The present invention relates to the field of sensors interception of radio broadcasts, whether for civil or military applications.

The document US 2006/058035 A1 discloses a communication signal receiver, which can detect the presence of a radar signal in a received signal.

In the civilian world, the field of interception of radio emission sensors corresponds to control of the radio frequency spectrum can be achieved by frequency regulators, to control the frequency allocations and to detect, locate and identify the radio emissions not -autorisées or disruptive.

In the military, this domain is monitoring functions and listening radar emissions and / or radio to electronic warfare purposes (intelligence function, position keeping function, warning function vis-à screw threatening emissions).

The radio transmissions are classified into two types: communication or COM emissions in this (radio, television, mobile telephony, microwave links, satellite communications, tactical military radio ...) and the RAD or radar emissions following (surveillance, navigation, guidance, fire control, ...)

Because, firstly, their spectral regions disjoint and, secondly, their different temporal modulations techniques implemented to intercept and analyze radio signals RAD and COM are adapted to the nature and forms of RAD and COM wave and led to the realization of sensors dedicated to the interception RAD and dedicated to the COM interception sensors.

In addition, this interception requires the implementation of several sensors RAD or more COM sensors to cover all or part of the spectral domain with the HF, VHF, UHF and SHF.

However, the spectral and temporal separation of these two types of signals has declined over time: in fact, the increase in processing capabilities, especially those of the computers has enabled to operate more waveforms complex even broadband, which was previously impossible. This has opened new possibilities for communication programs (increased bandwidth and therefore flow rate) and radar emissions (more complex waveforms thus better protected from interference and to extract information of target intercepted).

Accordingly, a pickup sensor receives a mixture of COM signals and signals RAD. If it is a RAD sensor, COM signals are interference signals which must be removed for analysis of the RAD signals and, conversely, if it is a COM sensor, the RAD signals constitute signals parasites that must be removed to analyze COM signals. The absence of spurious signals of removal treatments disturb or degrade the useful signal and therefore the quality of the analysis thereof.

In addition, we often seek to characterize the radio emission of whatever origin (COM or RAD) which involves the use of two sensors (RAD and a COM sensor sensor) on the same platform (vehicle terrestrial, aircraft, surface ships ...).

But this increases the complexity of integrating the platform. Hardware resources (antennas, RF receivers and processing units) are separated, this leads to an increase of mass, volume and power consumption, as well as an integration of antenna arrays of the FDR sensor and sensor COM problematic on the same platform.

Moreover, the vast expanse of the spectral range to be covered involves the use of multiple antenna arrays for each sensor and each RAD COM sensor implemented.

In conclusion, the juxtaposition of a COM sensor and a sensor RAD is less and less relevant in terms of both performance as integration constraints on the platform.

There is therefore a need for an interception of radio emission sensor to achieve the desired integration.

For this the invention relates to a sensor intercept radio signals according to claims.

The invention consists of a sensor RAD / COM integrated simultaneously treating emissions communication and radar emissions, the architecture enables a pooling of maximum resources from the receiving antennas to the primary processing means, only the means for secondary processing being dedicated to be the processing of COM signals is in the processing of signals RAD.

The RAD / COM integrated sensor according to the invention offers the possibility to control the radio frequency spectrum and / or intercept both radar and communication transmitters as multifunction weapon systems in electromagnetic environments increasingly dense tangle with radar bands and communication.

RAD sensor / integrated COM according to the invention has the further advantage of being modular and a more compact solution that juxtaposing a receiving sensor RAD COM and a receiving sensor.

The invention and its advantages will be better understood on reading the detailed description that follows of a particular embodiment, given solely by way of example, this description being made with reference to the accompanying drawings in which:

Figure 1 is a schematic representation of sensor usage context RAD / COM integrated according to the invention;

- Figure 2 is a schematic representation in block form and high-level architecture of the FDR sensor / integrated COM according to the invention;

Figure 3 is a schematic representation in block form of the sensor of the receiver stage RAD / COM integrated Figure 2;

Figure 4 is a schematic representation in block form of the primary stage of analysis treatments sensor RAD / COM integrated Figure 2;

Figure 5 is a schematic representation in block form of the stage of secondary processing of the FDR sensor / COM integrated Figure 2;

Figure 6 is a time-frequency graph illustrating the different segmentations performed in parallel by the stage of primary treatment of Figure 4; and,

Figure 7 is a graph instantaneous time-frequency-band illustrating the various patterns used by the discriminating means of the floor of primary treatments of Figure 4.

As shown in Figure 1, the sensor 10 intercept the radio transmissions according to the invention, also known as RAD / COM sensor integrated in the following, receives and analyzes the radio signals transmitted by different transmitters, they correspond to radar emissions (emission RAD) or communication emissions (COM emissions).

10 RAD / integrated COM sensor presents an architecture for performing a primary analysis of the radiofrequency signals received independently of the nature of the signals (RAD or COM emissions) and then continue this analysis (secondary analysis) by two types of treatments, a dedicated emissions RAD and another dedicated to COM emissions. In this way the number of common components between RAD and COM processing chains is maximized, particularly the acquisition hardware components and computing resources to the primary analysis. This rationalization of material present the advantage of reducing the bulk of the integrated sensor and to facilitate its installation on a wearer.

As shown in Figure 2, the sensor 10 includes a receiver stage 100. This is a stage of pooling receiving material resources (antennas, radio frequency receivers, analog to digital conversions and digital processing).

The sensor 10 comprises a stage of treatment 200 primary analysis for the primary analysis of the digital signals delivered by the receiver stage 100. Several treatments are applied to the signals regardless of their nature (RAD emission or emission COM) and waveform (continuous or pulse, narrowband or broadband). The primary analysis, independent of the nature of the intercepted emission permits for each elementary signal to characterize, including its duration, arrival (TOA or "Time Of Arrival" in English), its instantaneous band, its center frequency, amplitude and possibly its direction of arrival (DOA or "Direction of arrival" in English). The last treatment carried out at the stage 200 is, from the characteristics of the basic signal and a priori knowledge, to estimate the type (RAD, COM, or undetermined) the emitter that generated the signal.

The sensor 10 comprises, downstream of the stage 200 of pooling a number of treatments, a stage 300 of secondary analysis treatments. The stage 300 includes one or more processing chains 400 of the signal and / or data signals dedicated to COM and one or more channels 500 to signal processing and / or data signals dedicated to RAD. The latter treatment may be carried on the floor 300 is a melting process, performed by a merge module 600, data between the outputs of the different channels, as well as COM 400 RAD 500, not outputted only consolidated information.

The architecture of the sensor 10 will now be presented in more detail, with reference to Figures 3 to 5.

RECEIVING FLOOR 100

As shown schematically in Figure 3, the floor 100 pools the physical resources required for the interception of both emissions and RAD COM emissions.

The stage 100 comprises one or more antenna arrays 1 10. Each network comprises at least one antenna.

The stage 100 includes antenna arrays NR 1 10. Each antenna array comprises antenna elements NVA. Each antenna converts incident electromagnetic signals into electrical signals.

In the preferred embodiment, each array is an antenna array in polarization diversity by frequency subrange. The antennas of the same antenna array are thus divided into two groups: the antennas of a first group are adapted to a polarization antennas and a second group are adapted to the orthogonal polarization from the first group. For example, if the antennas of the first group are adapted to horizontal polarization, the antennas of the second group will be adapted to a vertical polarization. As a further example, if the antennas of the first group are adapted to a right circularly polarized antennas of the second group will be adapted to a left circular polarization. This enables receiving electromagnetic signals irrespective of their polarization.

The number of networks is a function of the desired frequency coverage for the sensor 10. Each antenna array is associated with a particular frequency domain.

NR network number of antennas, the number of NVA network antennas as well as the geometry of the antenna array are selected by the skilled person depending on the compromise between the integration constraints antenna arrays on platform, masks between antennas, the reception sensitivity, cost, and when the characterization of radio signals requires an estimation of the arrival direction, accuracy and protection ambiguities of estimating the direction arrival.

NR x NVA electrical signals are outputted from all of the sensor networks.

The stage 100 includes a switch antenna 120, common to the different networks of antennas 1. 10

The antenna switch 120 takes as input the electrical signals from the antennas of each NVA NR networks. The antenna switch 120 selects an antenna array through the NR antenna arrays and outputs the electrical signals NVA antennas of the selected antenna array.

The antenna switch 120 allows to inject adjustment signals in place of the signals from different antennas.

The stage 100 finally comprises a radio frequency receiver 130 to NVA reception channels. The radio frequency receiver 130 takes as input the electrical signals NVA switch output antenna 120, that is to say from the NVA antennas of the selected network.

Preferably, it is a synchronous multichannel radio frequency receiver.

The radio receiver 130 converts the electrical signals within a BPI instantaneous bandwidth located around a radio frequency receiver in which the receiver tuning frequency 130 is set dynamically. BPI band of the receiver 130 and the multi-tone instantaneous dynamic that results must be adapted to the environment to observe.

The signal obtained at the output of each receiving channel is digitized after amplification, filtering and, according to the reception frequency, of one or more frequency transpositions or not.

The radio receiver 130 outputs NVA synchronous digital signals corresponding to the BPI band and the receiver tuning frequency.

STAGE TREATMENT OF PRIMARY ANALYSIS 200

The floor of primary analysis processing 200 is a digital signal processing chain, carrying out primary analysis of the elementary signals received in the BPI band regardless of type, that is to say no assumptions about the nature of this signal.

In the embodiment shown in Figure 4, the floor 200 includes a module 210 to time-frequency segmentation of the signal for each of NVA synchronous reception channels of the radio frequency receiver 130.

When executed, the module 210 takes as input the digital signal of the selected channel.

The digital signal is cut synchronously with other channels in time and frequency. This division is for example performed using a filter bank or a Fourier transform.

The module 210 segments the signal with at least two time-frequency analysis of resolutions implemented simultaneously. In Figure 6, which is a time representation T - frequency F n resolutions {Ti, Fi} are simultaneously used to segment the field playing time - instantaneous bandwidth BPI centered around the frequency fo. This allows the sensor 10 to be adapted to all possible waveforms.

Resolutions frequency analysis used may change from BPI band to another in order to be most appropriate as possible to the waveforms of communications programs and radar transmissions may be intercepted.

Par exemple, on peut choisir, in the VHF domaine d'utiliser la-resolution temps fréquence {800 με, 12.5 kHz}, 212 houses in the figures 6, conjointement a resolution-temps fréquence {10 με, 1 MHz }, 214 houses in the figure 6.

To facilitate the implementation of the association step (performed by the module 240 shown below), the frequency resolutions must be integers from smallest multiple frequency resolution and temporal resolutions must be integer multiples of the smaller temporal resolution.

The stage 200 then includes a module 220 for detection and enumeration.

When executed, this module carries out detection in each of the time-frequency cells are to determine the presence or absence of at least one useful signal more or less buried in noise.

The detection may be performed conventionally by an estimate of the covariance matrix calculated on the NVA channels and comparison of eigenvalues ​​of the covariance matrix obtained with a detection threshold.

The detection threshold typically corresponds to a signal to noise ratio selected to obtain a detection probability and a probability of false alarm desired. To determine this threshold, it is necessary to estimate the noise level.

Compared to conventional solutions, the estimate of the noise level here is advantageously carried out by averaging, for all boxes distributed in the instantaneous bandwidth BPI with a slice of time considered, the lowest values, on the assumption that all of the boxes will not be occupied by a useful signal.

This principle allows for a more accurate estimate of the noise level and also increase the capacity count of the number of co-channel sources (ie the number of useful signals in the same square time- frequency) compared to methods that seek to differentiate the eigenvalues ​​useful vis-à-vis the eigenvalues ​​of the noise signals in a single channel.

Alternatively, another method of detection may be performed by the module 220, of comparing the output level energy beam formation NVA channels compared to a detection threshold or by comparing the trace the covariance matrix of each cell compared with a detection threshold.

In yet another embodiment, the detection can be performed by thresholding the energy integration of neighboring squares (incoherent integration on a time / frequency domain) when the cost calculation of covariance matrices is too important to be systematically .

The stage 200 then includes a module 230 for grouping detection and characterization of these groupings.

When executed, the module 230 estimates the primary features of the detections from the covariance matrix. These primary features are for example the date of arrival, direction of arrival, polarization level, the band, the duration, the frequency, the signal to noise ratio, etc. of the detected signal.

In particular, the treatment of the arrival direction estimation for each detection exploits the diversity of polarization antenna arrays to separate co-channel sources (concurrent issue in analysis time-frequency interval), using conventional algorithms.

Alternatively, the processing for estimating the co-channel source direction of arrival is replaced by an estimation processing of the single-source direction of arrival.

Then the module 230 combines the detected signals into an object of synthesis. For a COM signal, such an object is known under the name of plot, but for a signal RAD, such an object is known under the name pulse.

The module 230 aggregates all detections from a single segmented signal when they are contiguous in time and / or frequency and may belong to the same basic electromagnetic signal. For this estimation similarities between neighboring detected signals is performed, for example by calculating a Mahalanobis distance on the primary characteristics, in particular the direction of arrival and polarization.

Alternatively, the estimation of the arrival direction is suppressed and the combination of the detected signals is performed only on proximity criteria time, frequency, and optionally level.

The synthesis of object then summarizes the characteristics of groups either through a melting characteristics of each of the detected signals, or via a new estimate of the characteristics of the time-frequency area representing the combination of the detected signals.

The stage 200 then includes a 240 combination module to choose the best object of synthesis, representing a useful signal of the n time-frequency analysis of resolutions implemented.

When executed, the module 240 takes as input the synthetic objects issued by the module 230 and compares the concurrent objects in time and frequency for holding the m objects providing a summary of the incident electromagnetic signal with the best signal to noise ratio , that is to say with the resolution most suitable time-frequency analysis of the incident electromagnetic signal. M the selected objects are forwarded to the next module for estimating signal.

The stage 200 then includes a module 250 for estimating the IQ signal.

When executed, the module 250 takes as input each of the m objects synthesis representing m detected useful signals, and the individual signals from the channels of the radio frequency receiver 130.

The module 250 isolates in the BPI band, the elementary signal corresponding to each of the m objects. The isolation of each element signal is performed by conventional procedures of filtering / decimation frequency, possibly supplemented by spatial filtering treatments, particularly through formation.

This step outputs, for each object of synthesis, the object of synthesis together with the corresponding baseband signal, called signal IQ signal.

The stage 200 then includes a module 260 to characterize the signal.

When executed, the module enables 260, from the IQ of each object signal to refine the primary characteristics of the elementary signal developed by the module 230, including the estimated arrival date, duration signal and the signal band. The estimated date of arrival, the duration and the band is performed using conventional methods.

The stage 200 includes a module 270 then discrimination signals.

When executed, the module 270 label from primary characteristics, each tone associated with synthetic objects either COM transmitting or broadcasting RAD or indeterminate emission when the signal is unknown or COM is an ambiguity between emission and RAD program.

At the output, the needle unit 270 to the secondary processing suited. Thus, an object of synthesis, with the signal IQ, classified as qu'émission COM or indefinite emission is oriented 400 secondary analysis processing adapted to communication channel emissions while an object of synthesis, with his IQ signal, classified as qu'émission RAD or indefinite emission is directed towards the channel 500 of secondary analysis processing adapted to the radar emissions.

To perform this discrimination between COM signals and signals RAD, the unit 270 compares the estimated time, frequency and modulation band of elementary signals received vis-à-vis duration templates transmit / transmit frequency / modulation band.

As shown in Figure 7, different templates COM or COM and RAD RAD are predefined and stored in a database 280. These templates synthesize a priori knowledge of waveform COM and waveforms RAD.

The development of templates is based on a data model for describing, according to common parameters, the elementary signals, both the communication programs that radar emissions.

Thus, the primary analysis processing implemented correspond to the detection of treatments, finding and estimation of the signal. These treatments are performed with several simultaneous time-frequency analysis of resolutions, depending on each frequency range, in order to be adapted to all COM waveforms and RAD that these waveforms are continuous (CW COM , FMCW radar, ...) or pulse (bearing EVF, TDMA burst, radar pulses ...), narrow band or broadband. The primary analysis treatments also have a capacity of separating concurrent sources or co-channel to analyze the overlapping emissions. Estimates of elementary signals carried by these primary analysis treatment (duration, bandwidth, level, direction of arrival, polarization, extraction of the signal itself ...) are oriented towards the secondary analysis suitable treatment primary function estimates and prior knowledge about the radio broadcasts.

FLOOR 300 SECONDARY TREATMENT

The output of the stage 200 of primary analysis treatments to an integrated analysis of signals regardless of their type, the floor 300 of secondary analysis processing includes, as shown schematically in Figure 5, a chain 400 dedicated COM emissions analysis and a chain 500 dedicated to the analysis of RAD programming.

Each of these channels 400, 500 performs conventional treatments that are found in the COM receiving sensors and the receiving sensors RAD respectively.

The channel 400 and includes:

A module 410 COM systematic identification technology;

- A module 420 technical identification on request;

430 a deinterleaving module;

A module 440 for tracking a transmitting source; and,

A module 450 for locating the source COM tracked.

The channel 500 and includes:

- A module 510 Intrapulse analysis;

520 a deinterleaving module;

530 a tracking module of a transmission source;

A module identification 540 technical RAD tracked source; and a module 550 for locating the source RAD tracked.

The location module 450 and the location module 550 can be advantageously shared. This or these tracking units used to estimate the position of the COM and RAD transmitters and can use conventional localization techniques such as, for example, location by scrolling when the supporting platform of a sensor 10 is movable relative to sources of fixed signals or location by triangulation using multiple 10 networked sensors.

Technical identification of a source of emission establish a list of issuers candidates by comparing the analyzed technical characteristics of said transmission source and specifications of emissions that can be encountered in use and stored in a database such a database based on a modeling of communications waveforms and modeling of radar waveforms. Thus the technical identification RAD led to a list of modes and sub-modes RAD emissions and technical identification leads to COM COM transmission modes list. For military applications, these modes and RAD emission sub-modes and COM transmission modes can be attached to a probable list of issuers, platforms or known weapon systems.

In the RAD / COM integrated sensor 10, the secondary treatment is adapted to provide the characteristics of the tracks 600 useful to a fusion module for, when executed, to produce a consolidated information on each intercepted transmission, including information its nature (radar transmitter or transceiver radio) and its characteristics.

Generally, the melting operation for synthesizing in a meta-object, different types of data respectively from the RAD and COM treatments. This also eliminates any redundancies when the elementary signals were referred by the module 270 to both the 400 secondary analysis treatment chain adapted to communication issues and the chain of 500 secondary analysis of treatments tailored to radar emissions.

For example, if an elementary signal sent to channels 400 and 500 for secondary treatment analysis led to the technical identification of a transmission mode COM and no technical identification of a RAD mode, or sub-mode, then the track after the workflow RAD will be eliminated.

For example even if the location of the source and the RAD COM source are consistent geographically and / or temporally and if the intersection of lists of modes or sub-modes RAD and COM transmission modes identifies a platform common form, then the data from each processing chain are grouped to not present the operator a high-level object having an operating direction for the latter.

It should be noted that with the sensor 10, the frequency coverage of interception and measurement accuracy of the arrival direction are identical for radar transmissions and for communication emissions.

The sensor 10 comprises at least one calculation unit comprising calculation means such as a processor, and memory means such as RAM and ROM memories, the memory means storing the instructions of specific computer programs execution by the computing means.

In particular, the modules of the stages of primary or secondary treatments analysis just described correspond to clean computer program instructions to be executed by the calculation means. When called, a module allows the implementation of a corresponding step.

Thus, what has been described above in terms of device could also be described in terms of process, including method of treating primary analysis and secondary analysis treatment process. Preferably, these programs are executed in real time on the signals delivered by the receiving stage.

ADVANTAGES

The advantages of the sensor presented above with respect to conventional solutions juxtaposing COM sensors and RAD sensors are numerous.

The sensor 10 allows a reduction in necessary resources, the latter being streamlined and pooled to perform a reception and primary characterization of elementary signals, regardless of their nature.

The pooling of resources and treatment requires taking into account the diversity of characteristics of radar emissions and radio broadcasts:

Diversity polarization: vertical polarization, horizontal polarization, circular polarization ...

Diversity waveforms: waveform continuous (FMCW radar ...) or pulse (EVF bearing TDMA burst, pulse radar ...).

Variation of the band of the waveform: narrowband or broadband.

In addition, various origins signals intercept makes these polymorphic in the sense of their spectral and temporal coverage. This requires that the primary analysis of the signal is able to adapt to any form of signal.

The reduced requirements facilitates integration of the sensor 10 on a single platform reducing mass, volume and power consumption.

The sensor 10 allows a better recognition of the complexity of the radio environment: the elementary signals extracted from the juxtaposition and tangle communications emissions and radar transmissions are separated more efficiently.

Moreover primary RAD / COM integrated analysis allows, from the primary characteristics of individual signals and a priori knowledge, to guide these elementary signals to the COM treatments and treatments RAD most suitable. This allows these specific secondary treatments or RAD COM not to be disturbed by the presence of COM or RAD signals, respectively.

The monitoring time of an interception (also called "tracking Issuer") has to deal with a large number of ambiguities due to the presence of many similar issuers in the spectrum. This requires treatment of duty make the most of the specifics of each intercepted signal. It follows that it is necessary that the sensor has a secondary analysis to a specific data processing to monitoring ARD programs and other specific monitoring of emissions COM.

This sensor is particularly suitable for the development of radio emissions. Indeed, there is no a priori assumptions about the waveforms intercepted (only RAD or COM only). The sensor 10 is able to process all radar and radio broadcasts, as well as current coming. Issuers will seek to optimize spectral efficiency (allocation of frequency allocation by regulators increasingly difficult because of the growing needs of communication channels), duration (for privacy needs, and sharing niche) that the peak power emitted (spreading time / frequency).

WE CLAIM

1.sensor (10) for intercepting radio signals, characterized in that it is adapted to analyze radar emissions (RAD) and communication emissions (COM), and in that it comprises:

- a stage (100) for reception, and able to scan a radio signal incident;

- a stage (200) of primary common analytical treatment and own preprocessing the digitized signal to determine a plurality of primary features of the incident radio signal; and,

- a stage (300) of secondary analysis treatment, comprising a chain (400) of digitized signal analysis processing pretreated dedicated to communication emissions (COM) and a chain (500) of analytical processing of the digitized signal pretreated dedicated to radar emissions (RAD)

the floor (200) of primary analysis comprising a treatment module (270) of own discrimination applying the digitized pre-processed signal in the channel inlet (400) dedicated to communication emissions (COM) and / or the input of the chain (500) dedicated to radar transmissions (RAD), according to the primary characteristics determined for the radio signal incident.

2.sensor (10) according to claim 1, wherein the receiver stage comprises:

- a plurality of antenna arrays (1 10), each antenna array comprising at least one antenna and being associated with a particular frequency domain;

- an antenna switch (120) for selecting the electrical signals generated by an array of antennas of the plurality of antenna arrays; and,

- a radiofrequency receiver (130) of which the number of lanes is equal to the number of antennas of the selected antenna array, each channel being capable of digitizing the electrical signal output from a corresponding antenna of the selected antenna array, to preferably the receiver is synchronous.

3. - sensor (10) according to claim 2, wherein each antenna array is an array of polarization diversity antennas by frequency subrange.

4. - A sensor according to claim 2 or claim 3, wherein the floor (200) of primary analysis treatment comprises a module (210) of time-frequency segmentation of a digital signal for each channel of the radio receiver (130), preferably the module (210) segmentation being adapted to segment the digital signal according to at least two different time-frequency analysis of resolutions, preferably the time-frequency analysis of resolutions being variable according to the ranges frequency.

5. - A sensor according to claim 4, wherein the floor (200) of primary analysis treatment comprises a module (220) of own detection to verify the presence of at least one useful signal mixed with noise in each of the time-frequency squares of the or each segmented signal output by the module (210) of time-frequency segmentation, preferably the noise being estimated by determination of the lowest eigenvalues ​​of the covariance matrix on all cells of time-frequency analysis of the corresponding segmented signal.

6. - The sensor of claim 5, wherein the floor (200) of primary analysis treatment comprises a module (230) of grouping own detections to estimate, for each detection, a plurality of primary features from a date arrival, a direction of arrival, polarization, a level, a tape, a period, a frequency, a signal to noise ratio of the radio signal corresponding to the incident considered detection and aggregate detections together to form an object synthesis, for the implementation of adjacency criteria in time and / or frequency and / or similarity according to the values ​​of one or more primary features.

7. - A sensor according to claim 6, wherein the floor (200) of primary analysis treatment comprises a module (240) association to choose the best object of synthesis from synthetic objects derived from each of the signals segmented output of the module (210).

8. - A sensor according to claim 4, wherein the floor (200) of primary analysis treatment comprises a module (250) for estimating the IQ signal associated with the best object of synthesis, the IQ signal is isolated from the digital signal of the corresponding channel of the radio frequency receiver (130) by filtering / decimation and possibly by spatial filtering.

9. - A sensor according to any one of claims 1 to 8, wherein the module (270) of discrimination determines what channel transmitting the preprocessed signal, based on the primary characteristics and a priori knowledge on the forms of communication wave emissions (COM) and the waveforms of the radar transmissions (RAD).

10. - sensor (10) according to claim 9, wherein the a priori knowledge takes the form of a plurality of templates transmission time / frequency transmission / elementary band signal modulations.

January 1. - sensor (10) according to any one of claims 1 to 10, wherein the stage (300) of analysis of secondary treatment comprises a module (600) of own fusion to generate a consolidated information on each program cleared from the outputs of the chain (400) dedicated to communication emissions (COM) and the chain (500) dedicated to radar transmissions (RAD), in particular by allowing the removal of any redundancy when the elementary signals were referred by the module (270) discrimination towards both the chain (400) dedicated to communication issues (COM) and the chain (500) dedicated to radar emissions (RAD).