FR2904696A1 - Signal filtering method for aircraft, involves retaining signals received after and before emission of pulses, respectively, during period lower to time necessary to one pulse to travel distance moved/returned between antenna and obstacle - Google Patents

Abstract

The method involves retaining signals received immediately after the emission of an electromagnetic pulse and before the emission of another electromagnetic pulse, during a period lower or equal to the time necessary to the former pulse to travel the distance moved/returned between an airborne antenna and an obstacle. The antenna emits a synthetic aperture radar (SAR) type wave. The retained signals are not corresponding to backscattering by ground. Independent claims are also included for the following: (1) a method for detecting the interferer with the ground by an airborne antenna (2) a method for emitting and receiving electromagnetic signals.

Description

 Signal filtering method, method for detecting a scrambler on the ground and method for transmitting and receiving electromagnetic signals emitting a frequency hopping waveform

The present invention relates to a filtering method, among the electromagnetic signals received by an antenna, signals not corresponding to the back-diffusion of an electromagnetic pulse previously emitted by the antenna by an obstacle located at a known distance. The present invention also relates to a method of detecting a jammer on the ground by an airborne antenna emitting electromagnetic pulses. The present invention also relates to a method for transmitting and receiving electromagnetic signals for an airborne radar emitting a frequency hopping waveform. It applies for example in the field of radar.

Embedded on board an aircraft, a radar operating in SAR mode, according to the Anglo-Saxon expression of Synthetic Aperture Radar, provides a two-dimensional image of the ground clearly showing the configuration of the installations. The SAR waveform is a coherent waveform adapted to receive the backscattered signals from the ground. The final image reflects the retro-diffused power, which depends on the nature of the terrain. It is in fact a mapping of the coefficient of electromagnetic backscattering of the soil. To build the final image, a power scale is associated with a color palette, usually from black to white through all shades of gray or green, for example, a color shade characterizing a power level. For example a signal reflected by bitumen is returned with little power and can be associated with the black color. A signal reflected by a metallic material is returned with a lot of power and can be associated with the white color.

SAR processing is based on a Doppler signal analysis to achieve very fine resolution along the axis transverse to the direction connecting the phase center of the antenna in the center of the image area. In fact, to get the same resolution only using the only angular selectivity of the antenna, it would require a huge antenna width making it totally unsuitable for airborne applications. An antenna operating in SAR mode is then equivalent to a huge virtual antenna that would work in conventional mode, hence the Anglo-Saxon name of Synthetic Aperture Radar.

SAR radar modes are particularly sensitive to jamming. Indeed to ensure fine resolutions, the radar illumination time and the emitted band are important, leaving enough time for a jammer to generate a noise signal whose spectrum is included in the frequency band used. A jammer is a device for listening and emitting electromagnetic waves. When it receives signals that it recognizes as having been emitted by a radar, it emits signals at the same carrier frequency, which are supposed to interfere with the back-diffusion of the received signals. Thus, the radar receives the signals emitted by the jammer and interprets them as his own retro-diffused impulses. But their digital processing leads to unusable data. For example, in the case of a radar operating in SAR mode and exposed to a jammer, the image is devoid of any contrast, totally monochrome in green for example, not distinguishing any installation on the ground.

To perform anti-scrambling and / or scrambling detection processes, it is necessary to have a scramble noise reference devoid of a useful radar signal. This reference is obtained by recovering the signal contained in a clear zone. A clear area is a signal area devoid of a wanted signal, i.e. a signal area that does not contain backscattered radar signals. If a jammer emits, a clear area is then composed of thermal noise and interference noise. In the absence of a jammer, a clear zone consists only of thermal noise, the statistics of which are well known. Thermal noise is a white noise, Gaussian and known power. Thus, from N _eCh noise vector samples (Xi) measured in clear zone it is not only possible to perform anti-jamming treatments based for the most part on the measurement of an autocorrelation matrix of the noise, but it is also possible to perform jammer detection treatments based on a power threshold. Once the noise measured in a light zone, an autocorrelation of the noise f is simply deduced from the following relation (3): r = -

NOT

-Σ ^χ , ^χ

(3) ech '-

In relation (3), Xj ^H designates the transposed and conjugated vector of Xj. The index i represents the dimension according to which the correlation is estimated. Most of the time, I index the space paths of the radar.

But even if estimating the autocorrelation matrix of the noise is easier in clear area, because it is not necessary to filter backscattered signals, isolate a clear area in which we can measure the N _ΘCh samples of noise vectors (Xi) is not without difficulties in SAR mode.

Conventional methods implemented in the medium or high frequency recurrence radar modes exploit one or the other of the two principles set out below.

The first of these principles consists in deliberately creating clear time zones by interposing passive listening phases between consecutive phases of transmission of a pulse / reception of the signals. Subsequently, the coherent block designates each transmission phase of a pulse / signal reception.

This solution based on listening phases of clear areas interposed between successive coherent blocks can not be transposed to the SAR modes. Indeed, in the high recurrence frequency (HFR) or recurrence frequency (MFR) air / air modes, the consecutive coherent blocks correspond to relatively low illumination times compared to the jammer reaction time. . Thus, even a sophisticated jammer can hardly stop transmitting during the clear-zone listening phase by the radar, during which the radar stops transmitting, and can just as hard to wait for the radar pulse transmission phases. to transmit: the jammer is obliged to transmit even during the listening phase of the light zone by the radar and therefore he is obliged to expose himself to the detection. By contrast, SAR modes are characterized by a low frequency of recurrence. The duration of a single coherent block corresponds to the entire duration of the image taking, this duration of illumination being even greater than the resolution is fine. A clear zone listening phase, whether inserted at the beginning or at the end of the block, then gives information far too far in time with respect to the information used to form the image. In particular, a sophisticated scrambler can exploit this shift to modulate its scrambling signals.

The second of these principles consists in exploiting Doppler light zones naturally present in the part of the Doppler spectrum devoid of a clutter signal.

This solution based on light areas in the Doppler frequency domain is also not transferable to SAR modes. Indeed, in the modes of recurrence frequency of medium to high, the frequency of recurrence is sufficiently high to obtain a clear spectral zone devoid of ground echoes. In contrast, SAR modes are characterized by a low recurrence frequency. It is then difficult to obtain a clear area Doppler frequency, especially for large shooting distances.

The aim of the invention is in particular to overcome the above-mentioned difficulties by highlighting a naturally occurring temporal clear zone in the low-frequency radar modes of recurrence, without it being necessary to interpose a clear zone listening phase of artificial way, and by proposing means to exploit this clear temporal zone. It should be noted that this solution is not applicable to radar modes with medium or high frequency of recurrence, for which this natural clear time zone does not exist. For this purpose, the subject of the invention is a method of filtering, among the electromagnetic signals received by an antenna, signals which do not correspond to the back-scattering of an electromagnetic pulse previously transmitted by the antenna by an obstacle located at an antenna. a known distance. Only the signals received immediately after the transmission of the pulse and before the transmission of the next pulse, during a duration less than or equal to the time required for the first pulse to travel the return distance between the antenna and the obstacle. This is done after each pulse emitted. For example, the obstacle may be the ground.

The antenna may be airborne and the signals retained may be those that do not correspond to back-scattering by the ground. For example, the antenna may emit a SAR waveform.

The invention also relates to a method of detecting a jammer on the ground by an airborne antenna emitting electromagnetic pulses. The method comprises a sampling phase of the signals received by the antenna immediately after the emission of a pulse in a time less than - where h is the altitude of the antenna carrier and cc is the speed of light . It also comprises a phase for estimating the noise power P from the signal samples. A scrambler is detected if the power P is greater than a predefined threshold of thermal noise.

Advantageously, the noise power P can be estimated from N _eCh samples (Xi) of signals received by the antenna immediately after the emission of a pulse by P = ΣX, ² -

¹ cch '= 1

The invention also relates to a method for transmitting and receiving electromagnetic signals for an airborne radar emitting a frequency hopping waveform. Each emission of a pulse at a frequency level, that is to say at a given carrier frequency, is followed by a reception phase of the signals by demodulating at the same frequency for a duration less than or equal to - where h is the altitude of the radar carrier and it is the speed of light.

For example, the emitted waveform may be of SAR type

The main advantages of the invention are that it makes it possible to measure the noise without modifying the transmission / reception cycle of the radar, whether it is a conventional radar or a frequency hopping radar. In particular, frequency of recurrence is not changed. This minimizes the impact on other signal processing performed by the radar. Based on a temporal analysis of the signals throughout the transmission / reception cycle of the radar, the invention also makes it possible to follow the emission cycle of a jammer, whose jamming signals appear and disappear simultaneously in the flow of signals. processed signals.

Other characteristics and advantages of the invention will become apparent with the aid of the following description made with reference to appended drawings which represent:

FIG. 1 schematically illustrates an area imaged by an airborne SAR radar and by a graph the antenna pattern associated with the area;

FIG. 2 illustrates by a graph the method according to the invention inserted in the transmission / reception cycle of a radar,

FIG. 3 illustrates by 3 graphs the application of the method according to the invention to a radar operating in SAR mode with frequency hopping.

Figure 1 illustrates on the one hand schematically an area imaged by an airborne SAR radar and on the other hand graphically the appearance of the power retro-diffused by the ground.

Figure 1 illustrates in a diagram an aircraft 20 carrying a radar operating in SAR mode. At a given instant, the zone imaged in the vertical plane of the aircraft 20 oriented according to its displacement is the area illuminated by the radar beam between an angle of incidence with the antenna θ _mιn and an angle of incidence with the antenna θm _ax . The angular range between θ _mιn and θ _max corresponds to an antenna gain variation of a maximum of 3 decibels around an angle of incidence with the antenna θ ₀ which corresponds to the angle of incidence of the maximum gain of the antenna. the antenna. Thus, the ground being materialized by an axis 22 representing the distance to the ground and whose origin O is the projection of the position of the aircraft 20 on the ground, the imagined zone extends in front of the aircraft 20 between a distance D _mιn ground corresponding to the angle of incidence θ _min and a ground distance D _max corresponding to the angle of incidence θ _max , D _mn and D _max being also represented on the axis 22. FIG. 1 also illustrates by a graph the power of the signal retro-diffused by the ground as a function of the distance to the projected position O of the aircraft 20, in a system of axes where the point O and the axis 22 have simply been reported below the diagram described above for the sake of clarity and where an axis 23 represents the power returned by the ground. A curve 24 represents the power of the backscattered signal and shows a maximum power P ₀ which corresponds to the angle of incidence θ ₀ of antenna maximum gain, itself corresponding to a distance D ₀ with respect to the projected position of the aircraft 20. A major element of the present invention is to remark that no ground target likely to be part of a SAR image can be at a distance from the aircraft 20 which is less than the altitude h of the aircraft 20. In the absence of jammer, the distances between 0 and h are devoid of radar signal and the corresponding distance boxes contain only thermal noise. Thus, a clear zone 21 is highlighted in which the backscattered power is zero. Contrary to what could be envisaged in FIG. 1, the light zone is not located only vertically of the airplane 20: any point in the space situated at a distance from the airplane 20 below its altitude h is part of The corresponding distance boxes can then be dedicated to the calculation of the autocorrelation matrix of the noise or to the detection of jammers. Indeed, in case of backscattering of a signal between 0 and h, it is necessarily a scrambling signal. This is easily detectable by comparing the measured power with a power threshold corresponding to the perfectly known statistical thermal noise.

Thus, the invention proposes to sample signals coming from a distance between 0 and h, where h is the altitude of the airplane 20. It is sufficient to sample signals received immediately after the emission of an impulse in a time less than - where c represents the speed of the light, - then being the time required for a pulse to travel c the distance between the aircraft 20 and the ground. Then it is a question of measuring and recording N _eC noise vector samples (Xj) from these sampled signals. Then, the measured N _eCh samples make it possible to estimate 1 ^ a power P = ^ ^x . If the power P is greater than the threshold of

N _e ch '= 1 thermal noise known elsewhere, it can detect a jammer. Otherwise, this makes it possible to calculate the autocorrelation matrix of the noise f according to the relation (3) and to make the anti-jamming according to a filtering method known elsewhere.

FIG. 2 graphically illustrates how the method according to the invention can fit into the transmission / reception cycle of a radar without disturbing it. Radar pulses 30, 31 and 32 are represented as well as the power of their backscattered signal by curves 33, 34 and 35, respectively, in a system of axes similar to that of FIG. 1, in which the abscissa represents the distance to the ground and the ordinate represents the power of the signal. A distance 36 represents the shooting distance, which is the distance from the antenna to the center of the imaged area. This distance is between the distances D _min and D _max described above. A distance 37 indirectly represents the Doppler recurrence period, which is the time interval between the emission of two successive pulses. Indeed, during the Doppler recurrence period, an electromagnetic pulse travels the distance 37 at the speed of light. In the radar mode shown in FIG. 2, the shooting distance is less than the radar recurrence period. It is said then that the radar mode is unambiguous. Indeed, a ground target of the imaged zone necessarily returns to radar the echo of a pulse before the emission of the next pulse. Thus, on reception of an echo, there is no ambiguity on the reflected pulse, it is necessarily the last emitted. As previously explained, light areas 38, 39 and 40 appear immediately after the transmission of the pulses 30, 31 and 32 respectively. In this radar mode, the invention is therefore limited to sampling and recording the distance boxes of the light areas 38, 39 and 40: immediately after the emission of a pulse, the signals received in a time interval less than the time required for an electromagnetic wave to travel the return distance between the carrier and the ground are recorded. The distance boxes thus recorded make it possible to obtain the desired clear areas. FIG. 3 schematically illustrates how the method according to the invention can be applied to an ambiguous radar mode, in which the shooting distance can be greater than the radar recurrence period. Thus, on reception of an echo, there is possible ambiguity on the reflected pulse: it is not necessarily the last impulse emitted but perhaps a previous impulse.

FIG. 3 illustrates, for example, the application of the method according to the invention to a radar operating in SAR mode with frequency hopping. In a system of axes similar to that of Figures 2 and 3, the abscissa representing the ground distance and the ordinate representing the power of the signal, the graph 71 illustrates the radar transmission cycle. In the same axis system, graph 62 illustrates the reception cycle of the radar when operating in conventional frequency hopping mode. Finally, again in the same axis system, the graph 63 illustrates the reception cycle of the radar when operating in frequency hopping mode according to the invention.

Graph 71 illustrates the transmission cycle of a radar operating in frequency hopping SAR mode. Radar pulses 50, 51 and 52 are emitted, the power of their backscattered signal being represented by curves 53, 54 and 55 respectively. Indeed, in SAR mode, to increase the accuracy in distance it is necessary to increase the bandwidth transmitted around a carrier frequency. But this is difficult to achieve with current radars. Therefore, the bandwidth required to achieve the desired distance accuracy is divided into adjacent frequency subbands, each subband being centered on a clean carrier frequency. Such a mode of operation is called a frequency hopping mode. But it is physically impossible for a radar to emit subbands that exactly follow each other in time. The subbands are necessarily time separated and can only be virtually connected by appropriate digital processing. The three pulses 50, 51 and 52 constitute the division into sub-bands of the frequency band necessary to obtain the desired accuracy in distance. Thus, the three pulses 50, 51 and 52 are re-transmitted cyclically at regular intervals and a distance 61 indirectly represents the Doppler recurrence period, which is the time interval between the emission of 2 pulses of the same frequency. Over the Doppler recurrence period, the entire frequency band required for the distance accuracy is emitted by the radar. The inevitable temporal spacing between the subbands is materialized by the spacing between the three pulses 50, 51 and 52. Zones 56, 73 and 72 represent the zones imaged by the pulses 50, 51 and 52 respectively, a distance 60 representing the shooting distance of each pulse. It appears that the area imaged by a pulse at a certain frequency is interposed between the emission of other pulses at different frequencies. This explains the particular reception cycle of a frequency hopping radar, this cycle being detailed below.

Figure 62 illustrates the reception cycle of the same SAR radar when operating in a conventional frequency hopping mode. Immediately after the emission of a pulse at a certain frequency, the radar changes frequency in order to demodulate the signal received and corresponding to the back-diffusion of the previous pulse. Thus, a phase 65 immediately after the pulse 51 represents a demodulation phase at the frequency of the pulse 50, a phase 66 immediately after the pulse 52 represents a demodulation phase at the frequency of the pulse 51 and a phase 67 immediately after the pulse 50 represents a demodulation phase at the frequency of the pulse 52. A duration 64 represents the transmission time of a pulse at a certain frequency and demodulation at reception at the frequency of the pulse previous. However, according to what is explained above, light areas 57, 58 and 59 shown in graph 71 and corresponding to pulses 50, 51 and 52, respectively, occur naturally immediately after the emission of each of these pulses. More precisely, the zone 57 is a clear zone for the frequency of the pulse 50, the zone 58 is a clear zone for the frequency of the pulse 51 and the zone 59 is a clear zone for the frequency of the pulse 52 Indeed, in light zone 58 the radar still receives the signals represented by the curve 53 corresponding to the backscattering of the pulse 50. This is why the radar demodulates at the frequency of the pulse 50 in the light zone 58. In light zone 59 the radar receives the signals represented by curve 54 corresponding to the back-diffusion of the pulse 51. This is why the radar demodulates at the frequency of the pulse 51 in the light zone 59. And, although it is not shown in FIG. In the clear zone 57, the radar receives the signals represented by the curve 55 corresponding to the back-scattering of the pulse 52. This is why the radar demodulates at the frequency of the pulse 52 in the light zone 57. Therefore, a radar operating in a conventional frequency hopping mode is not able to exploit the light areas highlighted by the invention, because in these clear areas it demodulates in reception at a different frequency from those corresponding to the clear areas.

The graph 63 illustrates the reception cycle of the same SAR radar but when it operates in a frequency hopping mode capable of exploiting the light areas highlighted by the invention. This mode of operation is part of the present invention. It involves cutting the demodulation phases at a single frequency in reception in two demodulation phases at two different frequencies, the first being the frequency of the light zone. Thus the demodulation phase 65 at the frequency of the pulse 50 is preceded by a brief demodulation phase 68 at the frequency of the pulse 51. Likewise, the demodulation phase 66 at the frequency of the pulse 51 is preceded by a short demodulation phase 69 at the frequency of the pulse 52 and the demodulation phase 67 at the frequency of the pulse 52 is preceded by a brief demodulation phase 70 at the frequency of the pulse 50. The duration of the phases 68, 69 and 70 is chosen so that they cover only the light areas as described above, that is to say so that during this time a pulse can not travel more than the distance equivalent to the altitude of the carrier in return. For example, in the case of airplane 20 in FIG. 1 flying at altitude h, this duration may be equal to or less than - where c is the speed of light. It is important to note that a backscattered signal received during a clear area does not disturb this clear area. Indeed, the back-scattered signal is received at a frequency different from the frequency for which the zone is clear. Thus, the demodulation phase 68 may for example be used for the anti-scrambling of the signal 73 resulting from the backscattering of the pulse 51 as previously stated. Another advantage of the present invention, in particular applied to a frequency hopping radar mode, is that it does not allow a very sophisticated jammer to distinguish between the transmission phase and the light-zone listening phase of the transmission. radar, the light zone listening phase does not exist as such. Such an intelligent jammer can not wait for the transmit phase of the radar to transmit, transmitting and listening to clear area being closely nested. Such a jammer can not escape the detection and bypass the anti-jamming.

Claims (7)

. A method of filtering, among the electromagnetic signals received by an antenna (24, 33, 34, 35), signals not corresponding to the backscattering of an electromagnetic pulse previously transmitted by the antenna (30, 31, 32) by a obstacle situated at a known distance, characterized in that only the signals received immediately after the emission of the pulse and before the emission of the next pulse are retained for a duration (21) less than or equal to the time required at the first impulse to travel the distance back and forth between the antenna and the obstacle. 2. The filtering method according to claim 1, characterized in that the antenna is airborne and the signals retained are those that do not correspond to the retro-diffusion by the ground (22). 3. Filtering method according to claim 2, characterized in that the antenna emits a waveform of SAR type. 4. A method of detecting a jammer on the ground by an airborne antenna emitting electromagnetic pulses (30, 31, 32), characterized in that it comprises: a sampling phase of the signals received by the antenna immediately after the emission of a pulse in a time period less than Where h is the altitude of the carrier (20) of the antenna and c is the speed of light; a phase for estimating the noise power P from the signal samples; a scrambler being detected if the power P is greater than a predefined threshold of thermal noise. A method for detecting a jammer according to claim 4, characterized in that the noise power P is defined from N _eCh samples (Xi) of signals received by the antenna immediately after the emission of a pulse by P = ΣX, ² - ^ ech '= 1 A method for transmitting and receiving electromagnetic signals for an airborne radar transmitting a frequency hopping waveform, characterized in that each transmission of a pulse (50, 51, 52) at a frequency level is followed a reception phase of the signals (70, 68, 69) by demodulating at the same frequency level for a duration less than or equal to - where h is the altitude of the carrier (20) of the radar and cc is the speed of light . 7. A method of transmitting and receiving electromagnetic signals according to claim 6, characterized in that the transmitted waveform is of the SAR type.