DE3639500A1 - Radar receiver for mobile radar apparatuses having an antenna array with clutter suppression which acts in a two-dimensional manner - Google Patents

Abstract

The invention describes a method for two-dimensional suppression of clutter signals of all types in the case of a radar receiver for a coherent radar which moves on a platform and has an active antenna array for detecting moving targets with an adaptive processor, in the case of which the favourable properties of the optimal processor are maintained but (as a result of a pretransformation operation using characteristic vectors of the characteristic values of the clutter covariance matrix as auxiliary channels for interference suppression) the complexity of computation operations and the computation time are reduced to such an extent that practical implementation becomes possible. The invention can be used in aircraft on-board radars. <IMAGE>

Description

The invention relates to a method for the two-dimensional, ie in space and time effective suppression of clutter signals of all kinds, in particular ground clutter in echo signals for locating moving target objects by scanning the backscattered echo signal field generated by transmitting M transmission pulses with a radar receiver for a moving coherent radar device with a, in particular phase-controlled, group antenna with N sensors, in which the echo signals received by the sensors are amplified, mixed down, digitized, ordered in reception channels into complex input signals, and then a target signal is obtained from the input signals by signal-adapted filtering and suppression of clutter. The invention further relates to the design of a radar receiver for performing the method according to the preamble of claim 1.

Finding moving targets near the ground, such as low-flying aircraft, Vehicles can by an elevated point of view, for. Legs flying platform with which the radar device is moved, e.g. B. the on-board radar of an aircraft can be significantly increased, as influences such as shadows in undulating terrain and on The radar lobe pitching near the ground can be reduced. The searched target signal of the moving target must however from the entire echo, including all interference signals (clutter) field by suppressing clutter of all kinds by means of a Clutter filter can be obtained.

Regarding the clutter echoes arise with z. B. flying be radar devices have some disadvantages, the following of which both of particular importance for the development of one Clutter filters are:

• a) The different clutter elements are located under different azimuth and elevation angles relative to the flight axis of the radar. The radial component of the relative speed between the clutter element and the flight axis of the radar is therefore proportional to the cosine of the angle between the location vector of the clutter element (seen from the radar) and the flight axis. Since the relative radial speed between the clutter element and the radar determines the Doppler frequency of the radar echoes, such moving clutter echoes, in particular from a homogeneous background, are Doppler-colored, i.e. the received clutter signals have all possible Doppler frequencies between v _max (airspeed) and - v _max . In horizontal flight, curves of constant Doppler frequency form a family of hyperbolas that open in the direction of flight (or against). If one walks along a distance ring, one runs through the entire clutter Doppler spectrum from v _max to - v _max .
• b) When flying over different terrain formations the radar device counts zones with different rear spreading behavior, d. H. the clutter background is temporal and spatially variable. It follows that the clutter suppression system should be adaptive to interference, d. H. it must have the ability to cope with changing disturbance conditions adapt.

From a) it follows that clutter signals of several parameters, namely azimuth, elevation, distance and Doppler frequency hang. However, there is a fixed link between the location coordinates of each clutter element and its Doppler frequency.

Such a range azimuth Doppler network is shown in FIG . With ϕ the azimuth angle, with v the clutter radial speed, with r / h the distance / flight altitude ratio and with HNr the number of the respective hyperbola.

If you restrict yourself to a range ring, you can see that the clutter power is a function of two parameters: azimuth ϕ , Doppler frequency fD and clutter radial speed v , and can be represented for a radar target range by a two-dimensional power spectrum. In FIG. 2 this is an example of a Bewegtclutterspektrums for the viewing direction of the transmission lobe φ _B = 0 °, ie in the direction of flight, provided. There is a maximum main lobe clutter at ϕ = 0 ° and the associated clutter frequency v = v _max . Any flight destinations to be discovered are in the direction of the beam ϕ = 0 ° and at all possible speeds - v _max . . . + v _max , that is, along the upper left edge in FIG. 2. Along the diagonals, one sees the side lobe clutter distributed over all Doppler speeds. CNR means the clutter-to-noise ratio, M the number of echo pulses, N the number of sensors.

So if you consider a certain distance ring, there remains a dependency of the clutter performance on two parameters: azimuth ϕ and Doppler frequency f _D (or relative speed v) . A filter to suppress such interference signals must therefore be effective in two dimensions, ie in ϕ and v (or in space and time).

It follows that the re-controlled echo field in two dimensions sions, and that must be recorded spatially and temporally. These requirements are met by a coherent pulse Doppler radar with phased array antenna, the Sensors as spatial scanning and the echo pulse (coherent Echo sequences) can be viewed as a time sample. As Clutter model is converted from the clutter echo samples to a spatial Time-covariance matrix formed.

Most of the previously known methods of suppression in moving radars are based on a modification of the known one 2-pulse clutter cancellers.

On the one hand, there are processes that use the "clutter" method locking ". With these methods the maximum of the Clutter spectrum in the restricted area of the 2-pulse canceller pushed so that a certain obliteration of the main clubs clutters is done. This principle is used in the TACCAR System found (M.I. Skolnik "Radar Handbook" McGraw Hill, 1970) pp. 17-32 /. A disadvantage of TACCAR and related processes is that there is no motion compensation, so that clutter spectrum covered by the main lobe is relatively high Doppler bandwidth has only that of the 2-Plus Canceller can be suppressed wisely. Nebenzipfelclutter has from Main lobe clutter different Doppler frequencies and is about not suppressed at all.

On the other hand, there is the so-called DPCA technology (displaced phase center antenna). Here the movement of the platform of the  Radars compensated by the fact that between two pulses of a reception array to a second one, in the direction of flight behind it attached array is switched so that the platform speed between two pulses is just canceled. The method has the disadvantage that with a comparable up used around 6 dB of transmission energy for clutter suppression get lost.

A preliminary stage for adaptive 2-dimensional clutter suppression for moving radars is in Andrews "Radar Antenna Pattern Design for Platform Motion Compensation "IEEE Trans, AP, Vol. Ap-26, No. 4, July 1978, pp. 566-571. Here the Ge Weights of two evaluation sets of a sensor array such opti that the difference between their clutter responses becomes minimal. Adaptive implementation is not covered. All of these known method is common that only two on top of each other following echoes are used for clutter suppression.

In the publication by Brennan, L.E., Mallett, I.D., Reed, I.S .: "Adaptive Array in Airborne MTI Radar", IEEE AP-24, No 5, Sept 1976, pages 607-615 it is noted that the moving clutter problem (like other types of interference sub pressure) in principle by the optimal detector (likelihood Ratio test = LR test) can be solved.

The optimal detector is known for the likelihood Ratio test (LR test probability test) given:

With

x vector of the received signals R clutter covariance matrix of the spatio-temporal samples s vector of the target signals n vector of clutter and noise values

given.  

If the print on the left exceeds a threshold, the decision "target available" applies, otherwise not. Equation (1) is a very general rule that can be used for any decision problem. In the case of the discovery of flight destinations in moving clutter, R has the dimension ( MN) × (MN) , the vectors x , n and s the dimension MN .

As is well known, the optimal receiver consists of a two-stage filter, a noise suppression component and a target-adapted filter (pre-whiten and match). The first part consists of the inverse covariance matrix of the clutter signals. In the case of 2-dimensional signal sampling, the order of such a matrix can assume dimensions that are no longer suitable for real-time use. An example: A small antenna has N = 32 sensors, the coherent signal processing works with M = 32 coherent pulses. The order of the matrix is N × M = 2¹⁰ = 1024. The matrix contains (N · M) ² = 1,048,576 elements, for their inversion you need (N · M) ³ = 2³⁰≈10⁹ arithmetic operations. These dimensions cannot be mastered by today's computing capacity and will hardly be in the future.

Basically, the 2-dimensional signal ver work no particular problems; the limitation of profit is due to the receiver noise (as in 1-dimensional Case). To realize the optimal detector too driving effort is so high that he is currently with the available devices neither in terms of costs nor in terms of computing time to cope with.

The object of the invention is to provide an adaptive receiver for a coherent radar with an active array antenna create with a way of working that has the favorable properties of the optimal detector, but in the one to be operated Computational effort is reduced so far that it can be realized can.

The invention solves the problem with that by Features of claim 1 characterized method and with  a radar receiver according to the characterizing features of the Claim 11.

The invention works on the principle of match and post-whitening for two-dimensional clutter suppression, one or more search channels being formed by lobe formation and Doppler filtering of the received signals, which are still subject to all interference components in terms of direction and amount, followed by clutter suppression. Instead of carrying out the known probability ratio test according to equation (1) at full rank N · M , the LR test is used in the transformed area according to the invention. For this purpose, the received signals are subjected to a linear transformation using the eigenvectors of the clutter covariance, so that the resulting vector space (or: the number of signal values to be evaluated per unit time) is considerably reduced. As a result, the outlay on subsequent signal processing, interference suppression, spatial-temporal signal-adapted filtering is reduced to an economically feasible level. A further development of the method according to the invention can be gathered from the characterizing features of claim 5.

The invention thus enables the creation and operation of a clutter-adaptive filter to suppress clutter of all kinds, especially ground clutter, in radar devices, which are on platforms moving relative to the earth Find. Such a clutter filter enables the discovery of moving targets such as planes and vehicles aboard one flying radar platform provided the targets are in their Doppler frequency from that of the disturbing background differentiate.

Theoretical preliminary investigations in the publications by Klemm, R .: "Suboptimum Clutter Suppression for Airborne Phased Array Radars", PROC RADAR 82, London, UK, October 1982 and Klemm, R .: "Adaptive Clutter Suppression for Airborne Phased Array Radars", IEE Proc. Pts. F and H, Vol. 130 No. 1, February 1983, pages 125-132 have revealed an important property of the covariance matrix of the spatial and temporal samples of the clutter echoes. If the number of sensors is N and the number of echo values is M , one obtains an (N × M) (N × M) cova riance matrix; however, the number of cluster inherent values is a maximum of N + M. From this it follows on the one hand that the number of eigenvalues, especially for large values of N and M, is small compared to the number of samples or the order of the covariance matrix. It also follows that the minimum number of clutter intrinsic values is reached for a given total number of samples for N = M , ie if the number of echo pulses is equal to the number of sensors. This finding is implemented according to the invention by using the eigenvectors of the space-time covariance matrix of the clutter sample values associated with the clutter eigenvalues as auxiliary channels for interference suppression. It is thus possible to derive a processor on the basis of the optimal detector in order to find moving targets in front of a moving clutter background in such a way that the favorable properties of the optimal processor are retained, but due to the suitable transformation of the received signals, the effort in arithmetic operations and computing time of the subsequent signal processing is retained is reduced so far that a practical implementation is possible.

To carry out the invention, the covariance matrix R of the spatio-temporal samples must be determined in principle, but for this purpose, in a simplification according to the invention, using the eigenvectors of the covariance matrix, a linear transformation of the received data is carried out, with which the received signals, ie clutter suppression, are then evaluated , is made.

The invention is carried out in such a way that instead of carrying out the LR test according to (1) at full rank N · M , only a transformed reduced test is carried out according to equation (2):

The transformed quantities are defined as follows:

x _T = T * x (2a)
s _T = T * s
n _T = T * n
R _T = T * R T

Since the target signal s and thus also s _{T is} initially unknown, the control vector b ( ϕ , v ) takes its place:

Since the matrix I anticipates the signal integration (lobe formation, Doppler filtering) and the sidelong contributions to the target signal can be neglected compared to the main lobe contribution, the following is:

with e as the unit vector, where R _T ^-1 e represents the first column of R _T ^-1 and (2) simplifies to

T is an NM × L matrix (L N + M) . It is chosen according to the invention so that

• a) no signal energy is lost
• b) the clutter echoes with high CNR are received
• c) L « MN , so that the desired saving in computing time, arithmetic operations and hardware expenditure is achieved
• d) L is not less than the number of clutter eigenvalues (N + M) .

If these conditions are met, the same behavior can occur of (3) as expected from the optimal detector in equation (1) will.

The structure of the transformation matrix represents the core of the invention described here. In order to satisfy the conditions set out, the matrix I , which is a rectangular matrix with N · M columns and L rows, must consist of two parts:

• 1) 1 main channel, consisting of lobe formation (spatial) and Doppler filter (temporally, a Doppler to be suppressed channel) for signal adaptation  
• 2) L - 1 auxiliary channels from spatial + temporal samples, which serve to measure the interference and span the clutter subspace of R as well as possible so that the clutter echoes are received with the highest possible CNR (clutter / noise ratio). For this, according to the invention, the eigenvectors of a clutter covariance matrix that are intrinsic to the clutter are used.

The signal adaptation according to point 1) is a target signal adjusted weighting taken into account, d. H. the elements of a Columns of the matrix must have the spatial-temporal samples correspond to an expected target signal. The signal adaptation is a vector with that of a particular target signal expected partial signals

s _{n, m} ( ϕ , ν ) ∼ exp j β _n exp j ω _D t _m (4)

in which

n = sensor index m = pulse index τ _m = delay of the m th echo ω _D = Doppler frequency due to the relative speed between the radar and the target b _n = phase related to the phase center at the n th sensor due to the direction of incidence

For a regular rectangular sensor arrangement parallel to the Earth will

where x _n , y _{n represent} the sensor coordinates and ⊖ the elevation.

The first column of the transformation T thus looks as follows:

in which

is clubbing and the coefficients

c _m ( v ₀) = exp j ω _D (v ₀) τ _m (8)

represent the coefficients of a Doppler filter. Thus poses the dot product

y = x * t ₁

a club formation with a Doppler filter in cascade. Such a "search channel" is adapted to a target in a specific direction ϕ ₀ and with a specific radial speed, which depends on the target speed, the platform speed and the direction of flight of the target.

The remaining L -1 columns of T are used to estimate the clutter interference and should be designed so that they are received with the highest possible clutter / noise ratio.

The assignment of the L -1 clutter vectors in T plays a central role for the invention. First, a preview is made.

Received clutter signals are statistically described by a space-time covariance matrix R , as previously introduced. As is known, this matrix can also be represented in the following way:

where λ _{i is} the eigenvalues, e _{i are} the eigenvectors of R. If you rank the eigenvalues according to size, you get a series of multiple eigenvalues that represent the channel noise and, as a rule, N + M eigenvalues that belong to the clutter. An example of this can be seen in FIG. 3 for a linear antenna with N _x sensors and M transmission pulses. It can be seen as a law that neglecting clutter fluctuations means that the number of intrinsic clutter values is less than or equal to the sum of the number of sensors and echo pulses: L N + M. The minimum number results for N = M , ie L 2 N. The prerequisite here is that the pulse repetition frequency is not less than the Nyquist frequency of the clutter echoes.

Since the clutter intrinsic values usually stand out clearly from the noise, R can be divided into a clutter and a noise component:

The two matrix components are unitary to one another, so that the left component represents the vector subspace of the clutter, the right component that of noise. The invention is based on the fact that the remaining L -1 columns of T are filled with the L -1 eigenvectors e _i c , which belong to the clutter eigenvalues, ie the L -1 largest eigenvalues. The auxiliary channels formed in this way just span the clutter vector space, so that in principle all clutter is detected, but the matrix T brings about a considerable reduction in the order of the subsequent signal processing. With this transformation matrix T and equations (2a) and (3), the principle of the adaptive processor for suppressing moving clutter is given. Now, however, it is still necessary to carry out the inversion of the matrices R _T (m) which cause clutter suppression and are not known. It should be continuously adapted to the constantly changing clutter conditions.

Usually you ask all possible target Doppler frequencies when the direction is set at the same time, so that for the calculation rule (3)

stands, where m denotes a certain target Doppler frequency. This also means that M form matrices

T * _m R T _m (m = 1 ... M)

have to be inverted. T _m is the transformation that contains the m th Doppler filter.

The inversion of the matrices R _T (m) can be carried out very effectively if one takes into account that all L × L matrices R _T (m) have a common (L -1) × (L -1) submatrix, namely the covariance matrix of the auxiliary signals. So the matrices have the shape

With

a Clutter performance in the search signal b vector of the cross-correlation values between search and auxiliary signals D covariance matrix of the auxiliary signals

Only a and b also depend on the frequency m .

We assume that D ^{-1 is} known. Then the inverse matrix can be given as follows:

in which

Δ = D - 1 / a b b *. If D is known, then Δ ^{-1 is obtained} as follows

The algorithm therefore comprises the following steps:

a) Estimate D , a and b for all target Doppler frequencies m = 1. . . M
b) Calculate D ^-1
c) Calculate Δ _m ^-1 for all m = 1. . . M (Eq. 13).
d) Calculate R ^-1 for all m = 1. . . M (Eq. 12).

Step b requires (L -1) ³ multiplications
Step c requires approximately L ² M multiplications
Step d requires approximately L ² M multiplications.

Overall, there are somewhat over (L -1) ³ + 2 L ² M multiplications.
However, this effort can again be considerably reduced by applying the invention - eigenvectors of a clutter covariance matrix as auxiliary channels for interference suppression. The use of clutter eigenvectors in the matrix R now has the advantage that each one of the eigenvectors comes in with the weight given by the associated eigenvalue, in other words, there are "important" and "less important" clutter eigenvectors. The spectrum of the clutter in turn depends on the nature of the reflective background. With increasing directivity, the number of large eigenvalues decreases. One can therefore come to a less complex (computing time, hardware effort) method by saving some of the eigenvectors belonging to smaller eigenvalues. Inverting the matrices R _T (m) costs a few N ³ operations, so that, for. B. halving the dimension (L / 2 × L / 2 instead of L × L) saves computing time by a factor of 8. He is used according to the invention. In the case of a radar receiver which works with auxiliary channels with eigenvectors, a matrix of the L -1 first, ie the eigenvalues belonging to the largest eigenvalues, is used, as can be seen from the characterizing features of claims 5 and 6. It has been found that a reduction of the auxiliary channels in the assignment with eigenvectors, ie the reduction in the number of degrees of freedom L compared to N + M by up to half or up to a third, still leads to very good results, ie not yet noticeable losses in profit to lead. This is due to the fact that in the Jacobi algorithm used for the eigenvalue decomposition the eigenvalues are sorted by size and when the vector space spanned by the clutter eigenvectors is clipped, the small eigenvalues are automatically affected. In FIG. 4, the dependence of the gain of SCR (signal / Clutterverhältnis) of the number of Clutterfilterfreiheitsgrade L at constant N and M represents Darge.

The gain in signal / clutter ratio is plotted above the target speed normalized to the unique speed range. At the point of the clutter adjustment, the associated clutter element is interpreted as the target (here at v / v _max = 0.4). The clutter spectrum represents a projection of a spatial clutter distribution in the Doppler frequency range. This distribution was measured with the ELRA system and used here to obtain the most realistic conditions possible. As you can see, a reduction in the number of degrees of freedom from L = N + M = 24 to 12 does not result in a loss of profit.

In the following tables 1 to 3 an effort comparison of the required arithmetic operations between optimal detector method, auxiliary channel method with L = N + M degrees of freedom and half-dimension L / 2 reduced auxiliary channel method with eigenvectors is made. The auxiliary channel methods are referred to as sub-optimal methods.

The effort for club formation and Doppler filtering is approximately the same in all cases. In the suboptimal case, these operations are carried out before the clutter suppression, in the optimal case afterwards. The cost of estimating R _t is much lower than for the large matrix R ; this point is ignored. The effort for filter calculation, ie the required complex multiplications for the auxiliary channel methods and the optimal method are compared. It is assumed that N = M and thus L = 2 N. Then the numbers of multiplications for different values of N listed in Table 1 result for the optimal method. For the method according to the invention with L -1 auxiliary channels from the eigenvectors, the significantly smaller number of arithmetic operations already listed in Table 2 already result. This reduced number of arithmetic operations allows the two-dimensional clutter suppression to be realized in the first place. A further significant reduction in arithmetic operations by the factor is possible according to the invention by operating the auxiliary channel method with a reduction in the degrees of freedom by half, as can be seen from the analog arithmetic example in Table 3. If you need matrix inversion for L = 32 62 559 arithmetic operations with the simple auxiliary channel method, the effort is reduced by half the degrees of freedom to 7 471 arithmetic operations.

Table 1

optimal detector

Table 2

suboptimal detector with L -1 auxiliary channels at L = M + N

Table 3

suboptimal detector with L -1 auxiliary channels for eigenvectors and reduction of degrees of freedom

The eigenvector method in principle requires that the eigenvector decomposition of R would have to be carried out in real time to determine the auxiliary channels in the matrix T. However, this again makes the process more complex. According to the invention it is therefore proposed that the eigenvector decomposition of a matrix is carried out off-line and the eigenvectors thus obtained are introduced into the signal processor as auxiliary channels. This model matrix can e.g. B. obtained by longer averaging in a specific area extended in azimuth and elevation and updated from time to time. Another possibility proposed according to the invention is to determine such a model matrix by calculation. According to the invention, the case of an omni-directional background is provided here, which contains all possible azimuth and Doppler frequency components in the same way and thus covers all possible inhomogeneous cases.

The matrices R _T (m) of equation (3) can be estimated by averaging over a clutter area extended in azimuth and distance. A processor according to (3) is therefore adaptive to interference. Corresponding embodiments of the invention can be found in the features of claims 5, 7 to 9 and 11.

An even simpler non-adaptive receiver is obtained according to the invention by using the matrix for omnidirectional clutter transformed according to (2) for R _T (m) in equation (3). The method is somewhat less powerful than the adaptive method described above, since even in the case of a highly directive clutter situation, the entire clutter range is always masked out in azimuth and Doppler frequency. Furthermore, unpredictable influences such as sensor coupling and multipath propagation are not taken into account, as is the case with an adaptive system. Nonetheless, the non-adaptive method is of practical interest because of its simplicity.

An example of the same clutter distribution as in FIG. 4 can be seen in FIG. 5. As you can see, there are optimal profit curves for L = 12 and even L = 18, which are even cheaper than for L = 24. This can be explained by the fact that the target signal vector space is already clipped by the clutter filter due to an excessive number of clutter channels. A covariance matrix for omnidirectional clutter was used as a model for calculating the eigenvectors.

Fig. 6 shows the comparison with conventional methods, ie antenna lobe formation and temporal clutter filter in series. Large losses can be seen over almost half the Doppler frequency range.

Insofar as measurement curves are shown in the explanatory FIGS . 1-6, the data and parameters on which the measurements are based are dimensioned as follows:

CNR Clutter-noise ratio at each Sensor20 dB f ₀ Sendefrequenz3 GHz v ₀ Radarplattformgeschwindigkeit200 km / h T sampling interval (transmission interval) 1/2 f _max (maximum Doppler frequency)

Type of receiving antenna linear group in flight direction N number of sensors 16 M number of transmission pulses 8 L order of the reduced system 24 h flight altitude 500 m r distance 1000 m d sensor distance λ / 2

Type of transmit antenna linear in flight direction A _s transmit aperture (N -1) d ϕ B viewing angle 60 ° B _r relative clutter bandwidth 0

The invention is described below using exemplary embodiments explained in more detail.

Fig. 7 shows an overview of the two methods for Bewegtclutterunterdrückung which have been explained above.

Fig. 7a shows the structure as it is derived directly from the LR test is known (pre-whiten and match), and as the optimum detector. The received data are freed from the clutter in the space-time range (x, y, t) . Then a search channel is formed (lobe in direction ϕ _s , Doppler filter for speed v _s ).

Fig. 7b shows a block diagram of the auxiliary channel method (match and post-whiten). The received data are fed to a search channel ( d _s , v _s ) and several clutter channels (possibly several sets ϕ _c , v _c ). The search signal is then released from the clutter.

The principle of the new radar receiver can be seen in FIG. 8. The output signals of the N sensors pass through the usual coherent reception channels (amplifier, mixer, AD converter) and are then stored in the echo memory according to the sensor, distance and delay. By forming a weighted sum over the complex input signals (b ₁, b ₂... B _N ,), a search lobe with a lobe-forming network for all echo delays (and distance elements) is formed. The echo values of the search lobe are transformed into the Doppler frequency range using a DFB filter bank. The Doppler filter bank (weighted sums over the echo delays) carries out a spectral analysis of the received signals, ie the M values of the search lobe are examined together according to the individual Doppler frequencies, ie speeds. Together with the search lobe, the outputs of the Doppler filter bank represent M search channels. Lobe formation and Doppler filter bank are known devices and processes in radar technology.

The output signals of the receive channels are also fed to the auxiliary channels from a matrix of L -1 eigenvectors belonging to the largest eigenvalues. These each have the dimension NM and therefore act both in the sensor and in the echo pulse dimension. The L -1 outputs of the auxiliary channels are linked to one output of the Doppler filter bank and together form the reduced signal vector space in which the subsequent signal processing and evaluation takes place in an L × M matrix. The signal vector x is multiplied by the first column of R _T ^-1 (m) (clutter suppression). This is done separately for each of the M frequencies. The largest of the M signals of the remaining frequency channels is selected and brought to the threshold decision and displayed.

In particular, covariance matrices R _T (m) of dimension L × L between each output of DFB and all auxiliary channels are estimated (covariance estimate KS) . The (L -1) × (L -1) covariance sub-matrix of the auxiliary channels only needs to be estimated once; further M vectors of dimension L contain the cross-correlation values between the M search channels and the L -1 auxiliary channels and the performance of the search channels. The estimate must be made from clutter areas that are extensive in azimuth and distance in order to avoid influences from individual target echoes.

The M matrices are inverted in a special processor (FB filter calculation). The first columns R _t ^-1 (m) e represent the M evaluations sought, ie the clutter filter.

The maximum of the squares of the amounts of the filtered data is compared with a threshold and when exceeded to the end evaluation (display, fine location, track logic, etc.) passed on.

The filtering or evaluation part BN works in such a way that each search channel 1. . . M (ie the M outputs of the Doppler filter bank DFB in FIG. 8) is passed to a summing element via filter coefficients. The summation elements are also connected to all L -1 auxiliary channels at other inputs via additional filter coefficients. The filter coefficients cause the auxiliary channels to estimate the clutter signal in the respective search channel and to subtract it from the current search channel signal. The clutter-free echo signals appear at the outputs, sorted by Doppler frequencies. The evaluation network carries out the calculation rule x _T R _T ^-1 (m) e in equation (3a).

As shown in FIG. 8, the amount squares are formed from the output signals of the evaluation network BN by means of squaring units. With the aid of a comparison circuit MAX , the maximum of the M signals is sought, fed to a threshold value detector and finally to the display.

In a covariance estimator, the output signals 1. . . M of the Doppler filter bank DFB and the L -1 auxiliary channels, the covariance matrix D of the auxiliary channels according to the equation (11) and the cross-correlation vectors b and the a for all M target Doppler frequencies (m = 1... M) calculated. The correlation estimates can be calculated by averaging over a distance and over time.

The correlation values determined in the covariance estimator are fed to a circuit for filter calculation FB . The filter coefficients, ie the first columns of R _T ^-1 (m) , are calculated in the circuit according to the algorithms given by equations (12) and (13). These are fed into the evaluation network. The adaptation speed, ie the speed at which the filter coefficients are recalculated, can be independent of the actual filtering process, ie the operation carried out by the evaluation network. It depends on the speed with which the clutter characteristics change as a result of the platform movement.

In the case of the non-adaptive method, the blocks "covariance estimation" and "matrix inversion" are saved. In contrast, the coefficients of the first columns of the precalculated inverse matrices R _T ^-1 (m) for omnidirectional clutter are stored in the "filtering" block.

From the estimated covariance matrix R _T (m) , an eigenvector decomposition can be carried out in parallel with the adaptation and filtering process, which brings the auxiliary channels up to date in larger periods. Alternatively, eigenvectors calculated completely off-line, e.g. B. those of a matrix for omnidirectional clutter, are kept in a memory and are used as auxiliary channels.

So the receiver contains three parts, in different Speed down:

• - The evaluation of the data with search and auxiliary channel coefficients and the filtering must be strictly real time.
• - The adaptation of the filter (covariance estimate, matrix inv sion) must be able to follow changes in the clutter environment, can generally be slower than the data filtering.
• - The auxiliary channels are either fixed by external data adjusted or also to the changing environment customized. This adjustment can be slower than that Adaptation run.

A radar receiver realized according to the invention detects moving radar targets (aircraft, driving witness) against a clutter background from a moving platform out. The receiver basically performs the following functions out:  

• - spatial and temporal integration of target signals (clubs education, Doppler filter bank)
• - Estimation of the clutter properties
• - Calculation of a clutter filter
• - Clutter suppression
• - Discovery and notification.

The receiver is adaptive, i. H. he can differ Liche clutter ratios, e.g. B. when flying over various Adjust landscape shapes automatically.

As a result of the adaptive method, the interference suppression also works when the radar is stationary, namely against moving (clouds) and unmoving interference echoes (ground). The receiver is on par with the optimal LR detector (gain in signal-clutter ratio). In comparison with the optimal detector, a receiver according to the invention represents a simplification and reduction of the effort by orders of magnitude. As a result, implementation is possible at all.

Because the receiver is a modification of the optimal detector he closes the principle of motion compensation implicitly with a.

In side view operation with omnidirectional transmitter, linear The receiving group and before homogeneous clutter distribution are the Clutter echoes right-left-symmetrical. With that, the Clutter signals, the filter calculation and the filtering real. This results in a computer saving by a factor of 4. This also applies to pure transverse operation with directive transmitter antenna.

The radar receiver according to the invention for moving radar devices has the following characteristics:

• - adaptivity
• - optimal motion compensation  
• - Optimal clutter suppression with reduced effort
• - Any number M of echo pulses can be used for clutter filtering.

With the invention described here, the computing effort for adaptation and filtering can be significantly reduced, so that a realization for not very large antennas with today's Technology and using modern algorithms (systolic arrays) is made possible.  

Abbreviations and names

β _n spatial phase position e unit vector φ azimuthal angle φ _B λ of sight L order of the transformed system η discovery thresholds m temporal index M number of echo pulses n vector of Clutterechos n _T transformed Cluttervektor n sensor index N number of sensors R Clutterkovarianzmatrix R _T T * R T R _T estimate of R _T S vector of the complex target signal components S _T transformed target signal vector T transformation matrix to reduce the order of the system leevolution m delay of the m th echo pulse v relative radial speed between clutter and radar v _max plate shape speed x vector of the complex received signals x _T transformed received signal vector x _n Sensor coordinate y _n Sensor coordinate CNR clutter / noise ratio SNR signal / noise ratio a Clutter power in the search channel b Vector of the cross-correlation values between search and auxiliary channels D Covariance matrix of the auxiliary channels f _D Doppler frequency L Number of properties used as auxiliary channels vectors

Claims (11)

1. A method for two-dimensional, ie in space and time, effective suppression of clutter signals of all kinds, in particular ground clutter in echo signals to find moving target objects by scanning the backscattered echo signal field generated by sending M transmission pulses with a radar receiver for a moving radar device with a , in particular phase-controlled group antenna with N sensors, in which the M echo signals received by the sensors are amplified in reception channels to complex input signals, mixed down, digitized, ordered and then a target signal is obtained from the input signals by signal-adapted filtering and suppression of clutter, characterized in that from the input signals in a known manner by forming lobes and Doppler filtering in the M frequencies, M search channels with M outputs, which deliver M complex search signals, and in parallel to this from the input signals L -1 at L N + M auxiliary channels for the S free of interference is generated by means of eigenvectors of a covariance matrix from spatial and temporal clutter echo samples, and clutter is then suppressed by linking the output of each search channel to all auxiliary channels in a rating network. 2. The method according to claim 1, characterized in that as auxiliary channels arithmetically or by measurement eigenvectors a covariance matrix can be used. 3. The method according to claim 2, characterized in that a for omnidirectional clutter precomputed covariance matrix ver is applied.   4. The method according to claim 1 or 2, characterized in that the Covariance matrix is estimated during radar operation. 5. The method according to any one of claims 1 to 4, characterized in that

• a) for a spatial and temporal target signal integration, a search lobe is formed from the N input signals of the N sensors by forming a weighted sum and subsequently M echo values of the search lobe are transformed into M Doppler frequencies in a Doppler filter bank and the M outputs of the Doppler filter bank M search channels, which deliver, form and form the complex M search signals
• b) that for the elimination of interference components (clutter) according to direction and frequency, all input signals further a network with a dimension (L -1) (N · M) , each column forming an auxiliary channel being formed by an intrinsic vector of the covariance matrix, which supplies L -1 transformed complex auxiliary signals,
• c) the L -1 auxiliary signals together with a search signal each form a reduced signal vector space and a search signal with all L -1 auxiliary signals in each case in a column of an L × M evaluation matrix that meets the calculation rule x _T * · R _T ^-1 (m ) e is sufficient, be summarized, and
• d) if necessary, the evaluation matrix is determined adaptively by estimating covariance matrices R _T (m) between each search signal and all L -1 auxiliary signals and from the first columns R _T ^-1 (m) e of the inverted M matrices the evaluations sought for clutter suppression are formed, and from the filtered signals obtained the largest amount is selected and brought to the threshold decision (target, no target) and display.

6. The method according to claim 5, characterized in that when determining the L × L covariance matrices R _T (m) of auxiliary signals and search signal, the all L × L covariance matrices R _T (m) together (L -1) × (L -1 ) - Submatrix of the covariance matrix of the auxiliary signals, that is the matrix of the L -1 eigenvectors belonging to the largest eigenvalues determined only once, e.g. B. is estimated and used for all search signals. 7. The method according to claim 5, characterized in that the Covariance matrix of the auxiliary channels and a search channel during of the radar operation is estimated from received echo data and the first column of their inverse to suppress clutter in the search channel is used. 8. The method according to any one of claims 5 to 7, characterized in that the estimation of the covariance matrix R _T (m) , its inversion and the clutter suppression for all search signals given by a Doppler filter bank are carried out simultaneously. 9. The method according to any one of claims 5 to 8, characterized in that a by Averaging over an azimuth and distance Area estimated covariance matrix is used. 10. The method according to any one of claims 5 or 6, characterized in that the first column of the inverse of for omnidirectional clutter before calculated covariance matrices R _T (m) of the auxiliary and search signals for clutter suppression is used in the associated search signal. 11. Radar receiver for performing the method according to one of claims 1 to 10 with a device for a two-dimensionally effective clutter suppression for moving coherent radar devices with a phase-controlled group antenna in particular with N sensors connected to N receiving channels coherent receiver with echo signal memory, characterized That a lobe-forming network with a Doppler filter bank is provided, to which the M complex input signals from N reception channels with echo signal memory for lobe formation and transformation into the Doppler frequency range are supplied, and the number of M outputs corresponding to the number M of transmission pulses of the radar for the transformed complex Search signals is equipped; that a network with L -1 at L N + M auxiliary channels from eigenvectors of a spatio-temporal covariance matrix of the clutter echo samples is provided with inputs for all complex input signals from the echo signal memories and L -1 outputs for the complex auxiliary signals; that each individual output of the Doppler filter bank with all L -1 auxiliary channel outputs via an L · M evaluation network, which carries out the calculation rule x _T · R _T ^-1 (m) e , where x _T transformed current input signal vectors, R _T (m) transformed covariance matrix of all L -1 auxiliary signals, each with a search signal, m represent a specific target Doppler frequency and e a unit vector, and a covariance estimation device which is equipped with inputs for the search signals coming from the Doppler filter bank and the auxiliary signals coming from the network of eigenvectors and their Output for transmitting the data to a filter calculation device, in which the evaluation factors for the evaluation network can be calculated using selected algorithms, and the output of the filter calculation system is connected to the evaluation network.