METHOD FOR DETECTING AN AIRCRAFT EQUIPPED WITH SURFACE ELEMENTS MADE OF METAMATERIAL OR "FREQUENCY SELECTIVE SURFACES" MATERIALS

Description

[0001] The invention relates to a method with the features of claim 1 and a radar system with the features of claim 14.

[0002] From the generic patent US 10 725 164 B1, a method for detecting a missile equipped with surface elements made of metamaterials or "frequency selective surfaces" materials is known. Two transmitting antennas send transmission pulses with different frequencies, and it is checked whether the transmission pulses are scattered by the missile. In the case of backscattering, radar pulses with a modified signal are backscattered by means of a control mechanism of the surface elements.

[0003] US patent 2016 / 178729 A1 discloses a VHF / UHF radar for detecting stealth aircraft. It generates two or three radar beams with different wavelengths, the beams being longer than the aircraft's length. Different beam orientations are used to determine the aircraft's attitude.

[0004] US patent 2015 / 029052 A1 reveals a quantum harmonic radar for detecting stealth aircraft.

[0005] From DE 10 2016 015 107 B3, a method for operating a radar system to prevent deception by third parties is known, wherein a connection is provided between at least one direction-finding receiver and a search radar. Several pulses with different codes and / or different signal forms are transmitted from a transmitting antenna, namely the search radar. Random numbers are superimposed on the transmitted pulses of the search radar using a digital modulation method. The direction-finding receivers and / or the search radar receive pulses scattered at a target. It is then checked whether these received pulses are scattered at a target and thus form a useful signal or are formed by decoy signals. For this purpose, the reception directions of the received pulses are determined using the at least one direction-finding receiver and / or the search radar.The system checks whether the receiving directions can be assigned to a common target area, thus forming a useful signal, or whether they are not assigned to a common target area and thus form a decoy signal. This allows a distinction to be made between a decoy signal and a useful signal. The functionality described here is based on the receivers in the search radar and direction finders recognizing the code used in the transmitted pulse. An additional method is proposed below, which allows the system in DE 10 2016 015 107 B3 to also detect the presence of hostile objects that would disrupt the signal similarity.

[0006] For several years, research and development in the field of Frequency Selective Surfaces (FSS) and Metamaterials (MM) has been conducted in electrodynamics. Applications can be found in antennas and radar cross-section (RCS) modification. To date, corresponding proposals for FSS and MM have been rather static and passive. This means that the structures are designed for specific applications and cannot change their properties during operation. However, efforts are now emerging to modify the properties of FSS and MM during use, or to adapt them to changing environments. Initial approaches in the civilian sector are evident in the literature. It is noted that controlling the reflection behavior is also possible with an applied control voltage, and that larger areas can be covered in this way.Furthermore, it can be inferred from the specialist literature that the production costs for manufacturing one unit of area will not be so high.

[0007] Many areas of application can be identified in the specialist literature, particularly in the military field. The problem can now be discussed using the following examples: Fig. 1 and Fig. 2 will be explained. Fig. 1 Figure 1 shows a missile 1 with a surface element 2 made of metamaterial, which is adaptively controllable. Furthermore, a search radar 3 is present, which emits transmit pulses 4 with a pulse duration 5, TP. The transmit pulse 4 is scattered by the missile 1 with the surface elements 2, resulting in a radar pulse 6 scattered by the missile 1 with a signal SC and a signal value C as a random variable. The signal SC 6 has an expected value E { C} = 0 or E {C} ≠The detection distance RD is 0 and is statistically independent and uncorrelated with the noise signals NA 22 and NB 23, which is the normal case in reality. The detection distance RD is indicated by the reference symbol 7.

[0008] Assuming that the surface of missiles 1, such as aircraft, drones, or missiles, is coated with a configuration of the aforementioned surface units 2, and that each surface unit 2 is supplied with a control voltage, the RCS can be modified during flight. Thus, according to an approach scenario, the relevant RCS can be reduced to decrease the detection distance RD 7 to a search radar 3. Generally, such measures for RCS reduction are described as "stealth" measures. If the RCS properties of the missiles are not modified, this can be described as a passive "stealth" principle. If, however, the RCS properties are adaptively modified, this can be described as an active "stealth" principle.

[0009] In the active "stealth" concept, where the reflection behavior of the individual surface units 2 is modified by a control voltage, an additional effect for reducing the detection distance can be achieved besides influencing the RCS. By varying the control voltage, the reflection can be altered in amplitude and phase within the radar pulse duration TP 5 of the transmit pulse 4. If all surface units 2 are now driven with noise-like control voltages, where there is no correlation between the individual control voltages, the scattered radar pulse 6 of the search radar 3 undergoes modulation in amplitude and phase. In the extreme case, the scattered radar pulse 6 loses its similarity to the transmit pulse 4. The search radar 3 thus receives the scattered radar pulse 6 and attempts to detect it from the instrument noise using signal-matched filters, so-called "matched filters".As long as the received radar pulse 6 is similar to the expected received pulse, the signal-to-noise ratio (SNR) at the output of the "matched filter" can be increased. This enables the search radar 3 to detect even small signal components from the noise.

[0010] If the scattered radar pulse 6 is modulated by the control voltages to such an extent that it resembles a noise signal, the matched filters become ineffective. This means that the signal-to-noise ratio (SNR) cannot be improved, and thus the probability of detection decreases. This results in a reduction of the detection distance, which gives the attacking missile 1 an advantage. Search radar systems 3 that use the pulse compression method are particularly affected by this type of active "stealth" because they have a relatively long pulse duration TP 5. The same active "stealth" principle can also be applied to tracking radar systems.

[0011] The invention is therefore based on the objective of providing a method in which a search radar or tracking radar is enabled, in addition to its existing capabilities, to detect the signal scattered by a target, which no longer bears any resemblance to the original transmitted signal, and to determine the location of the scattering body.

[0012] This problem underlying the invention is now solved by a method with the features of claim 1 and a radar system with the features of claim 14.

[0013] The method serves to detect a missile with surface elements made of metamaterials or frequency-selective surfaces (FSS) materials, wherein a transmit pulse is sent by means of a transmitting antenna, and it is checked whether the transmit pulse is scattered by the missile. In the case of backscattering, a radar pulse with a modified signal is backscattered by means of a control of the surface elements. The method is characterized in that two receive signals are received by means of two receive antennas, wherein the receive signals are sampled and samples n=1..N of the receive signals are generated. The receive signals A(n)+C(n) and B(n)+C(n) each have a noise component A(n) and B(n), respectively, and in the case of backscattering, each has an additional signal component C(n) influenced by the surface elements. The samples of the receive signals are first multiplied, and the multiplied samples are summed over a sliding time window to generate a signal. S = ∑ n = 1 N C n + A n ⋅ C n + B n is determined, whereby an expected value E S = ∑ n = 1 N E C n 2 + E C n ⋅ B n + E A n ⋅ C n + E A n ⋅ B n is determined by comparing the expected value E{S} with a threshold value to determine whether a common signal (SC) is present in both received signals or not.

[0014] The method is particularly useful for detecting a missile with surface elements made of metamaterials or "frequency selective surfaces" materials, wherein a transmit pulse is sent by means of a transmitting antenna, and it is checked whether the transmit pulse is scattered by the missile, wherein, in the case of backscattering, a radar pulse with a modified signal is backscattered by means of a control of the surface elements, characterized in that two received signals are received by means of two receiving antennas, wherein the two receiving antennas are placed close to each other and form a twin antenna, wherein the received signals are sampled and samples n=1.N of the received signals are generated, wherein A(n) describes a random variable of the noise signal voltage of the n-sample, wherein B(n) describes a random variable of the noise signal voltage of the n-sample, wherein C(n) describes a random variable of the radar pulse of the n-sample scattered by the missile, wherein the random variables of the received signals A(n)+C(n) and B(n)+C(n) each have a noise component A(n) and B(n) respectively, and in the case of backscattering each have an additional signal component C(n) influenced by the surface elements, wherein the samples of the random variables of the received signals are first multiplied and the multiplied samples are summed over a sliding time window and a signal . S = ∑ n = 1 N C n + A n ⋅ C n + B n is determined, whereby an expected value E S = ∑ n = 1 N E C n 2 + E C n ⋅ B n + E A n ⋅ C n + E A n ⋅ B n is determined by comparing the expected value E{S} with a threshold value to determine whether a common signal (SC) is present in both received signals or not.

[0015] The radar system serves to execute the method and comprises a transmitting antenna, two receiving antennas, and an evaluation circuit. Transmitting pulses can be sent via the transmitting antenna, and received signals can be received via the two receiving antennas. The received signals can be evaluated by the evaluation circuit as follows: the received signals can be sampled, and samples n=1...N of the received signals can be generated. The samples of the received signals can first be multiplied, and the multiplied samples can then be summed over a sliding time window to produce a signal. S = ∑ n = 1 N C n + A n ⋅ C n + B n is determined, whereby an expected value E S = ∑ n = 1 N E C n 2 + E C n ⋅ B n + E A n ⋅ C n + E A n ⋅ B n is determinable, wherein the expected value E{S} is comparable to a threshold value, whereby it is possible to determine whether a common signal (SC) is present in both received signals or not.

[0016] The radar system serves in particular to execute the method and comprises a transmitting antenna, two receiving antennas, and an evaluation circuit. Transmitting pulses can be sent via the transmitting antenna, and received signals can be received via the two receiving antennas, the two receiving antennas being positioned close to each other and forming a twin antenna. The received signals can be evaluated by the evaluation circuit as follows: the received signals can be sampled, and samples n=1...N of the received signals can be generated. Random variables of the received signals A(n)+C(n) and B(n)+C(n) each have a noise component A(n) and B(n), respectively, and in the case of backscattering, each has an additional signal component C(n) influenced by the surface elements. The samples of the received signals can first be multiplied, and the multiplied samples can be summed over a sliding time window to produce a signal. S = ∑ n = 1 N C n + A n ⋅ C n + B n is determined, whereby an expected value E S = ∑ n = 1 N E C n 2 + E C n ⋅ B n + E A n ⋅ C n + E A n ⋅ B n is determinable, whereby the expected value E{S} is comparable to a threshold value, whereby it is possible to determine whether a common signal (SC) is present in both received signals or not.

[0017] At least two receiving antennas are used. Two of these receiving antennas form a twin antenna. Several such twin antennas can be used. In particular, each direction-finding receiver can be equipped with a twin antenna. It is conceivable that the search radar could also be equipped with a twin antenna.

[0018] The two receiving antennas must be positioned close to each other so that both capture the scattered electromagnetic field in such a way that no significant differences in arrival time occur. The arrival time differences, and thus the distance between the two antennas 10a and 10b, should not exceed the coherence period of the signal SC. The coherence period is defined as the time it takes for autocorrelation to show a correlation between the original signal SC and a copy of the signal waveform SC shifted by a certain time. To reduce the mutual coupling of this "twin antenna" arrangement to a tolerable level, a corresponding design must be implemented. Elements are used in the vicinity of the two antennas 10a and 10b, as well as in the space between them, to minimize mutual coupling and interference.

[0019] An omnidirectional transmitting antenna, in particular, emits pulses of a specific duration and at a specific carrier frequency. The carrier frequency and its bandwidth must be selected so that no other radar systems, radio services, or other artificial sources transmit in this frequency range. Furthermore, background radiation should also be taken into account during frequency planning. The pulse duration is calculated by multiplying a predetermined number of samples (N) by a predefined sampling rate. These transmitted pulses can be coded or uncoded. Preferably, the transmitting antenna is located at the search radar. The transmitting antenna can also be part of the search radar. The transmitted pulses can thus be sent from a transmitting antenna of the search radar. Alternatively, an additional transmitting antenna can be located at a different location than the search radar.

[0020] The signals received by the receiving antennas are preferably filtered and amplified separately and converted analog-to-digital (A / D) 15, 16. Due to the electronic components in the respective receiving paths, the received signals each contain noise or noise components. This noise reduces the detectability of a signal SC of the radar pulse 6.

[0021] The A / D conversion is preferably operated by a common sampling frequency generation 17 so that jitter in the sampling clock has no influence on the operation.

[0022] It can be assumed that the additive noise signals in the respective receiving paths are statistically independent of each other, uncorrelated, and without DC components. This means that the respective expected values ​​of the N samples of the noise voltage values ​​A(n) and B(n), which are random variables, become zero. E { A ( n )} = E { B ( n)} = 0 , where n = 1 ... N This is the norm in radar receivers.

[0023] After the A / D conversion 15, 16, the corresponding samples of the two paths are, simplified for the nth sample Pa: [ C ( n ) + A ( n )] and Pb: [ C ( n ) + B ( n )] , First, the values ​​of 25 are multiplied together and summed over a sliding time window. The length of the time window corresponds to the number of samples N. With each summation, the time window is shifted by one clock cycle; that is, the oldest value is subtracted from the sum and the current value is added to the result.

[0024] The output signal of the summation is then fed to a threshold detector 18, which then decides whether a common signal SC of the scattered radar pulse 6 was present in both received signals. The signal SC of the radar pulse 6 is statistically independent and uncorrelated with the noise signals. The sampled signal voltages of the scattered radar pulse are random variables and are denoted by C and C(n), respectively. Furthermore, the following applies to the expected value of the signal voltage C and the random variable C. E { C} = 0 or E { C} ≠ 0 .

[0025] If a signal SC is received via both antennas 10a and 10b, a DC component is generated during multiplication. However, the formation of a constantly occurring DC component in the uncorrelated and statistically independent noise voltages of the two receiving channels A and B is suppressed by the absence of individual DC components in the noise.

[0026] Transmitter 9 sends pulses with the selected pulse duration TP at certain time intervals via transmitting antenna 8. The pulse repetition frequency PRF is set so low that the direction finders can reliably detect a pulse at all times. This eliminates the need for pulse discrimination by the direction finders. Since only the target generates a modulated (scattered) signal SC 6, it is possible to distinguish between the pulses of normal operation according to DE 10 2016 015 107 B3 and the pulse scattered by the target. Therefore, such modulated pulses indicate the presence of an aircraft with active stealth capabilities. If the transmitting antenna 8 is placed at the same location as the search radar antenna from DE 10 2016 015 107 B3, the direction finders can distinguish between clutter and flying object by simultaneously evaluating the coded radar pulses known from DE 10 2016 015 107 B3 and the received signal SC 6.Radar pulses scattered by clutter objects such as forests, mountains, or other obstacles are still similar to the transmitted radar pulse. Clutter is present when both the method described in DE 10 2016 015 107 B3 and the method presented here result in a detection.

[0027] The direction of incidence of signal SC or radar pulse 6 is determined by each direction finder and communicated to the search radar via the existing information pathway from DE 10 2016 015 107 B3. Using radiation methods as described in DE 10 2016 015 107 B3, the location of the aircraft can be determined.

[0028] In the following, important characteristics of the method known from DE 10 2016 015 107 B3 may be explained again, since these characteristics are or may be part of the present method.

[0029] The method uses a search radar and at least one direction-finding receiver. Preferably, several direction-finding receivers are present. The direction-finding receivers and the search radar are not located in the same place, but rather in different locations. This allows the target to be located from multiple directions. The search radar transmits several pulses at time intervals with different codes and / or different waveforms. The codes are generated randomly. The waveforms are varied randomly.

[0030] The codes are generated by a random number generator. For example, N-digit binary numbers can be used as codes. The search radar uses a random code and modulates the radar transmission pulse with this code. Specifically, one N-digit binary number is randomly selected from 2N < -M possible N-digit binary numbers, and the radar transmission pulse is modulated with this number using a digital modulation technique. Thus, the search radar has a choice from 2N < -M binary numbers. M is the number of binary numbers that have already been selected within a certain time period. All codes within M pulse repetition intervals are preferably different, where M is greater than or equal to 2. Therefore, if a binary number has been selected and is used for a pulse repetition interval, it should only be selectable again after a certain time. If N is chosen to be large, e.g., N=7 and M=2, then for each pulse repetition interval PRI, there are 126 selectable binary numbers.The previously selected number is blocked for the duration of M = 2 pulse repetition intervals, but can be selected again after the second pulse repetition interval. If the number N is sufficiently large, it can even be publicly known, meaning the opponent is allowed to know it. Conversely, M can be a secret fixed value or even determined by a random number generator in the search radar. However, M > 1 must hold true. With this measure, the pulse repetition interval is always assigned a unique number, and all pulses belonging to this pulse repetition interval can be assigned to this interval and thus to a specific time period.

[0031] As a further option, each number can be assigned a specific waveform. These waveforms should preferably be chosen so that they are orthogonal to each other. This allows the receiver to extract the specific, current waveform from a multitude of other received waveforms using a signal-matched filter. This also allows multipath propagation effects to be processed to such an extent that the desired received signal can be temporally separated from the other received signals.

[0032] A constant carrier frequency can be used for all transmission pulses. If, due to changes and / or fluctuations in the radar cross-section, the use of multiple transmission frequencies becomes necessary to increase the probability of target detection, the number of carrier frequencies can be increased. A number of K=4 is considered a typical nominal value. The use of multiple frequencies is also known as frequency diversity and serves solely to increase the probability of target detection in the case of highly fluctuating radar cross-sections. Since modern jammers are capable of simultaneously disrupting a wide frequency band with a noise signal, frequency changes are no longer applicable as electronic countermeasures. In this sense, a single transmission frequency is sufficient to keep the radar system simple to implement.

[0033] The search radar generates a burst consisting of Q-pulses, each preferably modulated onto the carrier frequency with a different randomly chosen binary number using a digital modulation technique. The pulse repetition interval can be constant or change from pulse to pulse within the burst duration. The decoy system, hereinafter referred to as the DRFM system, can only copy and modify a received radar transmit pulse if it arrives at the DRFM system's location. If a DRFM system wants to simulate a false target spatially in front of the actual target location using a decoy signal or false target pulse before the actual radar transmit pulse arrives, it must rely on the already received transmit pulses. If the search radar changes the pulse parameters from radar transmit pulse to transmit pulse, the DRFM system is no longer able to simulate false targets at a distance shorter than the actual distance to the target.This limits the feigning of false target distances to the spatial area behind the real target up to the Instrumented Range, i.e., the maximum clearly determinable distance.

[0034] The decoy system, or DRFM system, is typically installed on an aircraft and has a receiving and a transmitting antenna. This means that the DRFM system can only physically be located at a single, fixed point in space. The DRFM system can attempt to deceive the search radar into believing that the target is located at a different azimuth through side-lobe jamming. Side-lobe jamming means that the search radar's receiving antenna picks up a decoy signal via its sidelobes, but the radar assumes that the received signal originated from the antenna's main lobes. However, since the DRFM system's transmitting antenna can only be physically located at one point, a decoy attempt can be detected by determining the location of the electromagnetic field source. By combining the data, the search radar recognizes that the false target cannot possibly be within the main lobe of the search radar's receiving antenna.As a side effect, it is also possible to distinguish between ground clutter and the target if the ground clutter is located in the same resolution cell as the target. In this case, too, the sources of scattering are spatially separated from each other in the azimuth of the elevation.

[0035] A particular advantage of the invention is the use of at least one, and in particular several, direction-finding receivers. These direction-finding receivers can be passive. The opponent may even know the location of the direction-finding receivers.

[0036] The physical condition that the source of the deception signal, namely the DRFM transmitting antenna, is attached to an aircraft makes its location unambiguously determinable. Even if the DRFM system attempts to manipulate the phase front using antennas distributed above the aircraft, spatial deception is impossible, or only possible under theoretical conditions, due to the finite size of the aircraft and the use of multiple spatially distributed direction finders. In practice, these theoretical conditions are never met, as many parameters, such as the locations of the direction finders or the orientation of the DRFM antennas, will never be perfectly aligned.

[0037] A monostatic search radar is typically used as the search radar. The direction-finding receivers operate asynchronously to the monostatic search radar. Precise time synchronization is not necessary, as the direction-finding receivers do not need to measure the travel time of the radar pulse from the radar antenna, across the target, to the direction-finding antennas. The direction-finding receivers and the search radar receive the received pulses scattered at the target or emitted by the decoy device.

[0038] Several queries can now determine whether the received pulses are the desired signal or a decoy signal. First, it is checked whether the received pulses were received by the direction finder within a specific time window after transmission. The propagation time of a signal between the at least one direction finder and the search radar is determined, whereby the time window for receiving the pulse is calculated based on the propagation time and the size of the area to be monitored. Received pulses that reach the respective direction finder outside of this time window are not considered the desired signal. This time window is determined for each transmitted pulse and for each direction finder.

[0039] Furthermore, it is checked whether the received pulses have the corresponding code. If both of these conditions are met, the direction of reception of the received pulse is determined using the direction finder and also by the search radar. By comparing the direction of reception, it can be verified whether the received pulses originate from a common target area. If all the direction of reception can be assigned to a target area, then it is not a deceptive signal, but a legitimate signal. By three-dimensionally tracing an incoming electromagnetic field, namely the corresponding received pulses and the bearing values ​​measured from them, the search radar can also distinguish between ground clutter and the target by combining all the bearing values.

[0040] The disadvantages mentioned at the beginning are therefore avoided and corresponding advantages are achieved.

[0041] There are now numerous possibilities for designing and further developing the inventive method and the radar system. For this purpose, reference may first be made to the claims subordinate to claim 1. The invention will now be explained in more detail with reference to the drawing and the accompanying description. The drawing shows: Fig. 1 shows a schematic representation of a missile and a search radar, wherein a transmitted pulse is scattered by surface elements made of metamaterial on the missile and is received by the search radar in the form of a scattered radar pulse, Fig. 2 shows a circuit arrangement for determining whether a radar pulse scattered by the missile has been detected with a signal SC by two receiving antennas, and Fig. 3 shows a diagram with a simulation result, wherein a signal S is recorded as a family of curves over a number of realizations.

[0042] As already described in Fig. 1A transmitted pulse 4 is scattered by a missile 1, whereby surface elements 2 made of metamaterial on the missile 1 generate a noise signal from the transmitted pulse 4. To detect this, a twin antenna with two receiving antennas 10a, 10b is used, which receive a signal SC of the scattered radar pulse 6. The receiving antennas 10a, 10b must be placed close to each other so that both detect the scattered electromagnetic field, here in the form of the signal SC, in such a way that no significant differences in travel time occur. To reduce the mutual coupling of this "twin antenna" arrangement to a tolerable level, the design must be implemented accordingly.

[0043] Each direction finder and, if applicable, the search radar 3 are equipped with a twin antenna.

[0044] In addition to normal operation, as described in DE 10 2016 015 107 B3, an omnidirectional transmitting antenna 8 emits transmit pulses with a specific duration TP, 5. The pulse duration is calculated from the number of samples N, 25 and the predefined sampling rate, which is known from sampling frequency generation 17. The transmit pulses are generated by a signal conditioning unit 9. These transmit pulses can be coded or uncoded. Furthermore, the transmit frequency may deviate from that used for normal operation. The transmit frequency and its bandwidth must be selected such that no other radar systems, radio services, or other artificial sources transmit in this frequency range. Background radiation should also be taken into account during frequency planning. Preferably, the transmitting antenna 8 is located at the location of the search radar 3.

[0045] In Fig. 2The signal waveform is shown schematically. The instrument noise NA 22 and NB 23 is modeled as additive noise, as is standard practice in radar technology. The instrument noise NA 22 and NB 23 is expressed by the random variables A and B. The random variables A and B associated with the corresponding instrument noise NA 22 and NB 23 are statistically independent and uncorrelated.

[0046] The respective received signals are separately filtered and amplified via paths Pa, 12 and Pb 13, and converted to analog-to-digital (A / D) signals 15, 16. Noise NA, 22 and NB, 23 is introduced into the received signals by the electronic components in the respective receiving paths. This noise reduces the detectability of signal SC, 6.

[0047] The A / D conversion 15, 16 is operated by a common sampling frequency generation 17 so that jitter has no influence on the operation.

[0048] It can be assumed that the additive noise signals NA, 22 and NB, 23 in the respective receive paths Pa, Pb are statistically independent of each other, uncorrelated, and without DC components. This means that the respective expectation values ​​of the N samples of the noise voltage values ​​A(n) and B(n), which are random variables, become zero. E { A ( n )} = E { B(n )} = 0 , where n = 1 ... N This is the norm in radar receivers.

[0049] After the A / D conversion 15, 16, the corresponding samples of the two paths are, simplified for the nth sample Pa: [ C ( n ) + A ( n )] and Pb: [ C ( n ) + B ( n )] ,First, the values ​​are multiplied together (24) and summed over a sliding time window (26). The length of the time window corresponds to the number of samples N (25). With each summation operation, the time window is shifted by one clock cycle, i.e., the oldest value is subtracted from the sum and the current value is added to the result.

[0050] The output signal of the summation is then fed to a threshold detector 18, which then decides whether a common signal SC was present in both received signals.

[0051] The signal SC of the scattered radar pulse 6 is statistically independent and uncorrelated with the noise signals NA 22 and NB 23. Furthermore, the expected value of the signal voltage C, which is also a random variable, is given by: E{C} = 0 or E{C} ≠ 0 .

[0052] If a signal SC of the scattered radar pulse 6 is received via both antennas 10a and 10b, a DC component is generated during the multiplication 24. In contrast, the formation of a constantly occurring DC component in the uncorrelated and statistically independent noise voltages NA and NB is suppressed by the absence of the individual DC components in the noise.

[0053] Transmitter 9 emits pulses with the selected pulse duration TP 5 at certain time intervals. The pulse repetition frequency (PRF) is set so low that the direction finders can reliably detect a pulse every time. This eliminates the need for pulse discrimination by the direction finders. Since only the target generates a modulated scatter signal (SC), it is possible to distinguish between the pulses of normal operation (according to DE 10 2016 015 107 B3) and the pulse scattered by the target. Therefore, such modulated pulses indicate the presence of an aircraft 1 with active stealth capabilities.

[0054] The direction of incidence of the signal SC 6, or of the radar pulse 6 scattered by the missile 1, is determined by each direction finder and communicated to the search radar 3 via the existing information pathway from DE 10 2016 015 107 B3. The location of the missile 1 can be determined using radiation methods as described in DE 10 2016 015 107 B3.

[0055] The following section will explain the principle mathematically using expected values. This discussion serves to clarify the preceding description and is limited to the essentials.

[0056] The noise signal NA, 22 has an expected value E { A ( n )} = 0, where A(n) describes a random variable of the noise signal voltage of the n-sample. The noise signal NA, 22 has an expected value E { A} = 0 and is statistically independent and uncorrelated with respect to the noise signal NB and the signal SC.

[0057] The noise signal NB, 23 has an expected value E { B ( n )} = 0, where B(n) describes a random variable of the noise signal voltage of the n-sample. The noise signal NB has an expected value E { B} = 0 and is statistically independent and uncorrelated with respect to the noise signal NA and the signal SC.

[0058] Signal SC of radar pulse 6: The signal SC, 6 can have an expected value equal to zero or not equal to zero, i.e. E { C ( n )} = 0 or E { C ( n )} ≠ 0 , where C(n) is a random variable of the signal voltage of the signal SC, 6 scattered on the missile.

[0059] The following always holds true: Random variables A(n), B(n) and C(n) are statistically independent and uncorrelated, which is true in real systems.

[0060] The expected value of signal S after summing 26 is: S = ∑ n = 1 N C n + A n ⋅ C n + B n E S = ∑ n = 1 N E C n 2 + E C n ⋅ B n + E A n ⋅ C n + E A n ⋅ B n

[0061] Without signal SC, 6 : E S = 0

[0062] With signal SC, 6 : E S = ∑ n = 1 N E C n 2 > 0

[0063] A simulation will be used to graphically illustrate the situation ( Figure 3 This will be illustrated. The following probability density functions (PDF) for the random variables are assumed for simplification.

[0064] Noise NA 22: WDF : p A a n = 1 2 ; − 1 ≤ a n ≤ 1 0 ; sonst , mit E A n = 0

[0065] Noise NB 23: WDF : p B b n = 1 2 ; − 1 ≤ b n ≤ 1 0 ; sonst , mit E B n = 0

[0066] Signal SC 6: WDF : p C c n = 1 2 ⋅ A C ; − A C ≤ c n ≤ A C 0 ; sonst , mit E C n = 0

[0067] Furthermore: 0 ≤ A C ≤ 1

[0068] It is clear in Figure 3 A free zone 33 can be seen between the curve sets 30 without signal SC and 31 with signal SC for N > 4000. Figure 3 This applies to an SNR per channel of -10.5dB. Therefore, the threshold can be placed in the free zone 33 and it can be determined whether a signal SC, 6 has been received or not.

[0069] Furthermore, the course of the expected value 32 is also: E { S} = N · E {[ C ] 2<} drawn. Reference symbol list:

[0070] 1: Missile 2: Surface element made of metamaterial, which is adaptively controllable 3: Search radar 4: Transmit pulse 5: Pulse duration TP 6: (Radar) pulse scattered by the missile with signal SC and the signal value C as a random variable of the n-sample 7: Detection distance RD 8: Omnidirectional transmitting antenna 9: Transmit signal processing 10a: Receiving antenna, path Pa 10b: Receiving antenna, path Pb 11: Transmit pulse duration generation and calculation 12: Path Pa: Receiving unit (bandpass filter, amplifier, .... ) 13: Path Pb: Receiving unit (bandpass filter, amplifier, ....14: Time-of-flight adjustment unit to match the signal propagation times in paths Pa and Pb 15: Path Pa: Analog / digital conversion 16: Path Pb: Analog / digital conversion 17: Sampling frequency generation 18: Threshold detector 19: Path Pa: Schematic representation for coupling the additive noise signal NA 20: Path Pb: Schematic representation for coupling the additive noise signal NB 21: Not assigned 22: Path Pa: Noise signal NA with voltage value as random variable A of the n-sample. The noise signal NA has an expected value. E { A} = 0 and is statistically independent and uncorrelated with respect to the noise signal NB and signal SC. 23: Path Pb: Noise signal NB with voltage value as random variable B of the n-sample. The noise signal NB has an expected value. E { B} =0 and is statistically independent and uncorrelated with the noise signal NA and signal SC. 24: Multiplier 25: N: Number of samples to be summed 26: Summer over a time window with N samples with output signal S 27: Result, whether a signal SC is present or not 28: Signal S after the summer 29: N: Number of summations 30: Family of curves without signal SC with 300 realizations 31: Family of curves with signal SC with 300 realizations 32: Expected value of S: E { S} 33: Free zone for determining the threshold

Claims

1. Method for detecting a flying object (1) equipped with surface elements (2) made of metamaterials or "Frequency Selective Surfaces" materials, wherein a transmission pulse (4) is transmitted by a transmitting antenna (8), wherein a check is carried out as to whether the transmission pulse (4) is scattered at the flying object (1), wherein in the event of backscattering by controlling the surface elements (2) a radar pulse (6) having a modified signal (SC) is backscattered, characterized in that two receive signals are received by two receiving antennas (10a, 10b), wherein the two receiving antennas are placed close to each other and form a twin antenna, wherein the receive signals are sampled and samples n=1..N of the receive signals are generated, wherein A(n) describes a random variable of the noise signal voltage of the nth sample, wherein B(n) describes a random variable of the noise signal voltage of the nth sample, wherein C(n) describes a random variable of the radar pulse (6) of the nth sample scattered at the flying object, wherein the random variables of the receive signals A(n)+C(n) or B(n)+C(n) each have a noise component A(n) or B(n) and in the event of backscattering each have an additional signal component C(n) influenced by the surface elements, wherein the samples of the random variables of the receive signals are initially multiplied (24) and the multiplied samples are summed over a sliding time window and a signal is determined, wherein an expected value E S = ∑ n = 1 N E C n 2 + E C n ⋅ B n + E A n ⋅ C n + E A n ⋅ B n is determined, wherein the expected value E{S} is compared with a threshold value, wherein it is ascertained whether or not a common signal (SC) is present in the two receive signals.

2. Method according to Claim 1, characterized in that an A / D conversion (15, 16) of the receive signals from a common sampling frequency generation (17) is performed.

3. Method according to Claim 1 or 2, characterized in that the length of the transmission pulses is determined by multiplication from a predetermined number (N, 25) of samples and a predefined sampling rate.

4. Method for operating a radar system to prevent deception by third parties, wherein a connection is provided between at least one direction-finding receiver and a search radar, wherein a plurality of pulses having different codes and / or having different signal waveforms are transmitted by the search radar (3), wherein the direction-finding receivers and / or the search radar receive pulses (6) scattered at the flying object (1), wherein a check is carried out as to whether these receive pulses (6) are scattered at a target and thus form a useful signal or are formed by deception signals, wherein receiving directions of the receive pulses (6) are determined by the at least one direction-finding receiver and / or the search radar, wherein a check is carried out as to whether the receiving directions can be assigned to a common target area and thus form a useful signal or are not assigned to a common target area and thus form a deception signal, wherein the search radar (3) and / or the direction-finding receivers are / is each equipped with two receiving antennas in order to additionally carry out the method according to Claim 1.

5. Method according to Claim 4, characterized in that the transmitting antenna (8) is part of the search radar (3).

6. Method according to either of preceding Claims 4 and 5, characterized in that the pulse repetition frequency (PRF) of the transmitting antenna (9) is selected to be so low that the direction-finding receivers can always reliably register a radar pulse (6).

7. Method according to one of preceding Claims 4 to 6, characterized in that the codes are generated randomly.

8. Method according to one of preceding Claims 4 to 7, characterized in that all codes are selected to be different within M pulse repetition intervals, wherein M is greater than or equal to 2.

9. Method according to one of preceding Claims 4 to 8, characterized in that the signal waveform is randomly selected and varied from transmission pulse to transmission pulse.

10. Method according to one of preceding Claims 4 to 9, characterized in that a constant carrier frequency is used for all transmission pulses.

11. Method according to one of preceding Claims 4 to 10, characterized in that a check is carried out as to whether the receive pulse was received by the direction-finding receiver and / or by the search radar in a determined time window after the transmission pulse was sent.

12. Method according to one of preceding Claims 4 to 11, characterized in that the time of flight of a signal between the at least one direction-finding receiver (21, 22, 23) and the search radar (20) is determined, wherein the time window for receiving the receive pulse is calculated on the basis of the time of flight and the size of the area to be monitored.

13. Method according to one of preceding Claims 4 to 12, characterized in that a plurality of direction-finding receivers (21, 22, 23) are used.

14. Radar system for implementing the method according to one of the preceding claims, having a transmitting antenna (8) and having two receiving antennas (10a, 10b) and an evaluation circuit, wherein transmission pulses (4) are able to be transmitted by the transmitting antenna (8) and receive signals are able to be received by the two receiving antennas (10a, 10b), wherein the two receiving antennas are placed close to each other and form a twin antenna, wherein the receive signals are able to be evaluated by the evaluation circuit, wherein the receive signals are able to be sampled and samples n=1..N of the receive signals are able to be generated, wherein A(n) describes a random variable of the noise signal voltage of the nth sample, wherein B(n) describes a random variable of the noise signal voltage of the nth sample, wherein C(n) describes a random variable of the radar pulse (6) of the nth sample scattered at the flying object, wherein the receive signals A(n)+C(n) or B(n)+C(n) each have a noise component A(n) or B(n) and in the event of backscattering each have an additional signal component C(n) influenced by the surface elements, wherein the samples of the random variables of the receive signals are initially able to be multiplied (24) and the multiplied samples are able to be summed over a sliding time window and a signal S = ∑ n = 1 N C n + A n ⋅ C n + B n is able to be determined, wherein an expected value E S = ∑ n = 1 N E C n 2 + E C n ⋅ B n + E A n ⋅ C n + E A n ⋅ B n is able to be determined, wherein the expected value E{S} is able to be compared with a threshold value, wherein it is possible to ascertain whether or not a common signal (SC) is present in the two receive signals.