GB2177796A

GB2177796A - Passive sonar systems

Info

Publication number: GB2177796A
Application number: GB08407318A
Authority: GB
Inventors: Jochen Ziegenbein; Manfred Siegel; Gotz Bodecker
Original assignee: Krupp Atlas Elektronik GmbH
Current assignee: Atlas Elektronik GmbH
Priority date: 1984-03-21
Filing date: 1984-03-21
Publication date: 1987-01-28
Also published as: GB2177796B; GB8407318D0

Abstract

To estimate parameters such as range, speed, course and/or transmission frequency of a moving target S radiating acoustic pulses into the water, from a receiver E remote from the target, echoes received in an off-target direction I by reflections from discontinuities SC, SC2 in the sea are monitored. Each discontinuity in the direction I will reflect a pulse transmitted by the target S with a different Doppler shift, varying as a function of the bearing of the discontinuity relative to the target. The frequency-time characteristic of the total acoustic radiation received from the direction I is determined, and it is shown that the required parameters can then be deduced by curve-fitting techniques. It is further shown that if acoustic radiation from three directions is monitored, are on-target and two off-target, some of the parameters can be calculated directly from the three frequency-time characteristics, without resort to curve-fitting. <IMAGE>

Description

SPECIFICATION Passive method for the estimation of magnitudes of state of a moving target radiating acoustic pulses into the water The invention relates to a passive method for the estimation of magnitudes of state, such as range, speed, course and/or transmission frequency, of a moving target radiating acoustic pulses into the water, such as a ship, torpedo or the like, with active sonar, from a reception position remote from the target.

One known method of this kind, generally called "ping steeling technique" makes use of the effect of the multi-way propagation in the acoustic channel. In this case from the transit time differences between the acoustic pulse received directly, that is in the direction-finding direction receiver to transmitter, and the acoustic pulse or pulses received by way of deviating paths, firstly the location and, by subsequent time-integrating processing, also the speed of the transmitter are approximately determined. This method however presumes good knowledge as to the sound propagation conditions prevailing in each case. This method cannot be used in shallow-water regions with mostly slight knowledge of the properties of the shallow-water channel.

The invention is based upon the problem of indicating a passive method of the initially stated kind with which the magnitudes of state of a target can be determined with relatively high accuracy independently of the knowledge of the properties of the acoustic channel. The method is to be suitable especially for shallow-water regions. At the same time this method is to be practicable at the receiver end with usual antennae or bases such as known passive sonar installations comprise, and all constructional extra expense, especially for the antenna or base, is to be avoided.

In a method of the kind stated in the opening statement of Claim 1, the problem is solved by the features in the characterising part of Claim 1.

In the method according to the invention specifically the property of the shallow-water channel which was hitherto felt to be troublesome for acoustic location is exploited, namely the increasedly occurring reverberation with the effects of the Doppler-caused frequency spread in the reverberation of a location signal and the direction- and time-dependence of this frequency spread. The method according to the invention has the advantage that no additional construction expense is necessary. The method is realised without any constructional modification of conventional passive sonar installations with for example a cylinder base, exclusively by signal-processing means.Part steps in signal processing, beam-forming and frequency analysis necessary for the method are in any case already available in a series of known passive sonar installations, so that the method can be implemented with minimum extra expense in existing sonar installations.

The method according to the invention works very accurately. Even under unfavourable marginal conditions it supplies estimated values for the distance between the transmitter-carrying target and the receiver, with an error Qf less than 10%.

An advantageous development of the invention appears from Claim 5. By additional determination of the Doppler frequency of the transmitted pulse, contained in the direct signal, that is in the signal arriving on the direction-finding bearing, the course of the target can be ascertained with the magnitudes of state, transmission frequency and amount of the target speed, determined according to Claims 3 and 4.

An advantageous development of the method according to the invention appears from Claim 6. In the two non-target-pointing selective reception directions or direction channels of the sonar installation one obtains complementary extremes of the Doppler reverberation frequencies, that is in the one direction channel a maximum and in the other a minimum Doppler reverberation frequency is detected. By selective reception direction there are here understood ordinary angles of aperture 2s 3 of the reception characteristics. An improvement of the accuracy of detection of the Doppler reverberation frequencies is achieved with reduction of the angle of aperture 2; 3.

Angles of aperture 2 3=2 , ordinarily generated nowadays in passive sonars are quite adequate for good results. Due to the directed reception of the reverberation the limitation to a stationary or almost stationary receiver necessary in the case of undirected or omnidirectional reception, is eliminated. Rather the receiver itself can move without limitation. Th so-called intrinsic Doppler then occurring can readily be eliminated by computation, as the speed and course of the receiver are known.

An advantageous development of the method according to the invention also appears from Claim 7. By these measures one avoids measurement inaccuracies which can occur in the development of the method according to Claim 6 if the two non-target-pointing selective reception directions are unfavourably selected in relation to the as yet unknown course of the target.

An advantageous development of the method according to the invention appears from Claim 8. With the specified magnitudes of state target course, target speed and transmission frequency of the target transmitter it is possible by means of the stated measure to ascertain the distance to the target and thus the target location, with sufficient accuracy.

An advantageous development of the method according to the invention appears from Claim 9. By these measures it is possible substantially to increase the reliability of the range estimation. Disturbances in the reverberation structure and thus incorrect support points can be eliminated due to the multiplicity of support values from the most various directions of reception, utilised for the estimation.

An advantageous development of the method according to the invention here appears from Claim 11. The use of the magnitudes of state of the target, ascertained according to Claims 3 to 10, as starting values for the laying down of the parameters permits of considerably reducing the computation expense for the estimation method.

For the method as described above, after the direction of the target has been found, in principle only one single transmitted pulse suffices to detect the defined magnitudes of state of the target completely. The evaluation of further transmitted pulses in the manner described serves merely for the improvement of the result of estimation of the magnitudes of state.

If however a series of acoustic pulses radiated from the target is available, the magnitude of state of target range can additionally also be estimated according to the further development of the method according to the invention in accordance with Claim 12. In combination with the evaluation methods stated in Claim 9 or 10 then one obtains two separately ascertained results of the same magnitude of state of target range, with which then by means of an errorcompensation method it is possible further to improve the actual estimation result.

An advantageous development of the method according to the invention appears from Claim 13. This procedure has the advantage of managing with only one, but at maximum with only three selective reception directions which are offset in relation to one another at the azimuth by an angle amount. Since the entire reverberation structure is used throughout the whole reception duration for the ascertainment of the Doppler reverberation frequencies-and not only selected support values-disturbances in the reverberation structure can easily be recognised and readily be eliminated in the computation of the magnitudes of state. The electronic expense for the formation of the maximum number of three reception directions or what are called Preformed Beams is relatively slight.

An advantageous development of the method according to the invention appears from Claim 14. By the direction-finding of the transmitter by means of a separate direction-finding beam it is possible reliably to detect both the moment in time of the reverberation detection and a maximum and a minimum value of the Doppler frequencies, as a result of the very high signal-tonoise ratio, the maximum and minimum values lying symmetrically of the mean frequency of the transmitted pulse. The impulse function in the time course of the Doppler reverberation frequency, occurring in the direction-finding beam, renders possible a reliable ascertainment of the mean frequency, that is of the transmission frequency.

An advantageous development of the method according to the invention also appears from Claim 15, especially in combination with Claims 16 and 17. As also in the development of the method as described at the outset, here again the speed and course of the target are determined from the extreme values of the Doppler reverberation frequency, that is the maximum and/or minimum Doppler reverberation frequency. Since however, in contrast to this method variant, here for one reception direction in each case the entire time course of the Doppler reverberation frequency, called the reverberation Doppler frequency time curve, is determined over the entire duration of reception of the reverberation, disturbances in the reverberation structure can easily be recognised and the actual extreme values, which correspond to the minimum and maximum Dopplers, can be determined very much more reliably.

A further advantageous development of the method according to the invention appears from Claim 18. With certain spatial relationships of target course and selected non-target-pointing selective reception direction it is not possible by measurement technique to ascertain the minimum or maximum Doppler reverberation frequency. In this case it is however possible to determine the minimum or maximum Doppler reverberation frequency which cannot be detected in the first reception direction, from the Doppler reverberation frequency-time values in the further non-target-pointing selective reception direction according to the invention, which has a different spatial relationship to the transmitter.

A further advantageous development of the method according to the invention appears from Claim 19. By these measures the reliability of the calculated magnitudes of state of the target is substantially increased in that it is always that reception direction in which the unambiguous extreme value of the Doppler reverberation frequency occurs which is used for the calculation of the magnitudes of state.

An advantageous development of the method according to the invention also appears from Claim 23. By these additional method steps it is possible for the results of estimation for the unknown magnitudes of state to be iteratively substantially improved and thus for an extremely precise target detection and target determination to be achieved.

The method according to the invention is usable not only when the target is emitting acoustic pulses omni-directionally. Even with other types of transmission which are frequently used in active sonars, such as RDT-, CRDT- or SRDT-operations, in which a narrow transmitted beam is pivoted over a horizontal angular range, the method according to the invention delivers equally good results for the magnitudes of state of the transmitting target.

According to the further development of the method in accordance with the invention according to Claim 25 a conclusion is drawn as to a target with active sonar transmitting in RDT operation when a time shift occurs between the reverberation detection in the two non-targetpointing reception directions. From this time shift it is then additionally possible to calculate the period of rotation of the transmitted beam.

According to the further development of the method in accordance with the invention according to Claim 26 it is also possible to deduce a target with underwater transmitter working in RDT operation when in what is called the direction-finding beam, that is in the target-pointing selective reception direction, a time shift occurs between the arrival of the transmitted pulse and that of the reverberation. This time shift too is a measure for the speed of rotation of the transmitted beam.

The invention will be described in greater detail below by reference to examples of execution, clarified by the drawing, of a method for the passive estimation of magnitudes of state of a target.

Figure 1 shows an illustration of principle of a model of the reverberation space in water, in an arbitrarily selected momentary spatial relationship between a moving transmitter S and a stationary or moving receiver E, Figure 2 shows a similar illustration to Fig. 1 in the case of a receiver having a total of three selective reception directions-, Figure 3 shows an illustration of principle of the Doppler reverberation frequency-time curves, detected in the individual selective reception directions, in the case of an ODT transmitter, Figure 4 shows a similar illustration of Doppler reverberation frequency-time curves in the case of an RDT transmitter, Figure 5 and Figure 6 show a block circuit diagram of a circuit arrangement for the realisation of the method of estimation of the magnitudes of state, Figure 7 shows a block circuit diagram of a data extractor in the circuit arrangement according to Figs. 5 and 6, Figure 8 shows a block circuit diagram of the course of the method according to a second example of execution.

The method for the estimation of the unknown magnitudes of state of a moving target from a reception point remote from the target will firstly be explained by reference to the illustration of principle in Fig. 1. It is here presumed that the target is emitting acoustic energy, for example acoustic pulses, at intervals. The moving target therefore is a surface ship, in the preferred application of the method, ordinarily having an active sonar on board for location tasks, the acoustic transmitter of which emits acoustic pulses, for example narrow-band CW pulses. The target or surface ship with its so-called intercept transmitter is designated by S in Fig. 1.It is travelling at the speed vs unknown to the receiver on a course ks. The stationary or moving receiver E, in the preferred application of the method, is a submarine which is stationary or travelling at the speed VF on the course kE, with a passive sonar installation with which the acoustic pulses or intercept signals can be received. The use of the method presumes that the acoustic channel between transmitter S and receiver E possesses reverberation properties, which is correct especially for shallow-water regions such as the North Sea. The reverberation is caused by discontinuities in the water, which occur in sea water as a result of differences of temperature or salinity, air inclusions, particle or micro-organism content, and effect jumps in impedance.On impact of the acoustic energy radiated by the transmitter these discontinuities give rise to reflections and scatter. In concept therefore these discontinuities can be interpreted as imaginary scattering centres SC, which are subjected to sound waves from the moving transmitter with the frequency

by an acoustic pulse of mean frequency f,,,. ss, here is the angle between the direction of travel of the transmitter and the direction in which the respective scattering centre SC, is seen from the transmitter.A part of the acoustic energy is scattered in the spatially selective reception direction of the receiver so that these scattering centres SC, appear to the receiver as imaginary transmitters, adjacent along the reception direction or reception beam axis, with different frequency ffsc, If the receiver is stationary these different frequencies fsc, can be detected directly in the selective reception channel of the receiver.If the receiver is moving at the speed VE these frequencies are shifted by a further Doppler, called the intrinsic Doppler, which results from the movement of the receiver in relation to the scattering centres SCj, and are traced in the receiver as

wherein 0 is the angle between the speed vector VE of the receiver and the direction in which the imaginary transmitters are seen from the receiver, that is the selective direction of reception of the receiver. Since the speed vector of the receiver and the selective reception direction are known, the intrinsic Doppler in the reception channel can be compensated and thus the transmission frequency fsci of the imaginary transmitters SCí can be detected.

The intrinsic-Doppler-compensated frequencies fssci, which are identical with the frequencies fsc radiated by the imaginary transmitters SC,, will hereinafter be called Doppler reverberation frequencies.

With the method as described in detail below now the magnitudes of state of the target with intercept transmitter S, which are unknown to the receiver E, will be estimated. By magnitudes of state there are understood the course k5 and the speed v5 of the target S, the transmission or mean frequency fm of the target intercept transmitter and the distance R of the target S from the receiver E. With these magnitudes of state an unknown target S can be located from the reception point E and the target behaviour can be described completely by course, speed and transmission frequency.

The receiver E has at least one selective reception direction I, called a Preformed Beam or directed reception channel. This reception direction I is arbitrarily selected, but may not be directed immediately at the target S, which is hereinafter called "non-target-pointing''. By way of the selective reception direction I or the directed reception channel the reverberation produced by the acoustic pulses of time duration T by reason of the initialiy described physical phenomenon, in the water, is detected. This reverberation is a function of the time and is also called reverberation signal.The frequency spectra are formed from the reverberation received by way of the selective reception direction I, namely for a plurality of time moments of a time pattern running from the reverberation interpretation onwards, that is from the moment of reverberation detection in the reception direction I, the Doppler reverberation frequencies sc, contained in the frequency spectra are determined and allocated to the respective time moment t,. The multiplicity of these Doppler reverberation frequency-time values results in a diagrammatically represented Doppler reverberation frequency-time curve f,,=g(t), designated by I in Fig. 3. It is here assumed that the speed VE of the receiver is zero.If however the receiver E is travelling with the known speed VE on the known course kE, then a direction-dependent speed compensation must be carried out in the receiver for the compensation of the resultant additional Doppler, called intrinsic Doppler.

Independently of the gaining of the Doppler reverberation frequency-time values out of the reverberation in the pre-determined, non-target-pointing reception direction I, for the same reception direction the Doppler reverberation frequencies f, are calculated as a function of the time t.

is valid for the Doppler reverberation frequency f, at the time t=t,.

With the relationship fl,=ir-k5-a (2) visible from Fig. 1, and the equation 6,=arc sin ii(K2+1ì '. (FKtVk2-f2+1) (3) with K=(L cos a-1) (L-sin a) ' (4) c.t, L=1+ (5ì R F=L ' (6) being set, the result is f,=h(t,, R, V5, k5, fm) (7).

From equation (7) it can be seen that the Doppler frequencies f, to be calculated are a function of the independent variables t and of the parameters R, v5, ks, fm With equation (1) now the Doppler reverberation frequencies f, are calculated for a plurality of successive time points t, and assembled into smoothing curves f=h(t). The unknown parameters R, v,, k5, fm are here prestated as estimated values. The estimated values, which are assumed arbitrarily but with reference to reality, are varied each for one parameter, the variation steps being selected suitably, and a smoothing curve is produced for each estimated value.Now the variance q2 between the smoothing curves f=h(t) and the Doppler reverberation frequency-time values f,,=g(t) obtained from the measured values (Doppler reverberation frequency-time curve, as represented under I in Fig. 3) is calculated. Among the calculated variances the variance minimum is ascertained (LMS estimation). That estimated value of each set of parameters the associated smoothing curve of which produces the variance minimum is issued as the magnitude of state of the target S.

With the existing four parameters, which must all be varied successively in suitable steps, in practice the computation expense is very great. This however can be simplified substantially in that by means of the Doppler reverberation frequencies fsc, obtained from the reverberation one calculates the magnitudes of state for the mean frequency f,,, the target speed v5 and the target course k5, so that only the magnitude of state of target range R remains as a parameter in the smoothing curves with the time as independent variable. The variation of the estimated values for the individual parameter R and the variance calculation then require only a fraction of the previously necessary computation expense.

For the calculation of the magnitudes of state fm, v5 and k5 the receiver E, as represented in Fig. 2, receives an additional selective reception direction 0 which is directed at the target S.

The Preformed Beam or directed reception channel is therefore also called direction-finding beam.

The reverberation is now additionally detected in the target-pointing reception direction 0. In the manner as described the frequency spectra of the detected reverberation and thence the Doppler frequency-time values f,,=g(t) are determined. The time pattern t, begins here with arrival of the direct intercept signal, which on account of the direct reception coincides with the reverberation interpretation, that is with the moment of the reverberation detection. The course of the Doppler reverberation frequency-time curve resulting from the Doppler reverberation frequency-time values, for the target-pointing reception direction 0 is represented in Fig. 3 and there designated by 0.As may be seen there, the course of the Doppler reverberation frequency over the time is characterised by an impulse function which jumps at the moment t,=0 from a smallest to a greatest value-or equally conversely in the case of the opposite target course -and then remains constant. If the course of the target S lies on the straight line connecting receiver and target, the smallest and the greatest values correspond to the minimum and maximum Doppler reverberation frequencies.In all other cases these extreme values, at the zero point, hereinafter called f,,,(+O) and f,?,(-O), are smaller than the minimum and maximum Doppler reverberation frequencies f"|" and and f,,,,, respectively, but always lie symmetrically in relation to the transmission or mean frequency f-.

From the upper and lower extreme values at the point t= +0 the mean frequency fm is determined by fin 2 If,.x( ì+fnx(+ )l (8ì and thence the radial speed component of the target S as c . f (9).

From the Doppler reverberation frequency-time values gained in the non-target-pointing reception direction I an extreme value ft.x is determined which is either the maximum or the minimum Doppler reverberation frequency fmax or fm,n. With this extreme value f,,, and the calculated mean frequency fm the target speed is calculated by c v5=(fex-fm) # - (10), fm the frequency difference Af=fe, fm ordinarily being called Doppler shift or half Doppler band width.

From the radial speed component Vsrad and the speed V5 it is possible to calculate the course k5 of the target S by vsrad k5=arc cos - (11).

vs In the case of specific spatial relationships of target S and selected non-target-pointing selective reception direction I of the receiver E it is not possible to ascertain by measurement technique a minimum or maximum Doppler reverberation frequency fmin or fm,, Especially for great target ranges for greater times t, the signal-to-noise ratio is too small, so that the Doppler reverberation frequencies fall off greatly. Approximately fixed maximum or minimum Doppler reverberation frequencies f,,, (fmax or fain) would be affected by great errors which would considerably falsify the magnitudes of state to be estimated.In order even in these cases to achieve a reliable estimation of magnitudes of state with little error, the receiver E-as shown in Fig.

2-receives a further non-target E-as shown in Fig. 2-receives a further non-target-pointing reception direction II, in which likewise the reverberation is detected and, in the same manner as in the first non-target-pointing reception direction I, the Doppler reverberation frequencies fisc, are determined over a time pattern running from reverberation interpretation onwards, that is from the moment of the reverberation detection. An example of the resultant Doppler reverberation frequency-time curve f,,,=g(t) in the reception direction II is represented and characterised by II in Fig. 3.The second non-target-pointing reception direction II is pivoted through an angle in relation to the first non-target-pointing reception direction I and preferably lies symmetrically thereof, in relation to the target-pointing reception direction 0 as axis of symmetry..

The extreme Doppler reverberation frequencies f,,,, or fmin are determined in the same way from the Doppler reverberation frequency-time values f,,,=g(t) obtained in the further non-targetpointing selective reception direction II. If at least one maximum or minimum Doppler reverberation frequency occurs in each of the two non-target-pointing reception directions I and II, then the Doppler shifts Af=f,.,-f,, are ascertained therewith. The greatest Doppler shift is then used for the determination of speed v5 and course k5 of the target S according to equations (10) hod (11).

The production of the smoothing curves f,=h(t) and the variance calculation also tEs place in relation to that of the two non-target-pointing reception directions I and 11 in which the greatest Doppler shift Af x occurs. In the case where in the non-target-pointing reception directions I and II in all at least two equal greatest Doppler shifts Af,,,,, occur, as applies to the example shown in Fig. 3, that reception direction will be sought out in which the greatest Doppler shift occurs earlier in time. In Fig. 3 this would be the second non-target-pointing selective reception direction II in which the minimum Doppler reverberation frequency f,,,,,, is detected as first in time.

In the target speed calculation v5 according to equation (10) the amount of the speed v5 is signed and has a positive or negative sign according to the utilised extreme value f,,, or f,,rn Taking consideration of this sign and of the selected non-target-pointing reception direction I or II, equation (11) for the target course determination k5 can be written more generally as

wherein x is the selected non-target-pointing reception direction I or II and should be inserted as 1 or 2 as the case may be.

The magnitudes of state of the target S, such as v,, k5, f,,, and R, obtained by the scanning of the reverberation space with the aid of the three selective reception directions 0, I, II sketched in Fig. 2, which already display a quite good accuracy, can be improved iteratively by the following procedure: The range value R ascertained after production of the smoothing curves by means of the variance calculation as fixed estimated value is inserted as parameter and a smoothing curve f.=h(t) is produced by calculation of the Doppler reverberation frequencies according to equation (1).One of the other parameters, for example the mean frequency f-, is varied in stages starting from the calculated values at f according to equayion (8) and in each case the smoothing curve is calculated. By variance calculation for the Doppler reverberation frequency-time values f,,,=g(t) obtained from the reverberation and determination of the variance minimum amin an improved value is obtained for the parameter concerned, in examples for the mean frequency f,. With this improved estimated parameter value the calculation of the target range R takes place again as already described and one obtains a further improved estimated value for the target range. With this improved estimated value for the target range again smoothing curves are formed with corresponding variations of a further parameter, for example the target speed V5, whereafter the described method is repeated. In total the above-described method steps are iteratively repeated until the change of the continuously improved estimated values for the target range no longer exceeds a predetermined amount.

In the method explained by means of the Doppler reverberation frequency-time curve in Fig. 3 it is presupposed that the intercept transmitter of the target S is transmitting omnidirectionally.

In active sonar installations however the possibility frequently exists of changing the type of transmission. One of the most usual further types of transmission is what is called RDT (Rotational Directional Transmission) with the modifications CRDT and XRDT. In all these types of transmission a beamed transmitted ray or transmission beam is pivoted over a more or less large horizontal angle.In the case of the RDT transmitter a transmitted beam rotates over the full horizontal angle of 360". In the CRDT transmitter three transmitted beams, staggered by 1200 in relation to one another, are pivoted in the same direction over a horizontal angle of 1200. In the case of the XRDT transmitter four transmitted beams staggered by 90 in relation to one another are pivoted in the same direction over an angle range of 90".

Even with targets S with such intercept transmitters the stated magnitudes of state can be ascertained in the same manner. In Fig. 4 the Doppler reverberation frequency-time curves gained from the reverberation in the selective reception direction 0, I, II are represented by way of example for a target S having an RDT transmitter. The zero point of the time pattern for the ascertainment of the Doppler reverberation frequency-time values f,,,=g(t) from the reverberation is here fixed by the point in time of the arrival of the acoustic pulse or intercept signal in the receiver from the target-pointing reception direction 0.As Fig. 4 shows, the time point of the arrival of the direct signal and reverberation interpretation, that is the moment of commencement of the reverberation reception, in the target-pointing reception direction 0 do not coincide, but display a time stagger. From this time stagger the presence of an RDT transmitter is deduced.

The rotation period TUM of the transmitted beam is calculated as twice the time stagger. From the rotation period TUM it is readily possible to determine the angular speed zD of the transmitted beam.

As appears from the Doppler reverberation frequency-time curves in Fig. 4 in the non-targetpointing reception directions I and li, the moments of commencement of the reverberation interpretation in the two reception directions I and II do not coincide as in an ODT transmitter, but are likewise staggered in time in relation to one another. This time stagger too is characteristic of the presence of an RDT transmitter in the target. The time stagger corresponds exactly to the rotation period TUM of the transmitted beam of the RDT transmitter.

The calculation and estimation of the unknown magnitudes of state f,,, v5, k5, R take place in the same manner as described above for the configuration of an ODT transmitter. As shown by the Doppler reverberation frequency-time curves f,,,=g(t) in Fig. 4, under some circumstances ambiguities of the function occur in one of the non-target-pointing reception directions, here in the reception direction II. The reason for this is essentially that due to the step by step spatially non-simultaneous sound action upon the reverberation space, at specific moments two different frequencies of the reverberation spectrum can occur at the same time.The variance calculation is expediently effected in relation to the Doppler reverberation frequency-time values from that one of the two non-target-pointing reception directions in which no ambiguities arise. In Fig. 4 this would be the reception direction I. The calculation of the Doppler reverberation frequencies according to equation (1) and the production of the smoothing curves f,=h(t) must then of course be carried out with this selected reception direction taken into account.

In Figs. 5 and 6 there is diagrammatically represented a block circuit diagram of a possible circuit arrangement in the receiver E for carrying out the method as described of estimation of the unknown magnitude of state of a target with acoustic radiation.

The receiver E comprises a beam former 10 known per se by means of which three directed reception channels are produced so that the receiver E is sensitive only in three selective reception direction 0, I, II. The directed reception channels or beams are designated in Fig. 5 and below in conformity with the selective reception directions by 0, I, II. The middle reception channel 0, called the direction-finding beam, is oriented to the target S (target-pointing reception direction 0), the other two reception channels I and II (non-target-pointing reception directions I and II) lie symmetrically about the direction-finding beam 0. The received signals of the individual reception channels 0, I, II are processed separately. For this purpose after each reception channel 0, I, li there are connected an FFT-processor 11, a data extractor 12 and a minimummaximum locator 13.The allocation of these components is characterised by a numeral appended to the reference and selected according to the reception channels 0, I, II, so that for example of the components connected to the middle reception channels 0, the data extractor 12 is characterised with the reference 120 and the minimum-maximum locator 13 with the reference 130. The FFT processors 11 in each case estimate the amount spectrum Is,(f)l of the received signals s(t). The spectrograms are each fed to the data extractor 12. This decides whether an acoustic pulse is detected, and in this case extracts the beginning and end of the reverberation produced by it, also its course over the time.As a result all Doppler frequency-time values fn=g(n) are obtained, as represented in Figs. 3 and 4 as Doppler reverberation frequencies fsc over the time t, beginning with reverberation inerpretation at the moment t=O.

One possible form of embodiment of a data extractor 12 is represented in Fig. 7. From the spectrograms 1S0(f)I delivered at the moments n a maximum locator 14 extracts the frequency with the greatest amplitude, the Doppler reverberation frequency f0. The Doppler reverberation frequencies fo are fed by means of a gate member 15 to the respective minimum-maximum locator 13 if the scatter & does not exceed a predetermined amount rr,. For this purpose all Doppler reverberation frequencies f0 detected at the various moments n are written into a shift register 16 with serial input and parallel output.From the content in each case of the shift register 16 at each moment n the arithmetic mean f0 is formed by a mean value former 17.

From this mean value and each Doppler reverberation frequency fn, in a calculation stage 18 the scatter

is calculated. For this purpose the calculation stage 18 is connected on the input side with the output of the mean value former 17 and with each of the parallel outputs of the shift registers 16. The output of the calculation stage 18 is connected with one input of a comparator 19 the other input of which is occupied with the amount of the predetermined permissible maximum scatter #ref2.

The comparator 19 gives a pass command to the control input of the gate member 15 when #n2 < #ref2 is detected. The extraction procedure delivers for each direction channel 0, I, II a set of Doppler reverberation frequency-time values f,,=g(n), as represented as curves in Figs. 3 and 4 and there designated by 0, I and li.

The extracted Doppler reverberation frequencies f are each fed to the minimum-maximum locator 13 which issues the lowest and highest Doppler reverberation frequencies fml.l and f,,,.

The minimum-maximum locator 130 allocated to the middle direction channel 0 is followed by an adder/divider 20 which calculates the mean frequency f,,, according to equation (8). One input of a subtractor 21 is connected to the output of the adder/divider 20 and the other input is connected to one of the outputs of the minimum-maximum locator 130.The subtractor 21 calculates the frequency difference between one of the largest or smallest Doppler reverberation frequencies, called angle frequencies, at the moment n=0, and the mean frequency f,,,. The subtractor 21 is followed by a multiplier/divider 22 which is connected on the input side again with the output of the adder/divider 20 and, with the introduced speed of sound c in water calculates the radial speed component vSIJ" of the target speed v5 according to equation (9).

Each minimum-maximum locator 131 and 132 is followed by a subtractor 23 and 24 respectively, which is further connected with the output of the adder/divider 20. The subtractors 23 and 24 calculate from each of the extreme values of the Doppler reverberation frequencies t (f,,,,, or fnll.l as the case may be) the Doppler shift Af by difference formation Af=f0-t,, or f=f,,,,-f,,,. The Doppler shifts Af in each direction channel are compared with one another in a comparator 25 or 26 and the largest Doppler shift in each case is issued at the output.The outputs of the two comparators 25 and 26 are connected with the two inputs of a further comparator 27 which ascertains the larger Doppler shift At,,00 of the two issued Doppler shifts and at the same time issues the code number x for that reception channel I or II in which this larger Doppler shift Af,,, occurs. This code number x, which can assume "1" or "2" according to the direction channel, forms a control value for an elector 28, for example multiplexor, to which the Doppler reverberation frequency-time values present on the outputs of the two data extractors 121 or 122 are fed.Those Doppler reverberation frequencies f, and time values n which have been obtained from that direction channel the code number x of which is present on the control input of the elector 28 are fed to a range estimation processor 29 (Fig. 6).

The output of the comparator 27 where the maximum Doppler shift Afmax is present is connected with one of three inputs of a multiplier/divider 30, the further inputs of which are occupied on the one part with the speed of sound c and on the other by connection to the adder/divider 20 with the calculated mean frequency f,,. The multiplier/divider 30 calculates the target speed vs according to equation (10). The output of the multiplier/divider 30 and the output and the multiplier/divider 22 are connected with a divider 31 in series with which an arc cos network 32 is connected. The target course k5 calculated according to equation (11) can be tapped from the output of the arc cos network 32.To take account of the sign-affected magnitude of the target speed v5 and the selected reception channel an adder 33 is also connected in series with the arc cos network 32, which adder is connected on the other side with the output of a calculator member 34. The code number x ascertained by the comparator 27 and, from the output of the multiplier/divider 30, the sign of the maximum Doppler shift At,,00 are fed to the calculator member 34. The calculator member 34 calculates the second summand of equation (12), which is added in the adder 33 to the output value of the arc cos network 32.

The absolute course k5 according to equation (12), related to the line of connection between receiver E and target S, can be taken at the output of the adder 33.

The range estimation processor 29 comprises a smoothing curve calculator 35, a variance calculator 36, a memory 37 in the form of a shift register with serial input and parallel output and a minimum detector 38. All ascertained magnitudes of state, such as the mean frequency f, the target speed V5, the target course k5 and the angle a between the target-pointing reception direction 0 and the non-target-pointing reception direction I or II and the speed of sound c in water are introduced into the smoothing curve calculator 35. Moreover the smoothing curve calculator 35 receives estimated values R1 of target range selected arbitrarily but with reference to reality, which are varied in stages j= 1 to k.In addition the time values n of the selected nontarget-pointing reception direction I or II are fed to the smoothing calculator 35, which takes place by way of the elector 28. The smoothing curve calculator 35 now calculates for the time pattern n and for each estimated value q the course of the Doppler reverberation frequencies f, according to equation (1). The result is fred to the variance calculator 36, which also receives the Doppler reverberation frequencies fn, obtained from the reverberation, in the selected nontarget-pointing reception channel I or II.The variance calculator 36 calculates the variance of all smoothing curves f,=h(n, R,) in relation to the Doppler reverberation frequencies f, according to:

The variances v2(R,ì for the various estimated values R are stored meanwhile in the memory 37. From the memory content the minimum detector 38 ascertains the variance minimum and issues the pertinent estimated value R"l"l as ascertained target range R. The code number x set forth exponentially in the equations in Fig. 6 serves merely for the characterisation of that of the two non-pointing reception channels I or II in which the greater Doppler shift Af,,j., occurs and in relation to the Doppler reverberation frequency-time values fn=g(n) of which the variance calculation takes place.

In the block circuit diagram in Fig. 8 there is sketched a similar method for the estimation of target magnitudes of state which is modified to the effect that by means of signal processing the reverberation is detected not only in three but in a plurality of selective reception directions offset in azimuth in relation to one another each through an equal angular amount, called a beam fan 40. Of the reception directions the one, called the direction-finding beam 41, is directed at the target. The direction-finding beam is here preferably situated in the middle of the beam fan 40. The actual beam forming takes place in the "signal preparation" block 42 by appropriate processing of the output signals of individual antenna elements of a receiver antenna 43 connected with the block 42.In the block 42 furthermore a frequency analysis of the reverberation signals received in the individual beams and, in the case of a moving receiver E, the intrinsic Doppler compensation are effected. The sets of data of the intrinsic-Doppler-compensated Doppler reverberation frequencies fscl obtained from the individual beams are fed in association with the reception direction a, and time t, to a computer 44.

The computer 44 on the one hand calculates Doppler reverberation frequencies f, as a function of the reception direction a, and the time t, according to

and 2e-(1+e2)cos a, si=arc cos ------ (16) (1 +)-2e cosa, with R (17) r+ct, and on the other hand by means of the least-mean-square method forms the mean quadratic difference between the supplied Doppler reverberation frequencies fsc, and the calculated Doppler reverberation frequencies fj allocated by reception direction and time.The parameters of the magnitudes of state f,,, v5, k5 and R prestated as imaginary values in the first calculation of the Doppler reverberation frequencies f, are here iteratively amended until the said difference is a minimum. The parameters found for the maximum are issued as the sought magnitudes of state.

It is here advantageous to prestate the start values for the parameters in the estimation method as exactly as possible. For this purpose in a further computer 45 likewise connected with the block 42 the speed vector of the target is determined out of the available data sets of the intrinsic-Doppler-compensated Doppler reverberation frequencies fsci For this purpose at a specific moment t, the Doppler reverberation frequencies fsci are read out of the data sets of all beams or reception directions. Thence the iargest and smallest Doppler reverberation frequencies fma, and f,,, are eliminated.Thus now the computer 45 calculates the magnitude of state of transmission frequency f, according to fm = 2 ffm3X + fMIn) ( 1 8} and the magnitude of state of target speed v5 according to fmax fmlll v5=c. ----- (19).

2f,, Furthermore in the computer 45 the Doppler frequency f0 of the acoustic pulse entering in the direction-finding beam 41, that is the Doppler frequency of the direct signal of the acoustic pulse, is read out and the magnitude of state of target course k5 is calculated according to fD - fmin kS=arc cos -------- (20).

(fmax-fmin) The speed vector v of the target, prestated by target course k5 and target speed v5, and its transmission frequency f,,, are fed to the computer 44 as starting values for the estimation method.

In the computer 45 furthermore it is also possible for a starting value for the parameter of target range R to be calculated. For this purpose the computer 45 reads out of the available data sets in one reception direction a, a Doppler reverberation frequency fsc, and the time moment t, of its arrival, calculated from the arrival of the direct signal in the direction-finding beam 41, and calculates the target range R out of equations (15) and (17) using the magnitudes of state v5, k5, f, ascertained according to equations (18) to (20), which range is then given as starting value to the computer 44.

If more acoustic pulses of the target transmitter are available for evaluation, the magnitude of state of target range R can also be estimated in another manner. By means of the directionfinding beam 41 the bearing to the target S is continuously taken in relation to a reference direction, for example north, and retained as a function of the time. The bearing angle values x, as a function of the time t are fed to a computer 46. This eliminates the intrinsic movement of the receiver E from the measured values and from the compensated measured values determines the bearing angle variations A t in time. Moreover the computer 46 calculates the time bearing angle variation ## #t in radius or degrees according to

The values v5 and k5 are fed to the computer 46 by the computer 45.While the unknown magnitude of state R is prestated as imaginary value. In a least-mean-square estimation method the prestated parameter value is amended iteratively until the mean quadratic difference of the calculated and measured bearing angle difference is a minimum. The pertinent parameter value of the range R is issued as magnitude of state of target range R. The target range R, in the case of a known position of the receiver and known bearing direction 41 is a direct measure for the location of the target.

In the above estimation method the value of the parameter R ascertained from the computer 45 as described above can also be inserted as starting value, so that the requisite calculation expense is considerably reduced. Since now the magnitude of state of target range R has been determined by two separate routes, an error-compensation calculation can also be carried out between the two results to improve the estimation result.

In place of the least-mean-square estimation discussed here it is also possible to use other estimation methods, for example the maximum-likelihood estimation method. In each case the parameter values which fulfil the conditions of the estimation criterion are issued as the sought magnitudes of state of the target S. The estimation method can be carried out either onedimensionally or two-dimensionally. In the former case the Doppler reverberation frequencies f, are calculated as a function of the reception direction a, for a predetermined time moment t and compared with the association Doppler reverberation frequencies fsc, measured at the moment t, as a function of the reception direction a,.In the latter case the Doppler reverberation frequencies f, are calculated as a function of time t, and reception direction a, and compared with the corresponding measured Doppler reverberation frequencies fisc, The invention is not limited to the examples of execution of the method as described. If one is content with the estimation of the magnitudes of state of target speed, target course and transmitter frequency of the target transmitter and confines oneself to the measurement of the target from a stationary or almost stationary receiver, then it is possible to dispense with the electrical expenditure of beam forming for the production of narrowest possible reception beams or a receiver characteristic of high resolution in the azimuth.Of course then the necessity of the intrinsic Doppler compensation of the receiver is also eliminated, since the Doppler reverberation frequencies detected by the stationary receiver correspond directly to the frequencies fsc, radiated by the scatter centres SC,. In this case of undirected reception of the reverberation likewise the extreme Doppler reverberation frequencies f and f"",, and Doppler frequency of the transmitted pulse fD can be ascertained in the reverberation signal and-as stated above-the magnitudes of state of target speed V5, target course k5 and transmission frequency f, can be ascertained therefrom. The magnitude of state of target range R can then admittedly be ascertained only if several transmitted pulses from the target transmitter are available. As described above this is then determined iteratively by means of a suitable estimation method from calculated and measured bearing angle varations in time, the previously calculated magnitudes of state of target speed V5 and target course k5 being inserted as starting values.

Claims

1. Passive method for the estimation of magnitudes of state, such as range, speed, course and/or transmission frequency, of a moving target radiating acoustic pulses into the water, such as a ship, torpedo or the like with active sonar, from a receiver remote from the target, characterised in that Doppler frequencies (fisc) occurring in the reverberation are detected and in that the magnitudes of state (R, v5, k5, f,,) of the target (S) are determined by means of the Doppler reverberation frequencies (fisc,)'

2. Method according to Claim 1, characterised in that the Dopper reverberation frequencies (fsc,) are gained from the frequency spectra of the reverberation.

3. Method according to Claim 1 or 2, characterised in that maximum and minimum Doppler reverberation frequencies (f,,,, f,,,,) occurring in the reverberation are sought out and in that the magnitude of state transmission frequency (f,,) is determined as arithmetic mean of the maximum and minimum Doppler reverberation frequencies (f,,,,, f,,J.

4. Method according to Claim 3, characterised in that the magnitude of state of target speed (v5) is determined as the quotient, multiplied by the speed (c) of sound in water, of the difference of the maximum and minimum Doppler reverberation frequencies (f,,,, f,,,) and twice the transmission frequency (f,,).

5. Method according to Claim 4, characterised in that the Doppler frequency (fad) of the transmitted pulse is eliminated and the magnitude of state of target course (k5) is determined as the arc cosine of the quotient of the difference of Doppler frequency (fed) and transmission frequency (f,,) on the one hand and half the difference of the maximum and minimum Doppler reverberation frequencies (fmax, fmin) on the other.

6. Method according to one of Claims 1 to 5, characterised in that the reverberation is detected direction-selectively and in that the Doppler frequency (fD) of the transmitted pulse is obtained from the direct signal of the acoustic pulse arriving in the target-pointing selective reception direction (bearing direction 41) and the extreme Doppler reverberation frequencies (fmax, f,,,) are obtained from the reverberation signal received in each of two non-target-pointing selective reception directions extending on both sides of the bearing direction (41).

7. Method according to one of Claims 1 to 5, characterised in that the reverberation is detected in a plurality of selective reception directions offset in azimuth in relation to one another each by an angular amount, one of which is target-pointing, in that the Doppler frequency (fed) of the transmitted pulse is gained from the reverberation signal received in the targetpointing selective reception direction (bearing direction 41) and in that the extreme Doppler reverberation frequencies (fmax, fmin) are read out of the Doppler reverberation frequencies (fsc,) which occur in the reverberation signals detected in the non-target-pointing reception directions at a predeterminable time moment (t,) after arrival in the receiver of the direct signal of the acoustic pulse arriving in the bearing direction (41).

8. Method according to Claim 6 or 7, characterised in that with the magnitudes of state of transmission frequency (f,,), target speed (v5) and target course (k5) of the target (S) a selected Doppler reverberation frequency (f1) is calculated according to

wherein .S is arbitrarily assumed, and in that in a non-target-pointing reception direction (a,) the time interval (t,) from the arrival of the direct signal in the bearing direction (41) until the detection of the selected Doppler reverberation frequency (f,) is measured and the magnitude of state of range (R) is calculated from the measured time interval (t1) according to 2H :2) cos a ,5=arc cos (1 +t::2)-2c cos a with R #=- R+c.t1.

9. Method according to one of Claims 1 to 8, characterised in that for at least one predeterminable time moment (t,) after receipt of the direct signal of the acoustic pulse arriving in the bearing direction (41), for a plurality of non-target-pointing selective reception directions ((X1) each offset in azimuth by an angular amount the Doppler reverberation frequencies (f,) are calculated according to

and 2-(1 -e2)cos a, 6=arse cos (1 + E2)2E COS a, with R #=- R+c.t1 the unknown magnitudes of state forming parameters (R, v5, k5 f,,) being prestated as imaginary values, in that in a suitable estimation method, for example least-mean-square estimation, the parameter values are iteratively amended until the estimation criterion is fulfilled, for example the mean quadratic difference between the calculated Doppler reverberation frequencies (f,) and the associated Doppler reverberation frequencies (fsci) detected at this moment in the reception directions (a,) is minimalised, and in that the parameter values fulfiiling the estimation criterion are issued as sought magnitudes of state (R, v5, k5 f,,).

10. Method according to Claim 9, characterised in that the calculation of the Doppler reverberation frequencies (f,) is carried out for a plurality of time moments (tj), calculated from the arrival of the acoustic pulse direct signal arriving in the bearing direction (41) and the estimation method is carried out as two-dimensional estimation method for the calculated Doppler reverberation frequencies (f,=g(t,, a,).

11. Method according to Claim 9 or 10, characterised in that in the prestating of the imaginary values for the parameters the ascertained magnitudes of state (R, v5, k5, f,,) are used at least partially as starting values.

12. Method according to one of Claims 3 to 11, characterised in that the transmitter (S) is passively direction-located continuously from the receiver (E) and after compensation of any own movement of the receiver (E) the time variations of the bearing (eft) are measured, in that a time bearing variation (Aet/At) is calculated according to v sin sin k5 At R using the ascertained magnitudes of state of target speed (v5) and target course (us), the unknown magnitude of state of target range (R) being prestated as imaginary parameter value, in that by means of a suitable estimation method the parameter value is iteratively varied until the estimation criterion is fulfilled, and in that the parameter value which fulfils the estimation criterion is issued as magnitude of state of target range (R).

13. Method according to Claim 1 or 2, characterised in that the Doppler reverberation frequencies (f,,,) in at least one non-target-pointing selective reception direction are determined for a plurality of time moments of a time pattern (ti) running from reverberation interpretation onwards, in that independently thereof for this reception direction (I or II) the Doppler reverberation frequencies (f,) for the same time pattern (t,) are calculated as smoothing curve (f,=h(t,, R, V5, k5, f,,,)), the unknown magnitudes of state (R, v5, k5, f,,) of the target (S) forming parameters which can be prestated as estimated values, in that the estimated values are varied for at least one parameter in each case and for each estimated value a smoothing curve is produced, in that in each case the variance ((to) between each of the smoothing curves and the Doppler reverberation frequency-time values (f5,=g(t)) are calculated and in that those estimated values of the parameters of a smoothing curve for which the variance (tri2) is a minimum are issued as magnitudes of state of the target (S).

14. Method according to Claim 13, characterised in that the reverberation is detected simultaneously over one target-aiming selective reception direction (0) and the Doppler reverberation frequencies (fsc,) are determined for a time pattern (t,) running from the arrival of the transmitted pulse in the receiver (direct signal) onwards, and in that the largest and smallest Doppler reverberation frequencies (f,,x(+ ), f00(-0) are detected and their half sum is stated as transmission frequency (f,,) of the target (S).

15. Method according to Claim 14, characterised in that the radial speed component (Vs,d) of the target (S) is calculated as the product of the difference between the largest or smallest Doppler reverberation frequency (f,,,(+O) or f,0(-0)) and the transmission frequency (f,,) and the quotient of speed of sound (c) in water and transmission frequency (f,,).

16. Method according to Claim 14 or 15, characterised in that from the Doppler reverberation frequency-time values (f5,=g(t)) obtained from the reverberation in the non-target-pointing reception direction (I or II) the maximum and/or minimum Doppler reverberation frequency (f,,,, f,,rn) is ascertained and from this and from the transmission frequency (f,,) the speed (v5) of the target (S) is calculated as the product of the maximum Doppler shift (at,,,0) and the quotient of the speed of sound (c) in water and the transmission frequency (f,,).

17. Method according to Claims 15 and 16, characterised in that the course (k5) of the target (S) related to the target-pointing reception direction (0) is calculated as the arc cosine of the quotient of radial target speed component (v5rad) and target speed (us).

18. Method according to one of Claims 13 to 17, characterised in that the determination of the Doppler reverberation frequencies (fsci) as a function of the time (t,) is carried out in a further, non-target-pointing selective reception direction (II or 1), which is pivoted through a fixedly predetermined direction angle in relation to the first reception direction (I or II), preferably in such a way that the two non-target-pointing reception directions (I, II) lie symmetrically of the targetpointing reception direction (0).

19. Method according to Claim 18, characterised in that the maximum and/or minimum Doppler reverberation frequency (f,,,, f,,) in the further non-target-pointing reception direction (II or I) is determined, in that with the maximum and/or minimum Doppler reverberation frequencies (fax, f,,,) in the two non-target-pointing reception directions (I, II) the maximum Doppler shift f,,,,) is determined and with this the calculation of speed (v5) and course (k5) of the target (S) is carried out.

20. Method according to Claim 19, characterised in that the variance calculation in regard to the Doppler reverberation freluency-time values (f5,=g(t)) is carried out from that of the two non-target-pointing reception directions (I or II) in which the maximum Doppler shift (Af,,,,) is ascertained.

21. Method according to Claim 20, characterised in that on occurrence of several equal maximum Doppler shifts that reception direction (I or II) is selected in which the maximum Doppler shift (at,,,,) pertains to the smallest time value (t,).

22. Method according to one of Claims 13 to 21, characterised in that the calculated values for transmission frequency (cm), course (k5) and speed (v5) of the target (S) are used in the calculation of the smoothing curves (f,=h (t,, R, v,, f,,)) as estimated values and only the range (R) forms a parameter.

23. Method according to Claim 22, characterised in that the calculation of the smoothing curves with the obtained magnitude of state of target range (R) as estimated value and with at least one of the other magnitudes of state (k5, v,, f,,) as parameter is repeated with varied estimated value, in that in the same manner the variance calculation and the determination of the variance minimum are carried out, and in that with the then calculated value of the magnitude of state (k5, V5, f",) of the target (S) the above method steps are repeated with at least one further magnitude of state (k5, v5 f,,) as parameter until the variation of the magnitude of state (k5, v,, f,,) issued in each case does not exceed a predetermined value.

24. Method according to one of Claims 14 to 23, characterised in that the zero point of the time pattern in the non-target-pointing reception directions (I, II) is fixed by the time moment of arrival of the direct signal of the acoustic pulse arriving in the target-pointing reception direction (O).

25. Method according to Claim 24, characterised in that on occurrence of a time stagger between the reverberation detections in the two non-target-pointing reception directions, a target (S) with rotating transmitted beam is deduced and the time stagger is determined as rotation period (TUM) of the transmitted beam.

26. Method according to Claim 24 or 25, characterised in that on occurrence of a time stagger of the reverberation interpretation in the target-pointing reception direction (0) in comparison with the zero point of the time pattern a target (S) with rotating transmitted beam is deduced and the doubled time stagger is determined as rotation period (TUM) of the transmitted beam.

27. Method according to Claim 25 or 26, characterised in that the variance calculation in relation to the reverberation frequency-time values (f5,=g(t)) is carried out from that of the two non-target-pointing reception directions (I or II) in which no ambiguous Doppler reverberation frequencies (fscj occur.

27. Method according to Claim 25 or 26, characterised in that the variance calculation in relation to the Doppler reverberation frequency-time values (f5,=g(t)) is carried out from that of the two non-target-pointing reception directions (I or II) in which no ambiguous Doppler reverberation frequencies (fsc,) occur.

CLAIMS Amendments to the claims have been filed, and have the following effect: Claims 1-27 above have been deleted or textually amended.

New or textually amended claims have been filed as follows:-

1. A passive method of estimating one or more of the following characteristics, range, speed, course and transmission frequency of a moving target radiating acoustic pulses into water, such as a ship, torpedo or the like equipped with active sonar, by means of a receiver remote from the target, characterised by the steps of receiving reverberation signals being produced by the acoustic pulses as a result of distribution of the acoustic pulses through impedance discontinuities in the water, detecting reverberation frequencies occurring in the reverberation signals which frequencies are Doppler shifted compared to the transmission frequency corresponding to the position of each of the impedance discontinuities relative to the position of the target and - using selected reverberation frequencies to estimate the desired characteristic or characteristics of the target.

2. Method according to Claim 1, characterised in that the reverberation frequencies (fsci) are gained from the frequency spectra of the reverberation signals.

3. Method according to Claim 1 or 2, characterised in that maximum and minimum reverberation frequencies (f,,,,, f,,) occurring in the reverberation signals are sought out and in that the characteristics transmission frequency (f,,) is determined as arithmetic mean of the maximum and minimum reverberation frequencies (f,,,, f,s).

4. Method according to Claim 3, characterised in that the characteristics target speed (v5) is determined as the quotient, multiplied by the speed (c) of sound in water, of the difference of the maximum and minimum reverberation frequencies (f,,,, f,,,) and twice the transmission frequency (f,,).

5. Method according to Claim 4, characterised in that the Doppler frequency (fed) of the transmitted pulse is eliminated and the characteristics course (k5) is determined as the arc cosine of the quotient of the difference of Doppler frequency (fed) and transmission frequency (f,,) on the one hand and half the difference of the maximum and minimum reverberation frequencies (f,,,,, f,,,,) on the other.

6. Method according to one of Claims 1 to 5, characterised in that the reverberation signals are detected direction-selectively and in that the Doppler frequency (fD) of the transmitted pulse is obtained from the direct signal of the acoustic pulse arriving in the target-pointing selective reception direction (bearing direction 41) and the extreme reverberation frequencies (f,,,, fain) are obtained from the reverberation signal received in each of two non-target-pointing selective reception directions extending on both sides of the bearing direction (41).

7. Method according to one of Claims 1 to 5, characterised in that the reverberation signals are detected in a plurality of selective reception directions offset in azimuth in relation to one another each by an angular amount, one of which is target-pointing, in that the Doppler frequency (fed) of the transmitted pulse is gained from the reverberation signal received in the targetpointing selective reception direction (bearing direction 41) and in that the extreme reverberation frequencies (f,,j, f"l",) are read out of the reverberation frequencies ffsci) which occur in the reverberation signals detected in the non-target-pointing reception directions at a predeterminable time moment (t,) after arrival in the receiver of the direct signal of the acoustic pulse arriving in the bearing direction (41).

8. Method according to Claim 6 or 7, characterised in that with the transmission frequency (f",), target speed (v5) and target course (k5) of the target (S) a selected reverberation frequency (fj) is calculated according to

wherein 5 is arbitrarily assumed, and in that in a non-target-pointing reception direction (a,) the time interval (tl) from the arrival of the direct signal in the bearing direction (41) until the detection of the selected reverberation frequency (fl) is measured and the magnitude of state of range (R) is calculated from the measured time interval (t) according to 2c--(1 - C2) cos a rS=arc cos (1 - C2) - 2C COS n, with R #=- R+c.t1.

9. Method according to one of Claims 1 to 8, characterised in that for at least one predeterminable time moment (tl) after receipt of the direct signal of the acoustic pulse arriving in the bearing direction (41), for a plurality of non-target-pointing selective reception directions (a,) each offset in azimuth by an angular amount the reverberation frequencies (f,) are calculated according to

and 2e-(1 -e2) cos a, 8=arse cos (1 - E2)- 2E cos a with R #=-- R+c.t1 the unknown characteristics forming parameters (R, v,, k5, f,,) being prestated as imaginary values, in that in a suitable estimation method, for example least-mean-square estimation, the parameter values are iteratively amended until the estimation criterion is fulfilled, for example the mean quadratic difference between the calculated reverberation frequencies (f,) and the associated reverberation frequencies (fsc,) detected at this moment in the reception directions (a,) is minimalised, and in that the parameter values fulfilling the estimation criterion are issued as sought magnitudes of state (R, v,, k5, f,,).

10. Method according to Claim 9, characterised in that the calculation of the reverberation frequencies (f,) is carried out for a plurality of time moments (t,), calculated from the arrival of the acoustic pulse direct signal arriving in the bearing direction (41) and the estimation method is carried out as two-dimensional estimation method for the calculated reverberation frequencies (fl=g(tl, a,).

11. Method according to Claim 9 or 10, characterised in that in the prestating of the imaginary values for the parameters the ascertained characteristics (R, v5 k5, f,,) are used at least partially as starting values.

12. Method according to one of Claims 3 to 11, characterised in that the transmitter (S) is passively direction-located continuously from the receiver (E) and after compensation of any own movement of the receiver (E) the time Variations of the bearings (Ç) are measured, in that a time bearing variation (h t) is calculated according to Ai, V5 sin sin k5 At R using the ascertained magnitudes of state of target speed (v5) and target course (us), the unknown magnitude of state of target range (R) being prestated as imaginary parameter value, in that by means of a suitable estimation method the parameter value is iteratively varied until the estimation criterion is fulfilled, and in that the parameter value which fulfils the estimation criterion is issued as magnitude of state of target range (R).

13. Method according to Claim 1 or 2, characterised in that the reverberation frequencies (fsc,) in at least one non-target-pointing selective reception direction are determined for a plurality of time moments of a time pattern (t,) running from reverberation interpretation onwards, in that independently thereof for this reception direction (I or II) the reverberation frequencies (f,) for the same pattern (t,) are calculated as smoothing curve (f1=h (t1, R, vs, ks, fm)), the unknown characteristics (R, v5, k5, f,,) of the target (S) forming parameters which can be prestated as estimated values, in that the estimated values are varied for at least one parameter in each case and for each estimated value a smoothing curve is produced, in that in each case the variance (52) between each of the smoothing curves and the reverberation frequency-time values (f5,=g(t)) are calculated and in that those estimated values of the parameters of a smoothing curve for which the variance ((r2) is a minimum are issued as magnitudes of state of the target (S).

14. Method according to Claim 13, characterised in that the reverberation is detected simultaneously over one target-aiming selective reception direction (0) and the reverberation frequencies (fsc,) are determined for a time pattern (t,) running from the arrival of the transmitted pulse in the receiver (direct signal) onwards, and in that the largest and smallest reverberation frequencies (f0 (+O), f,,0(-0) are detected and their half sum is stated as transmissjion frequency (f,,) of the target (S).

15. Method according to Claim 14, characterised in that the radial speed component (vsrad) of the target (S) is calculated as the product of the difference between the largest or smallest reverberation frequency (fex(+ ) or f,,0(-0)) and the transmission frequency (f,,) and the quotient of speed of sound (c) in water and transmission frequency (f,,).

16. Method according to Claim 14 or 15, characterised in that from the reverberation frequency-time values (f5,=g(t)) obtained from the reverberation in the non-target-pointing reception direction (I or II) the maximum and/or minimum reverberation frequency (fmax, fmin) is ascertained and from this and from the transmission frequency (f,,) the speed (v5) of the target (S) is calculated as the product of the maximum Doppler shift (Af,,,,) and the quotient of the speed of sound (c) in water and the transmission frequency (f,,).

17. Method according to Claims 15 and 16, characterised in that the course (k5) of the target (S) related to the target-pointing reception direction (0) is calculated as the arc cosine of the quotient of radial target speed component (vsrad) and target speed (us).

18. Method according to one of Claims 13 to 17, characterised in that the determinatjion of the reverberation frequencies (foci) as a function of the time (t,) is carried out in a further, nontarget-pointing selective reception direction (II or 1), which is pivoted through a fixedly predetermined direction angle in relation to the first reception direction (I or II), preferably in such a way that the two non-target-pointing reception directions (I, II) lie symmetrically of the target-pointing reception direction (0).

19. Method according to Claim 18, characterised in that the maximum and/or minimum reverberation frequency (f,,,, f,,,) in the further non-target-pointing reception direction (II or I) is determined, in that with the maximum and/or minimum reverberation frequencies (f,,,,, f,,,,) in the two non-target-pointing reception directions (I, II) the maximum Doppler shift (at,,,0) is determined and with this the calculation of speed (v5) and course (k5) of the target (S) is carried out.

20. Method according to Claim 19, characterised in that the variance calculation in regard to the reverberation frequency-time values (f,,,=g(t)) is carried out from that of the two non-targetpointing reception directions (I or II) in which the maximum Doppler shift (at,,,0) is ascertained.

21. Method according to Claim 20, characterised in that on occurrence of several equal maximum Doppler shifts that reception direction (I or II) is selected in which the maximum Doppler shift (at,,,0) pertains to the smallest time value (t,).

22. Method according to one of Claims 13 to 21, characterised in that the calculated values for transmission frequency (cm), course (k5) and speed (v5) of the target (S) are used in the calculation of the smoothing curves (f,=h (t,, R, v,, k5, f,,)) as estimated values and only the range (R) forms a parameter.

23. Method according to Claim 22, characterised in that the calculation of the smoothing curves with the obtained magnitude of state of target range (R) as estimated value and with at least one of the other magnitudes of state (k5, v,, f,n) as parameter is repeated with varied estimated value, in that in the same manner the variance calculation and the determination of the variance minimum are carried out, and in that with the then calculated value of the magnitude of state (k5, V5, f,,,) of the target (S) the above method steps are repeated with at least one further magnitude of state (k5, v,, f,,,) as parameter until the variation of the magnitude of state (k5, v,, f,,,) issued in each case does not exceed a predetermined value.

25. Method according to Claim 24, characterised in that on occurrence of a time stagger between the reverberation detections in the two non-target pointing reception directions, a target (S) with rotating transmitted beam is deduced and the time stagger is determined as rotation period (TUM) of the transmitted beam.