time interval is selected too great it may no longer be possible to adopt the basis that during this measurement time the vehicle is driving with nearly constant vehicle speed, so that then it is no longer possible to make an indication of the momentary level of the vehicle speed.

The interval of the transducers is selected just like the time interval in dependence upon the transmission properties of the transmission stratum and is adapted according to the features of Claim 18 to the interference field to be expected. The dimensioning stated there of an interval of the transducers in dependence upon the interference wave length of two interfering own waves guarantees that the intensity patterns in the two sections partially overlap and a correlation of the intensity patterns can be ascertained. When applies in the art of water-borne sound by way of example in a shallow-water channel with a depth of about 40 m. and a mean frequency of 300 Hz an interval of about 100 m. produces sensible measurement results.From this it can be seen that the transducers can be arranged at the measurement location closely adjacently in relation to the measurement area to be monitored, which can have an extent of more than 10 km. Experiments in water-borne sound have shown that the time interval of less than 200 s.

suffices to carry out the first measurement of a pitch of an interference line. As frequency range a band width of 200 Hz about the mean frequency of 300 Hz has proved advantageous. The first target data of a vehicle which is approaching the measurement location can thus be ascertained at the measurement location after about 3 minutes, in relation to range, vehicle speed and course, after the vehicle has been detected. Further data on the movement behaviour are continuously possible as from then throughout the entire phase of approach of the vehicle in over-running or passing the measurement location and until departure from the measurement area, namely until the vehicle is no longer detectable.

By the dimensioning of the distance of the transducers and of the time interval in dependence upon the transmission properties in the measurement area, the measurement method is adapted to the mechanism of the production of the intensity patterns, whereby an optimilisation of the measurement results is achieved.

It is especially advantageous for the determination of the target data if the intensity patterns are as finely divided as possible, since then the time shift of the intensity patterns in the two sections can be detected especially well. According to an advantageous development of the method according to the invention in accordance with Claim 19 an improvement can be achieved in that the transducers are laid out within the transmission stratum at such a distance parallel to the limit plane that the own functions in the vertical direction have no zero point and the interference field is built up from as many own waves even of higher order as possible. This distance can be determined in that a transducer within the transmission stratum occupies different positions below the limit plane in the transmission stratum and each time the interference pattern of a noise source is recorded. The optimum distance is then found when the most interference lines lie in the section. The own functions of the transmission stratum can also be calculated easily by way of approximation. From this it is likewise possible to estimate the distance for the transducer arrangement.

For ascertaining the target data the wave energy radiated from the vehicle is subjected to a frequency analysis and a noise spectrum is derived therefrom, for example in the form of a short-time power density spectrum according to Claim 20. The noise spectrum of the vehicle is preferably weighted so that it would have a constant value over the frequency when no own waves were formed in the propagation of the wave energy. Such a calculation method for the corresponding normalisation of a noise spectrum was described by way of example in a report B.L. 4556, Krupp Atlas-Elektronik, "Detection of a plurality of basic frequencies of periodic signals in coloured noise" by G. Hermstrüwer, 1976.If this method is applied for example to ship noises where the noise spectrum over the frequency has a course of hump form, the hump is smoothed and a constant value of the spectrum over the frequency is established. Only at the moment when the propagation of the wave energy by own waves takes place do minima and maxima in the spectrum form over the frequency.

According to an advantageous development of the method according to the invention according to Claim 21, the pitch of the interference line is obtained in that the interference line is approximated by a straight line and the pitch of the straight line indicates the pitch of the interference line. The approximation is achieved when the straight line no longer intersects the interference line in the section, that is when no intensity maxima or minima are ascertained any more on the straight line and thus the straight line is tangential to the interference line, or distances of the straight line from the interference line in the frequency time coordinate system are a minimum. This method can be realised especially simply with the aid of a computer by regressive calculation.

For the determination of the pitch of the interference line within the frequency time coordinate system of the intensity pattern of one of the sections according to an advantageous development of the method in accordance with the invention according to Claim 22 a straight line is arranged at will in the section and the intensities are measured along this straight line. For approximation the straight line is turned and displaced in the time or frequency direction until the measured intensities are all equal. Then the straight line is approximate to an interference line. If the straight line is to approximate to an interference line formed from intensity maxima, it must be turned and/or displaced until the intensities are all of the same magnitude and by way of example possess adjacent maximum values within the section.Thus it is guaranteed that the intensities measured along the straight line do also in fact pertain to one and the same interference line, since they are all adjacent to one another and form a continuous line. For the explanation of this method one should imagine a three-dimensional co-ordinate system with a frequency axis, a time axis and an intensity axis perpendicular to this plane. The intensities are then represented as a relief over the frequency time plane. Interference lines are height lines in this relief. A section is placed by the straight line through the height profile. if all intensities along the straight line are equal, the straight line lies on a height line and is approximate to an interference line. If all intensities along the straight line are maximum values, the straight line lies on a crest.The interference lines in the case of an over-run are approximately straight lines until the measurement location is reached, in the case of a passing course, where the course of the vehicle has a transverse distance from the measurement location, they are hyperbolae the apices of which characterise the closest approach of the vehicle to the measurement location. The interference lines in the subsequent departure or removal of the vehicle from the measurement location possess converse pitch and mirror-symmetrical course in relation to the frequency axis.

An advantageous development of the method according to the invention in accordance with Claim 23 indicates an advantageous possibility for the calculation of the approximation. The intensities are measured along a straight line arranged at will in the frequency time co-ordinate system and their mean value is formed. Moreover these individual intensities are squared, the sum of the squared intensities is formed and divided by the number of the intensities measured along the straight line in the section. The difference between this result and the squared mean value is ascertained, the square root is extracted and divided by the mean value. This calculating operation supplies the relative standard deviation of the intensities along the straight line from their mean value.The straight line approximates to the interference line the more accurately the smaller is the relative standard deviation, it is turned and displaced in the frequency-time coordinate system until the relative standard deviation is a minimum.

To increase the security of measurement according to a development of the method according to the invention in accordance with Claim 24, in the frequency-time co-ordinate system a pattern is formed from a family of straight lines which all intersect at -0.1 times the value of the mean frequency. These straight lines have equidistant intervals on the frequency track of the mean frequency. The family is displaced with its intersection point in the time direction until it optimally approximates to the interference lines in the section and no longer intersects the interference lines but is tangential thereto. Then a connection between the middle of the section and the intersection point of the straight lines is produced and the pitch of this connection is measured which supplies the pitch of the interference line for the determination of the target data.By the application of a family of straight lines an averaging of the pitch of the interference lines is brought about which supplies a statistically more certain measured value of the sought pitch of the interference line.

For the determination of the time shift of the intensity patterns in the two sections according to an advantageous development of the method in accordance with the invention according to Claim 25 the comparison of the intensity patterns of the reception signals of both transducers is carried out with correlation technique means. The special advantage consists in that by this signal processing an automation is possible in a simple manner.

As initially explained, the method according to the invention is based upon the mechanism of propagation of own waves in a transmission stratum with dispersion properties, for example a shallow-water channel, and the interference thereof. As already stated, the number of forming own waves is dependent not only upon the radiated frequency but also upon the depth of the shallow-water channel or the vertical extent of the transmission stratum to its limit planes. In the case of a gradient within the measurement area, that is if the depth is not constant, errors can occur in the determination of the time shift of the intensity patterns and of the pitch of the interference lines if the vehicle is situated at a point the depth of which differs from the depth of the measurement location.

According to an advantageous further development of the method according to the invention in accordance with Claim 26 the ascertained radial speed component of the vehicle speed is corrected by twice the amount of the relative depth variation in the measurement area. Since these are only relative values, it is not necessary to know the depth itself. Only the gradient of the bottom must be used for correction, which can easily be ascertained in the measuring of the parameters of the shallow-water channel.

The following consideration illustrates the operation:- The marine craft surrounded by the interference field travels with the vehicle speed in a time a distance which just corresponds to an interference wave length. In dependence upon the depth of the shallow-water channel however the interference wave lengths are different, namely the shallower is the shallow-water channel the shorter is the distance between two interference maxima. If the marine craft is situated in an area shallower than at the measurement location, in the same time at the measurement location the interference maximum will travel a greater distance than at the location of the ship, since no gaps can occur in the build-up of the interference field and the interference field is determined solely by the channel parameters and not by the marine craft.The measured time shift is thus smaller and the vehicle speed ascertained therefrom is too great.

The pitch of the interference line is influenced in the same way by depth variations in the measurement area. Since the ratio of pitch and time shift affects the range-finding, the range is always correctly ascertained even in the case of depth variation and does not have to be corrected. The speed angle is calculated with the aid of the pitch value corrected according to the further development according to the invention of the method in accordance with Claim 27.

The manner of operation of the method according to the invention is here described preferably for use in the art of water-borne sound. In the same way passive measurements of the target data of a vehicle are possible in the monitoring of roads on land and in the air in regions where acoustic waves of the travelling noise penetrate into ground or air strata with dispersion properties and own waves are formed.

The invention is described in greater detail below by reference to examples of embodiment represented in the drawing, wherein: Figure 1 shows a measurement situation for the method for the determination of target data from a measurement location, Figure 2 shows a block circuit diagram in which the method is realised, Figure 3 shows a sketch for the explanation of the method in the case of an over-run and special passing course in relation to the measurement location, Figure 4 shows a detail of Fig. 1, Figure 5 shows a block circuit diagram for an intensity pattern unit represented in Fig. 2, Figures 6. 1 and 6.2 show a measurement situation and associated frequency-time diagram with interference lines in the case of an over-run of the measurement location by a vehicle travelling at constant vehicle speed, Figure 7 shows a frequency-time diagram in which the vehicle is approaching the measurement location at changing vehicle speed during the over-run, Figure 8 shows a geometrical survey for the explanation of the method in the case of a course of the vehicle travelling transversely of the measurement location, Figure 9 shows a block circuit diagram of an interference line calculator represented in Fig. 2, Figure 10 shows a transmission stratum with depth variation.

Fig. 1 serves for the explanation of the method for the determination of target data of a vehicle 1 which is travelling on a course 2 past a measurement location 3 at a vehicle speed V.

The course proceeds at a course angle 0 in relation to geographical north, which is hereinafter called reference direction N. The vehicle 1 is situated in relation to the measurement location 3 at a bearing angle 0, which is entered as right-pointing bearing in relation to 0. The vehicle speed V and its two mutually perpendicular speed components, namely the radial speed compo nent V, and the tangential speed component V# are illustrated. The radial speed component Vr lies in the direction of the connection line between the vehicle 1 and the measurement location 3. At the measurement location 3 there are three transducers 4, 5 and 6 which define an equilateral triangle having the side length d.For better recognisability the size ratios in relation to the distance d and the range between measurement location 3 and vehicle 1 are represented unrealistically. The range between vehicle 1 and measurement location 3 is as a rule several orders of magnitude larger than the distance d between the transducers 4, 5, 6. The transducers 4, 5, 6 receive the travel noise radiated by the vehicle 1 and convert it into reception signals.

Transit time differences #1, #2, #3 between reception signals of each two transducers 4, 5 or 5, 6 or 4, 6 are ascertained. Angles Oj, ej (i=1, 2, 3) to the central perpendicular to the connection of the respective transducer pair are calculated from the transit time differences 11, 12, 13. These angles 0 q are equal to the inverse sine of the transit time difference r-, divided by a maximum transit time difference d Tmax c where d is the interval and c the speed of propagation in the medium.For each transit time difference r1, 12, r, two angles ssj and ej result, as entered in Fig. 1. The angle lies between the central perpendicular and a connection to the vehicle 1, according to Fig. 1, the angle #i characterises what is called the mirror bearing and simulates a target direction-finding in which the supposed target is the true target reflected on the line of connection between the transducers. The angles #14 #1 are calculated from transit time differences Ti of the reception signals on the transducers 4 and 5.Angles #2, #2 are ascertained from transit time differences T2 between the reception signals on the transducers 5 and 6 and the angles #3 and #3 from the transit time differences r, of the reception signals on the transducers 4 and 6. In order to be able to separate out of the angles z3 and c those angles which point in the direction towards mirror-image targets, the angles 9, and , are converted into angle values in relation to the reference direction N. For this purpose in each case an angle /Rj with appropriate indexing, which is entered between the central perpendicular and reference direction N, is considered.The ascertained angle values (#1-ss1) and (#1-ss1) are compared with one another. The bearing angle # in relation to the reference direction is determined from equal angle values (#1-ss1)#(#2-ss2)#(#3-ss3), #=360 -(#1-ss1). The angles #1, #1 and ss1 are entered in the mathematically positive direction, the bearing angle (t) and the course angle ;' are ordinarily stated as pointing to the right, that is in the mathematically negative direction.The following table illustrates the ascertaining of the bearing angle: Index ss fi e (Dss) (c-fl) 1 12 49 131 37 119 2 72 109 71 37 359 3 312 349 191 37 121 #=360 -37 =323 .

Fig. 2 shows a block circuit diagram for an apparatus for carrying out the method. For the ascertainment of the bearing angle 9 high-pass filters 7, 8 and 9 are connected in series after the transducers 4, 5 and 6, by way of which filters the reception signals of the transducers 4, 5 and 6 are switched through to transit time calculator stages 10, 11 and 12. In the transit time calculator stages 10, 11, 12 the transit time differences Ti, T2, r3 of the reception signals of each two transducers 4, 5 and 5, 6 and 4, 6 are ascertained.From the transit time differences #1, #2, #3 angles #1 and #1 to the central perpendicular to the connection of the corresponding transducer pair are ascertained in series-connected angle-calculation stages 13, 14 and 15. In difference stages 16, 17 and 18 angle values (#@-ss@) and (#1-ssi) for each transducer pair are ascertained. The difference stages 16, 17, 18 are connected with a reference angle indicator 19 which prepares the three angles ss1, fi2, ss3 between the reference direction N and the central perpendicular of each transducer pair.In a series-connected comparator stage 20 the angle values (#i-ss1) and (#1-ss1) thus ascertained are compared and that angle value is issued which appears as difference value three times in equal magnitude. This angle value (01-fl) is required for the calculation of the bearing angle #.

In order to guarantee the most accurate possible determination of the bearing angle # the transit time calculator stages 10, 11 and 12 are followed by a minimum detector 21 in which it is ascertained which of the three transit time differences T1, T2, r3 is the least in amount. When the vehicle 1 is situated exactly on the central perpendicular to the connection of one of the transducer pairs, the transit time difference would be equal to zero.Since the sine of the bearing angle 0 is dependent upon the transit time difference, the calculation of the bearing angle 0 is the more accurate, the less is the deviation of the bearing angle 0 from the central perpendicular, since the sine has the greatest variations of its function value in the region around its zero point. The angles 93 and 3 are ascertained from the minimum transit time difference T2 in a further angle calculator stage 22.In a series-connected difference stage 23, which is connected with the reference angle indicator 19, taking consideration of the angle 33 between reference direction N and central perpendicular, the angle values (133-fl3) and (c3-fl3) are calculated and compared with the output signal of the comparison stage 20 in a comparator 24. The angle value (3-fl3( appears at the output of the comparator 24 and is deducted from 360 in a subsequent subtraction stage 25, supplying the bearing angle (p=3600-(Q-fl3).

For the determination of the bearing angle 0 the reception signals of the transducers 4, 5, 6 are first filtered as described in high-pass filters 7, 8, 9. In order to guarantee the most precise possible determination of the bearing angle 0 the received signal must be evaluated only in a frequency interval in which the phase speeds of the own waves are nearly equal. This is the case only at higher frequencies. Here the phase speeds moreover are approximately equal to the propagation speed c of the medium. The high-pass filters 7, 8 and 9 effect the separation between desired upper frequency interval and undesired lower frequency range. The lower limit frequency of these high-pass filters 7, 8, 9 is adapted to the requirements as just described.In place of the high-pass filters 7, 8, 9, band-pass filters can also be used with advantage. The signal-to-noise ratio can be improved by the upper band limitation.

The transducers 4, 5 and 6 are each connected with an intensity pattern unit 30, 31, 32. The reception signal are subjected therein to a frequency analysis and the time course of the intensities of the reception signals ascertained per frequency are stored in a frequency-time coordinate system. An intensity pattern is produced in dependence upon the frequency and the time which, in the case of propagation of the wave energy radiated by the vehicle 1 in the form of own waves and movement of the vehicle 1, possesses a course of equal intensities of fan form or hyperbola form. In every intensity pattern unit 30, 31, 32 at the same time a section of the intensity pattern is produced over a pre-determinable frequency range and a selectable time interval. These sections with time shift have equal patterns.The time shift is caused inter alia by the radial speed component V, of the vehicle speed V.

The determination of the vehicle speed V will be explained with reference to Fig. 3, namely for the special cases of overrunning or passing of the measurement location 3, where the course points parallel to the connecting line between a transducer pair. In the case of overunning the vehicle 1 approaches the measurement location 3 along a course on a prolongation of the connecting line between the transducers 5 and 6 with constant vehicle speed V=V,, namely with the approach speed V3, which is equal to the radial speed component V,. The tangential speed component VB is equal to zero. The interference field surrounding the vehicle 1 is received firstly by the transducer 6 and after a time dependent upon the interval d and the radial speed component Va=V,=V by the transducer 5.This time is equal to a time shift TIK of the intensity patterns of the reception signals from the transducers 5 and 6. Since the interval d of the transducers 5 and 6 is known all values for the determination of the vehicle speed V have been ascertained: V results from d -.

#IK In Fig. 3 a further vehicle 1' is represented on a course parallel to the connecting line between the transducers 5 and 6. Here the interference field is "pushed past" the transducers 5 and 6 with the radial speed component V, of the vehicle speed V. A time shift TIK of the intensity patterns is ascertained as if the transducers 5, 6 were to lie at the distance a=d.sinB on the connection between measurement location 3 and vehicle 1'. This connection and the central perpendicular on the interval d between the transducers 5, 6 include an angle 0. The time shift results as a d.sin# #IK=-=--.

Vr Vr The vehicle speed V is obtained by reason of the geometrical ratios from Vr V=--.

sin# Vr is known from the time shift measurement as d.sinss Vr= TIK and the vehicle speed can be ascertained as d.sinO d V= =.

TIK. sina TIK The interval d and the time shift TIK are measured values from which thus the vehicle speed V can be ascertained without knowledge of the transit time difference T2 or of the angle 0.

Fig. 4 serves for the explanation of the vehicle speed ascertaining with the assumption that the course 2 has any desired line in relation to the measurement location 3. In Fig. 4 a detail of the measurement situation according to Fig. 1 is illustrated. The detail shows the transducers 5 and 6 and the vehicle 1 proceeding on the course 2.As already represented in Fig. 1 and described, the connecting line between the vehicle 1 and the middle of the interval d of the transducers 5, 6 includes an angle a2 of which the complement to 180 is designated by a. This angle a is likewise entered in a triangle at the measurement location 3, the base line of which forms the interval d between the transducers 5, 6 and the one cathetus of which is equal to d.sin6. This detail representation serves for the explanation of the determination of the radial speed component V, of the speed V of the vehicle 1.The measured time shift TIK of the intensity patterns is caused by the radial speed component V, of the vehicle 1 and could be measured by an imaginary measuring arrangement the connection line of which points in the direction of the radial speed component V, and has the interval d.sin. The radial speed component V, could thus be calculated from the quotient of interval of an imaginary measuring arrangement 5', 6' and time shift T. The time shift TIK is measured. The interval of the imaginary measuring arrangement 5', 6' is determined with the aid of the transit time difference 12 two be measured additionally.With the aid of this transit time difference T2 it is possible to determine the sino, namely according to the equation: T2 sin ti= d/c Thus the interval of the imaginary measurement arrangement 5', 6', which however has been stated as d.sind, results as T2 d. =T2.c.

d/c Thus the formula for the radial speed component V, reads: T2.c V,= TIK The time shift TIK of the intensity patterns of two reception signals is ascertained with the aid of a correlator circuit 33 according to Fig. 2. The two inputs of the correlator circuit 33 are connected through a controllable change-over switch 34 with two of the three intensity pattern units 30, 31 or 31, 32 or 30, 32. The change-over switch 34 is connected by its control input together with the output of a maximum detector 35 which is connected in series after the three transit time calculator stages 10, 11 and 12. In the maximum detector 35 the greatest transit time T2 is ascertained and it is determined that the transit time T2 lies between the reception signals of the transducers 5 and 6. By the change-over switch 34 the sections of the intensity patterns of the reception signals of the same transducer pair are forwarded at the output of the intensity pattern units 31 and 32 to the correlator circuit 33. The intensity patterns of the reception signals of these two transducers 5 and 6 are used for the determination of their time shift TIK, since their time shift TIK is greater than the time shifts of the intensity patterns of the reception signals of the other two transducer pairs. Thus it is guaranteed that the relative accuracy of the determination of the time shift TK is at the greatest.

In the correlator circuit 33 the intensity distribution in time is correlated along a frequency track of the one intensity pattern within the time interval At with the intensity distribution in time of the same frequency track in the second intensity pattern in a correlation stage 36, that is multiplied and integrated for each time unit. This signal processing is carried out for all frequency tracks in the frequency range Af. The correlation functions thus obtained are deposited in an intermediate store 37 contained in the correlator circuit 33. An averaged correlation function is formed as to all correlation functions in a subsequently placed mean value former 38 and the time shift TIK of the intensity patterns is determined from the position of its maximum.

The correlator circuit 33 and a further output of the maximum detector 35 for the maximum transit time difference 12 are connected with a calculator circuit 40 in which the radial speed component T2 V,=c.

IIK is calculated. In the calculator circuit 40 the quotient of transit time shift 12 and time shift TIK of the intensity patterns of the reception signals of the same transducer pair is multiplied by the propagation speed c.

In Fig. 5 there is represented a principle of assembly of the intensity pattern unit 30. The intensity pattern units 31 and 32 are realisable in exactly the same way. The transducer 4 is followed, by way of a low-pass filter 39, by an analog-digital converter with series-connected store 41. The limit frequency of the low-pass filter is dimensioned so that it lies below the limit frequency of the high-pass filters 7, 8, 9. The time course of the filtered, digitalised reception signal is stored each time in time units T. A timing pulse emitter 42 controls the analog-digital converter and the store 41 appropriately.In a subsequently arranged FFT calculator circuit 43 spectrograms are produced from the stored reception signals after necessary filtration (aliasing filter) in accordance with the algorithm of the Fast-Fourier Transformation and subsequent amount squaring and normalisation, and stored. The FFT calculator circuit 43 is followed by a store circuit 44 which is connected with a frequency control circuit 45 and time control circuit 480 for the formation of the section. In the store circuit 44 the spectrograms are stored by lines over a time base which is screened in time units T, as the intensities are deposited over the frequency f per line. The store circuit 44 is connected with the timing pulse emitter 42. An intensity pattern represented as grey tone recording occurs in association with the time t as ordinates and the frequency f as abscissae.

In the frequency control control circuit 45 a frequency range Af about a mean frequency f0 is fixed so that a modulation measure of the intensities along all frequency tracks within the frequency range Af lies above a pre-determinable threshold. The frequency control circuit 45 contains a mean value circuit 46, a difference former 47, a modulation calculator 48 and a threshold-value calculator 49. The frequency control circuit 45 is connected with the FFT calculator circuit 43. In the mean value circuit 46 the intensities It are totalled along each frequency track and divided by their number N. The mean value I of the intensities per frequency track is obtained.In the subsequent difference former 47 the variance (T2 is calculated per frequency track, in that the difference between the intensities 1, on the frequency track and the mean value I of the intensities on the same frequency track is formed, squared and totalled. In the subsequent modulation calculator 48 the amount of modulation of the intensities of each frequency track is determined. The amount of modulation M is calculated as:

The modulation amount calculator 48 is followed by the threshold value calculator 49 in which it is determined for which adjacent frequency tracks the possibly smoothed modulation amount lies above a predeterminable threshold.At the output of the threshold value calculator 49 the mean frequency f0 and the frequency range Af are stated, within which the modulation amount for each frequency track lies above the threshold, for example a frequency range Af=200 Hz about a mean frequency fro=300 Hz.

The store circuit 44 is actuated by the frequency control circuit 45 for the formation of the section. Moreover the store circuit 44 is connected together with the time control circuit 480.

In the time control circuit 480, which is actuated by the timing pulse emitter 42, a time interval At of for example 200 s. is pre-stated. The time interval At includes a plurality of time units T and is so selected that at least one interference wave length is detected and for example two intensity maxima are to be noted on the frequency track of the mean frequency fO.

The frequency-control circuit 45 and time-control circuit 480 actuate the store circuit 44 and define the section of the intensity pattern. The intensity pattern in this section is moreover evaluated through the changeover switch 34 in Fig. 2 in an interference line calculator 50. The interference line calculator 50 in Fig. 2 consists of an approximation calculator 51 which is fed by the intensity pattern unit 31 just connected with the changeover switch 34, a simulation calculator 52 and a pitch calculator 53. In the approximation calculator 51 intensities of like power adjacent within the section are sought which form interference lines. In the simulation calculator 52, which is connected together with the approximation calculator 51, in a frequencytime co-ordinate system a straight line is simulated.This straight line is compared in the approximation calculator 51 with the interference line passing through the middle of the section.

The straight line in the simulation calculator 52 is turned and shifted in the time direction until deviations of the interference line from the straight line are a minimum. The pivot point of the straight line is preferably shifted in the time direction on a frequency track of -0.1f0. These deviations can be time and frequency deviations between the co-ordinates of the interference line and those of the straight line. This straight line represents the sought regression straight line. It is however likewise possible to approximate in the approximation calculator 51 not by regression but by comparison of intensities which occur in the interference pattern along the straight line. The straight line approximates to the interference line when all intensities measured along the straight line are equally great and preferably possess maximum or minimum values.

If straight line and interference line are brought into coincidence, the approximation calculator 51 gives a release signal to the pitch calculator 53 which is connected with the simulation calculator 52. The pitch calculator 53 takes over from the simulation calculator 52 the straight line in the frequency-time co-ordinate system and determines its pitch dt df =t', which indicates the sought pitch of the interference line.

The interference line calculator 50 is followed by a range calculator 55 which determines the range r between vehicle 1 and measurement location 3 from the pitch t', the time shift TIK and the transit time difference T2. The output of the correlator circuit 33 and the second output of the maximum detector 35 are likewise connected with inputs of the range calculator 55.

The range determination will be explained in greater detail below in connection with Figs. 6.1, 6.2 and 7.

Fig. 6.1 shows a measurement situation for an over-run in which the vehicle 1 is approaching the measurement location 3 on a direct or radial course with constant vehicle speed V or approach speed Va=V,=V. At the time moment to the vehicle 1 is situated at a distance r from the measurement location 3, if it retains its course with constant vehicle speed V. At the time moment tCpA it will have reached the measurement location 3.

With the method for the determination of target data it is intended to determine the range r, the vehicle speed V, the bearing angle d and the course angle ,. In this movement case the interference lines are approximately straight lines proceeding in fan form in the frequency-time co-ordinate system. Fig. 6.2 shows in a sketch of principle the progress of such interference lines in the frequency-time co-ordinate system for the starting phase. The interference lines are in reality slightly curved, but are here indicated diagrammatically as straight lines G1, G2 ....

Gn. If one moves on the frequency track of the mean frequency f0 the distances between the straight lines G1, G2, . ., Gn are determined by the interference wave length X(fo) and in dependence upon the approach speed V3=V,=V. The interval X(fo) V, is the smaller, the greater is the approach speed Va=Vr. If one considers an intensity maximum on one of the interference lines, for example the point on the straight line G1 at the moment to at the mean frequency fO, it can be said that in a time period AT totCpA just k intensity maxima occur at the interval X(fo) Vr On the frequency track f likewise k intensity maxima can be ascertained as far as the straight line G1.Intensity maximum number k is present for the frequency f on the straight line G1 at a time t, the intervals between the intensity maxima here amount to X(f) V, The time periods until the reaching of the measurement location 3 are: X(f0) X(f) t0-tCPA=k . - and t-tCPA=k . -.

Vr Vr The equation for t0-tCPA is resolved according to k and inserted into the equation for ttCpA.

Their results for t: X(f) t = (t0 - tCPA) # -- + tCPA.

X(f0) If this equation is differentiated according to the frequency f, one obtains the pitch of the straight line G1 dt X'(f) - = t' = (t0 - tCPA) . --.

df X(f0) If this equation is resolved according to t0-tCPA one obtains for the mean frequency f0 X(f0) t0 - tCPA = t' . --.

X'(f0) This time period t0-tCPA is however just the time which elapses until the vehicle 1 has reached the measurement location 3 with the approach speed Vr. Thus the following is valid: X(f0) r = (t0 - tCPA) . Vr = t' . -. Vr (A).

X'(f0) In this particular movement case the approach speed Va=V, is equal to the vehicle speed V which points in the direction of the connecting line between vehicle 1 and measurement location 3, like the radial speed component V, otherwise in every general movement case.

According to the block circuit diagram in Fig. 2 however the radial speed component V, has already been calculated in the calculator stage 40 which is here equal to the approach speed V3=V,:- d TIK In this special case of the approximation d 2 - c and thus T2 d Vr -= - (B) TIK TIK The range r is to be calculated in the range calculator 55 from equation (A) and equation (B) as follows: X(fo) T2 X(fo) d r=t' . . c .-=t' .. (C) X'(fO) TIK X (fO) TIK The interference wave length X(fo) is determined in advance by knowledge of the transmission stratum in which the measurement location 3 is situated. Likewise the derivation of the interference wave length X(f) is calculable according to the frequency f and to be determined in advance for the mean frequency fO. Thus equation (C) comprises only measurable values.Numerous experiments have shown that the quotient X(fo) Xt(fo) independently of the depth of the transmission stratum is always equal to approximately 1. 1 f,, although the interference wave length X(f) itself is very greatly influenced by the depth, the depth signifying the extent of the transmission stratum between its limit planes.

Fig. 7 shows a frequency time diagram in which the vehicle 1 is approaching the measurement location 3 in an over-run according to Fig. 6. 1 with two different vehicle speeds. By reference to this diagram of principle the method and its functional validity are also to be described in the case of varying approach speed VJ. The interference lines were approximated by straight lines.

In the lower region of the diagram we see straight lines the pitch of which is greater than in the upper region. After a time of 600 s. from the beginning of measurement the vehicle 1 has increased its vehicle speed V, since the pitch of the straight lines has decreased. The interval between the individual interference lines in this region is only half as great as in the lower region of the diagram. Thence it can be concluded that the vehicle speed V was doubled. The first measurement is commenced for example after the time of 100 s. A frequency range of Af=200 Hz is considered which is arranged about a mean frequency of fro=300 Hz. The time interval amounts to At=200 s., the section of the interference pattern in the first measurement case is characterised by the letter Y.Such a section of the intensity pattern has been formed in each of the intensity pattern units 30, 31 or 32. We assume that with the aid of the correlator circuit 33 a time shift TIK=20 s. is measured between two intensity patterns. The transit time difference of the reception signals of the same transducer pair, for example transducers 4 and 5, is assumed to be measured at T1=0.067 s., wherein d 100 m --0.067 s.

c 1500 m/s.

on account of the over-run according to Fig. 6.1, where the interval of the transducers amounts to d=100 m, the speed of sound m c=1500 -.

s In the interference line calculator 50 the pitch t' of the interference line passing through the middle of the section Y is determined. It amounts to t'=6.36 s/Hz. The quotient is determined as: X(fo) =1.1 fo=1.1 300 Hz.

X'(f0) From these measured values the range r is calculated according to equation (C) as follows: 100 r=1.1#300#6.36# --#10500 m.

20 The approach speed Vr or vehicle speed V is calculated according to equation (B): T1 1500 m/s.0.0675 Vr= C = = 5 m/s.=10 knots.

TIK 20 s.

The bearing angle # is obtained from the transit time difference T1 as #1 1500 m/s.0.0675 #=inv.sin(c.-)=inv.sin--=inv.sin 1 d 100 m.

(p= 900.

The course angle r is likewise equal to 90 , since there is no tangential speed component.

A renewed measurement is carried out after a time of 900 s. In the intensity pattern unit 30 a section Z is formed according to Fig. 7. The following measured values are ascertained: the time shift T,K=10 S., the transit time d difference rI=-=0.067 s.

c the pitch of the interference line t'=1.36 s./Hz the quotient 1.1# f0=1.1#300 Hz.

With these measured data the range is determined as r#4500 m.

and the approach speed as Vr=10 m/s.=20 knots.

bearing angle and course angle amount to: 0 = 90 .

From this example it is seen that independently of the preceding and succeeding movement behaviours of the vehicle, the target data are correctly determined.

Next the question arises whether the method for the determination of target data can be used even if the course of the vehicle 1 does not proceed over the measurement location 3, but passes the measurement location 3 at a transverse distance and the interference lines thus have a hyperbolic course. Substantiation is to be obtained by reference to the sketch of principle according to Fig. 8.

The vehicle 1 is situated at the moment t at a range r from the measurement location 3 at a bearing angle # in relation to the reference direction N. The course of the vehicle 1 proceeds at a course angle &gamma; in relation to the reference direction N and has a transverse distance q from the measurement location 3. The vehicle 1 is travelling at a vehicle speed V and has passed the transverse distance q at the position R at the moment tufa. Between the vehicle speed V and the radial speed component Vr there lies a speed angle a.

In the time AT=tItcpA the vehicle 1 has travelled at the vehicle speed V the distance s: s=V . AT.

According to Pythagoras's theoreum the following geometrical relationship is valid: r=q+s=q+V . ##.

(V . ##) is bracketed out and divided by V . ##: r r q -- # - = ##±- (I).

V # ## V V. ## Furthermore for the speed triangle of V, Vr and V# there is valid: Vr cosa -.

V For the triangle with measurement location 3, vehicle 1 and position R as apices there is valid: s V.## cos&alpha;=-=--.

r r Thus V # ## Vr --= r V is valid and for equation I one obtains: r q2 (Il) Vr V2 AT The first term AT on the right side of equation II is just the time which elapses when the vehicle 1 at the vehicle speed V travels the distance s. The second term q V2 Ar is a time period AT which would elapse if the vehicle were to travel the distance q from the measurement location 3 to the position R at an imaginary speed V* pointing in the direction of the transverse distance q, as shown by the following conversion:: With s=V Ar there results 2 2 q q ##=--=--. (III) V # ## s.V Furthermore according to Fig. 8 q - tana S which is inserted into equation Ill q AT= - tana (IV) V and according to Fig. 8 V -=V* tan&alpha; is valid which is resolved according to tana and inserted into equation IV. Thus: q2 q AT= -- V2. AT V* is valid.

Now it can be imagined that the vehicle 1 has reached its present position from the measurement location 3 with the radial speed component V, over the distance r after a time TIS=.

V, On the other hand it can also have reached this position in that it travelled from the measurement location 3 the transverse distance q with the imaginary speed V* in the time q -=AT V" and then the distance s. with the vehicle speed V in the time AT=t,tcpA Since however the vehicle 1 at the time t1 possesses the illustrated position independently of the path taken, r1s=Ar+AT= Vr However the quotient V, is just equal, according to equation (A), to the product of the pitch t' of one of the interference lines in the section at the mean frequency f0 and the factor X(fo) --1.1f0.

X(fo) Thus for any desired course of the vehicle 1 it is also valid that the distance r between measurement location 3 and vehicle 1 can be determined from the pitch t' of the interference line and the radial speed component V, in accordance with equation (A) on page 53 at any moment.

The range r is calculated in the range calculator 55 according to Fig. 2. The radial speed component V, of the vehicle speed V is present at the output of the calculator circuit 40. The vehicle speed V is calculated from the radial speed component V, and the tangential speed component V# according to Pythagoras's theorum in a vehicle speed calculator 60 which is connected on the input the range calculator 55:

side with the V= VV2,+V?t. calculator circuit 40 and through a multiplier circuit 61 with The product of range r and the time variation of the angle U IS caicuiatea In tne multiplier circuit i.I flis angie variation In time is trequently called angle speed 9. The geometric relationships between the speed components V,, V" of the vehicle speed V in relation to the measurement location 3 can be seen from Fig. 4. The angle a is determined from the transit time difference r2 in a calculator 62 following the maximum detector 35 in accordance with the equation c.T2 a=inv.sin-.

d The calculator 62 is followed by a differentiation circuit 63 in which the time variation of the angle # is determined per time unit T. The multiplier circuit 61 is connected by its section input with the differentiator circuit 63 and calculates the product of range r and time variation of the angle b. This product is equal to the tangential speed component VD=r. b.

For the calculation of the north-related course angle y the interference line calculator 50 and the differentiator circuit 63 are followed by an angle calculator stage 64 in which a speed angle a, as stated in Fig. 4, is calculated. The speed angle a lies between the radial speed component Vr and the vehicle speed V and reads

X(f0) &alpha;=inv.tan #.t'.### (D) X'(f0) This expression can be derived from the geometrical arrangement according to Fig. 4 and using equation (C) in the following way: V# #IK #IK X(f0) #2 tan&alpha;=-=- .r@##=-.##. --.t'.c.--.

V@ c.#2 c.#2 X'(f0) #IK thus X(fo) tana .t'. 0.

Xt(fo) this equation resolved according to a produces equation (D).

The vehicle speed V can also be calculated from the speed angle a and the radial speed component Vr according to the equation V, V= cosa The angle calculator stage 64 is followed by a totalling circuit 65 which receives as second input value the north-related bearing angle 6, from which the north-related course angle ,=6-a-180 is calculated according to Fig. 1.

Fig. 9 shows a modification of the interference line calculator 50. The approximation calculator 51 here contains for the ascertaining of the approximation of an interference line in the section of the intensity pattern and of a straight line produced in the simulation calculator 52, a selector circuit 69 which seeks out the intensities in the section which lie on the straight line, and a mean value former 70 in which the intensities I, are totalled along the straight line and divided by their number N:

In a squarer 71 the individual intensities I, along the straight line are squared and in a subsequent totaller 72 they are added and divided by the number N.One obtains the mean value of the squared intensities:

The mean value former 70 and totaller 72 are followed by a calculator circuit 73 in which the relative standard deviation #/I of the intensities li is calculated along the straight line from their mean value I according to the formula

Its output is connected with a monitor circuit 74 for the release signal of the approximation calculator 51. The monitor circuit 74 gives off a release signal when the relative standard deviation # I is as small as possible and smaller than a prederminable value. Then the intensities Ij along the straight line are nearly equal and the straight line approximates best to the interference line.The release signal is fed to the pitch calculator 53 in which the straight line t(f) simulated in the simulation calculator 52 is differentiated according to the frequency f. If no release signal is produced the straight line is turned in the simulation calculator 52 and/or shifted in the time direction until the relative standard deviation a is at the minimum.

It is likewise possible in the simulation calculator 52 in place of a straight line to simulate a family of straight lines which all intersect at the frequency -0.1f, and possess equal intervals on the frequency track of the mean frequency fOg The selector circuit 69 then seeks out the corresponding intensities, pertaining to the co-ordinates of the simulated straight line, from the section of the intensity pattern which intensities are further processed in the mean value former 70 and squarer 71 for each straight line. For all straight lines the relative standard deviation T is calculated and the straight lines are approximated to the interference lines in the section.The monitor circuit 74 generates a release signal when the relative standard deviation # T is at the minimum for all straight lines. From the simulation calculator 52 the straight line proceeding through the middle of the section is transferred to the pitch calculator 53.

The case of a depth variation in the transmission stratum is to be described below If the measurement location with the transducers is situated for example in a shallow-water region having no constant water depth, the determination of the radial speed component Vr from the time shift TIK is no longer independent of the ship's location and of the water depth prevailing at the ship's location.

Fig. 10 shows a sketch of principle of a shallow-water channel in which for the sake of simplicity a continuous depth variation is represented by two water depths H1 and H2 with a jump. By reference to this model a correction of the speed measurement is to be explained. In this model shallow-water channel two own waves interfere with one another, which in the region of the water depth H1 possess an interference wave length X, and in the region with the water depth H2 possess an interference wave length X2. The measurement location 3 is situated in the region with the water depth H1.If the vehicle 1 is situated in the region with the water depth H1, a time shift 'IK1 is measured at the measurement location 3 and together with the interval d of the transducers 4 and 5 supplies the vehicle speed V,=V according to equation (B) on page 54.

The vehicle 1 travels for example with its vehicle speed V in a time t1 such a distance as is just equal to the interference wave length XI. Since the vehicle 1 is surrounded by its interference field, an intensity maximum in the region with the water depth H2 will travel in the time t, a distance S2 which is less than the distance X1 and just equal to the interference wave length X2.

If the vehicle 1 is situated in the region with the water depth H2, then with vehicle speed V an interference maximum will travel in a time t2 a distance corresponding to the interference wave length X2. At the measurement location 3, where however in the same time t2 an interference maximum has travelled a distance corresponding to the interference wave length X1 with a measured speed V"*, there is measured X1 = V"".t2.The time t2 is determined from the interference wave length X2 and the vehicle speed V, and is X2 t2= - V If t2 is inserted into the equation for X1, one obtains V"* xl=-. X2.

V If this equation is solved according to the measured speed V"*, one obtains X1 X2 From the time shift tIK**, which is measured at the measurement location 3 when the vehicle 1 is situated in the region with the water depth H2, the measured speed V"" is known. The measured speed V** is greater than the vehicle speed V, namely V.X1 V**= X2 wherein X1 < X2.

From the above-mentioned article by Weston it is known that the interference wave lengths X1, X2 behave like the squares of the water depths H1, H2: X1 (H1)2 X2 H2 Then an estimation error for the speed V results as AV V-V"* AH (AH)2 - =-2.-+ - V"" V H1 H1 with AH--H1-H2 In most practical cases the relative depth variation AH H1 is small. Therefore the second term of this equation can be neglected. One obtains a correction factor of correct sign which is dependent solely upon the gradient of the bottom and equal to twice the relative depth variation.

On account of the variation of the interference wave lengths both the interference line pitch t' and the time shifts TIK are varied by the same factor. Since these two values according to the equation (C) enter as ratio into the formula for the range determination and the other values of this formula are not influenced by the depth variation, the range requires no correction.

For the calculation of the speed angle a the value of the interference line pitch must be corrected. Let t'" be the measured interference line pitch and t' the corrected interference line pitch value. Then the following correction is valid: t'=t'*.( H1 )2 H2 The speed angle a is then calculated with the corrected pitch of the interference line from equation (D) as follows: X(fo) (Hl )2 a=inv.tan . t'*. - . a.

Xt(fo) H2

Claims (28)

1. Method for the passive determination of target data, such as vehicle speed, distance and course, of a vehicle radiating self-generated wave energy, especially a marine vehicle, from a measurement location, in which the wave energy is received at the measurement location by transducers, and converted into electric reception signals, and a direction of incidence of the wave energy in relation to a reference direction is detected as bearing angle, characterised in that the measurement location within a measurement region is placed in a transmission stratum with dispersion properties for the wave energy radiated by the vehicle, in that at the measurement location at least two transducers are arranged spaced from one another, in that reception signals of each transducer are continuously subjected to a frequency analysis and intensities are stored away in dependence upon both the frequency and the time, in that in each case from the stored intensities of the reception signals of each transducer a section determined by a predeterminable frequency range and a predeterminable time interval is selected, in that a mutual time shift of the intensity patterns in the two sections is determined and in that on the one hand for the determination of the vehicle speed a radial speed component of the vehicle speed is obtained from the quotient multiplied by the interval of the transducers, of the sine of the bearing angle and the time shift, and/or on the other hand for the determination of the distance between measurement location and vehicle,

within one of the sections frequency-dependent interference lines are obtained from adjacent intensities of equal strength, and the frequency variation or pitch of at least one of the interference lines situated in the section is determined and the distance is obtained from the product of the pitch and the quotient, multiplied by the distance between the transducers, of the sine of the bearing angle and the time shift.

2. Method according to Claim 1, characterised in that a tangential speed component of the vehicle speed is obtained from the product of the distance and time variation of the bearing angle.

3. Method according to Claim 2, characterised in that the reception signals of the two transducers are used for the determination of the bearing angle and its time variation, in that a transit time difference of the reception signals is measured in a frequency interval placed higher than the frequency range and thence the bearing angle and its time variation are calculated, taking consideration of the interval of the transducers.

4. Method according to Claim 3, characterised in that a third transducer is set up at the measurement location in such a way that the three transducers define a preferably equilateral triangle in the transmission stratum parallel with its limiting plane, in that for the ascertainment of time shift and transit time difference the transducers are used by pairs.

5. Method according to Claim 4, characterised in that angles in relation to the central perpendicular to the distance between each transducer pair are calculated from the transit time differences, in that these angles are converted into angle values in relation to a common reference direction and compared with one another and in that the bearing angle is determined from the transit time differences pertaining to equally great angle values.

6. Method according to Claim 4 or 5, characterised in that the ascertained transit time differences and/or time displacements are compared with one another, in that for the calculation of the distance and vehicle speed the maximum transit time difference is combined with the time shift ascertained from reception signals of the same transducer pair or the maximum time shift is combined with the transit time difference obtained from reception signals of the same transducer pair.

7. Method according to Claim 6, characterised in that the transit time differences pertaining to equal angle values are compared with one another and the bearing angle and the time variation of the bearing angle are ascertained from the minimum transit time difference.

8. Method according to Claim 7, characterised in that the course is determined from the bearing angle plus a speed angle which lies between the radial speed component and the vehicle speed.

9. Method according to Claim 8, characterised in that the speed angle is calculated with the aid of the inverse tangent of the product of pitch, a factor and the time variation of the bearing angle.

10. Method according to one of Claims 1 to 9, characterised in that the frequency range is ascertained with its mean frequency in a manner in which along each frequency track a modulation measure of the stored intensities is determined within the time interval and a range of adjacent frequency tracks, for which the frequency course of the modulation measure lies above a threshold, is selected as frequency range.

11. Method according to Claim 10, characterised in that the frequency interval for the direction-finding is selected at such a frequency distance from the frequency range and its mean frequency that phase speeds of waves within this frequency interval, in dependence upon the frequency, are approximately constant and equal to the speed of propagation of the wave energy in the medium of the measurement region.

12. Method according to Claim 11, characterised in that the distance is calculated by the product of pitch, transit time difference, propagation speed and factor, divided by the time shift of the intensity patterns of the reception signals of the same transducer pair.

13. Method according to Claim 12, characterised in that the radial speed component is indicated by the transit time difference, multiplied by the speed of propagation, divided by the pertinent time shift.

14. Method according to Claim 13, characterised in that the vehicle speed is calculated by application of Pythagoras's theorum from the radial and tangential speed components, by squaring, totalling and root extraction.

1 5. Method according to Claim 12, characterised in that the interference wave length of two own waves interfering with one another in the transmission stratum is determined independence upon the frequency and its frequency derivation is formed at the mean frequency and in that the factor is formed from the quotient of interference wave length at the mean frequency and its derivation.

16. Method according to Claim 12, characterised in that the factor is selected as equal to 1. 1 times the value of the mean frequency of the frequency range.

17. Method according to one of Claims 1 to 16, characterised in that the time interval is selected as proportional to the interference wave length of two own waves interfering with one another in the transmission stratum, which develop by reason of the selected mean frequency, and comprises at least two interference lines on the frequency track of the mean frequency.

1 8. Method according to Claim 17, characterised in that the interval of the transducers is made less than half the interference wave length of two own waves interfering with one another in the transmission stratum which develop by reason of the selected mean frequency.

19. Method according to Claim 18, characterised in that the transducers are arranged within the transmission stratum at such a distance parallel to their limit plane that by reason of own waves of higher order within the section more than two interference lines are to be noted on the frequency track of the mean frequency.

20. Method according to one of Claims 1 to 19, characterised in that short-time power density spectra are formed in predeterminable time units from the reception signals of each transducer for frequency analysis and are stored in relation to a time base in each case as intensities in dependence upon the frequency, in that the time base is screened into time units and the time interval comprises a predeterminable number of time units.

21. Method according to one of Claims 1 to 20, characterised in that the pitch is obtained by approximation of a straight line to the interference line.

22. Method according to Claim 21, characterised in that within a frequency-time co-ordinate system of the intensity pattern formed by frequency and time units, intensities are measured along a straight line arranged as desired in the section, in that for the approximation of the straight line to the interference line the straight line is turned and displaced in the time and/or frequency direction until the intensities measured along the straight line have the minimum deviation from one another.

23. Method according to Claim 22, characterised in that within the frequency-time co-ordinate system of the intensity pattern, intensities are measured along a straight line arranged as desired in the section, in that for the approximation of the straight line to the interference line the mean value of the intensities, which are measured along the straight line, is formed, further in that the individual intensities are squared and added and this sum is divided by the number of measured intensities, in that thence the relative standard deviation of the intensities from the mean value is formed and in that the minimum departure of the straight line from the interference line is reached when the relative standard deviation is at the minimum.

24. Method according to Claim 22 or 23, characterised in that in the frequency-time coordinate system a pattern is produced from a cluster of straight lines intersecting at --O. 1 times the value of the mean frequency, with equidistant intervals on the frequency track of the mean frequency, in that in conformity with the frequency track of the mean frequency, the section and the pattern of the cluster are displaced in relation to one another along the time base until the individual straight lines of the cluster are tangential to the interference lines and no longer intersect them, in that the pitch of the connecting line between the intersection point of the straight lines and the centre point of the section indicates the pitch of the interference line.

25. Method according to one of Claims 1 to 24, characterised in that the time intensity distribution in the section of the intensity pattern allocated to the one transducer along each frequency track in the predetermined frequency range is correlated with the time intensity distribution of the intensity pattern allocated to the other transducer along the same frequency track over the entire time interval, in that the correlation functions of all frequency tracks are averaged and the time shift is determined from the position of the maximum of the averaged correlation function.

26. Method according to Claim 25, characterised in that in the case of a measurement region with depth variation between the limit planes the ascertained radial speed component is corrected by twice the amount of the relative depth variation, in dependence upon the relative depth variation related to the depth at the measurement location.

27. Method according to Claim 26, characterised in that the pitch is multiplied by the squared quotient of depth at the location of the vehicle and depth at the measurement location.

CLAIMS Amendments to the claims have been filed, and have the following effect: New or textually amended claims have been filed as follows:

1. Method for the passive determination of target data, such as vehicle speed, distance and course, of a vehicle radiating self-generated wave energy, especially a marine vehicle, from a measurement location, in which the wave energy is received at the measurement location by transducers, and converted into electric reception signals, and a direction of incidence of the wave energy in relation to a reference direction is detected as bearing angle, characterised in that the measurement location within a measurement region is placed in a transmission stratum with dispersion properties for the wave energy radiated by the vehicle, in that at the measurement location at least two transducers are arranged spaced from one another, the reception signals of each transducer are continuously subjected to a frequency analysis and intensities are stored away in dependence upon both frequency and time, in that in each case from the stored intensities of the reception signals of each transducer a section determined by a predeterminable frequency range and a predeterminable time interval for evaluating the intensity patterns is selected in that a mutual time shift of the intensity pattern in the two sections is determined and in that on the one hand for the determination of the vehicle speed a radial speed component of the vehicle speed is obtained from the interval between the transducers multiplied by the quotient of the sine of the bearing angle relating to the central perpendicular to the connection of the two transducers and the time shift, and/or on the other hand for the determination of the distance between measurement location and vehicle, within one of the sections frequencydependent interference lines are obtained from adjacent intensities of equal strength, and the frequency variation or pitch of at least one of the interference lines situated in the section is determined and the distance is obtained from the product of the pitch and the interval between the transducers, multiplied by the quotient of the sine of the bearing angle and the time shift.

5. Method according to claim 4, characterised in that bearing angles in relation to the central perpendicular to the distance between each transducer pair are calculated from the transit time differences, in that these bearing angles are converted into angle values in relation to a common reference direction and compared with one another and in that a north oriented bearing angle is determined from the transit time differences pertaining to equally great angle values.

8. Method according to claim 7, characterised in that the course is determined from the north oriented bearing angle plus a speed angle which lies between the radial speed component and the vehicle speed.

28. A method for the passive determination of target data substantially as herein described with reference to the accompanying drawings.