Communication device and detection method using said communication device

Abstract

The invention relates to a detection method using a communication device, comprising the following steps: - making an unambiguous estimation (E0) of a parameter representative of a position or a movement of at least one object; - making an ambiguous estimation (E2) of the representative parameter, using at least one reference signal having a comb-like structure; - comparing (E3) at least one value of the representative parameter obtained by the ambiguous estimation and at least one value of the representative parameter obtained by a previous estimation of the representative parameter; - based on the comparison, selecting (E4) at least one value obtained during the step of ambiguous estimation.

Description

[0001] DESCRIPTION

[0002] Title of the invention: Communication device and method of detection by said communication device

[0003] 1. Scope of the invention

[0004] The invention relates to the field of communication systems integrating radio detection functionalities, also known as ISAC systems (for "Integrated Sensing and Communication" in English) or integrated communication and detection systems.

[0005] 2. Prior art

[0006] The development of techniques aimed at merging communication and radio sensing functionalities into a single system (Integrated Sensing and Communication system) offers particularly interesting prospects in the field of wireless networks. More specifically, "integrated radio sensing and communication" refers to using radio waves transmitted within a communication system not only to establish communication between devices, but also to obtain information about the environment surrounding those devices. Radio sensing, for example, similarly to radar, involves obtaining information about the presence of objects (in the broadest sense: humans, animals, plants, vehicles, buildings, etc.).) in the environment of a piece of equipment and to determine characteristics associated with these objects (such as, for example, their positions, whether they are static or moving, their speeds of movement, possibly their shapes, etc.). One of the primary objectives of ISAC technology is to improve spectral and energy efficiency by using a single waveform for the simultaneous implementation of these two functionalities. Such integration is particularly relevant in the context of the development and deployment of new generations of wireless networks (notably 6G) designed to meet growing needs in terms of both high-speed communication and detection performance (for example, object detection).

[0007] Many modern wireless networks rely on orthogonal frequency-division multiple access (OFDM) techniques, which have the advantage of efficiently using the available frequency spectrum while remaining resistant to interference during data transmission.

[0008] Several methods are applicable in the context of detection when OFDM is used. One method involves using reference signals. These are predefined signals used in communication for the purpose of detection. Reference signals are of interest for detection applications due to their advantageous characteristics, including high passive detection performance, robust noise resistance, and good autocorrelation properties.

[0009] One problem with reference signals used for radio detection is their comb structure, meaning their periodic time and frequency structure, which can be adjusted by the comb size. The presence of empty subcarriers or resource elements in the reference structure reduces the unambiguous detectable range and introduces ambiguity. When estimating distance or velocity, these empty subcarriers create false peaks that lead to false detections. This results in false positives or phantom targets: i.e., the detection of a target that does not exist.

[0010] Prior art attempts to resolve this ambiguity problem, for example, by using a combination of different types of reference signals ([Khosroshahi 2024]). This solution is unsatisfactory because it introduces design complexity. Indeed, the use of multiple reference signals introduces challenges such as managing parameter configurations, coordinating different reference signals, and assigning these signals, all of which increase the system's design and computational load.

[0011] The present invention aims to provide a simple solution to reduce the ambiguity of communication systems incorporating detection functionalities.

[0012] 3. Description of the invention

[0013] The invention proposes a detection method using a communication device that includes an unambiguous estimation of a parameter representative of the position or movement of at least one object. The method further comprises the following steps:

[0014] - ambiguous estimation of said representative parameter using at least one reference signal having a comb structure;

[0015] - comparison between at least one value of the representative parameter obtained by said ambiguous estimation and at least one value of the representative parameter obtained by a previous estimation of said representative parameter;

[0016] selection of at least one value obtained in said estimation with ambiguity from said comparison.

[0017] The previous estimate with which the comparison is made corresponds to an unambiguous estimate or to an ambiguous estimate.

[0018] Unambiguous estimation allows for obtaining at least one accurate estimate, meaning an estimate in which the risk of detecting phantom targets or fewer objects than actually present in the environment is reduced or even eliminated. In particular, the ambiguity due to the comb structure of the reference signals (i.e., the presence of empty subcarriers in the reference structure) and the resulting detection errors are eliminated. Once this accurate estimate is obtained, the method according to the invention allows ambiguity to occur. The unambiguous estimate then serves as a reference for comparing ambiguous estimates with the unambiguous estimate in order to verify the accuracy of the ambiguous estimates and correct them if necessary. This comparison makes it possible to "cancel" the ambiguity that occurred during the ambiguous estimation step.

[0019] The method according to the invention avoids the need to systematically perform an unambiguous estimation. An initial unambiguous estimation can be performed, and as long as it is not required, the method can consist solely of ambiguous estimations. The ambiguity due to the comb structure of the reference signals is eliminated through the comparison and selection steps. The solution proposed according to the invention is thus less complex than those proposed by the prior art while still allowing for accurate estimation.

[0020] According to one characteristic, the ambiguous estimation is performed using only one type of reference signal, the reference signals having a comb structure.

[0021] By using only one type of reference signal, fewer resources are used. Design complexity is thus significantly reduced.

[0022] According to a characteristic, the selection includes a choice of a value of the representative parameter obtained during said estimation with ambiguity corresponding to the closest value of at least one value of the representative parameter obtained at or selected after the previous estimation.

[0023] According to one characteristic, the process includes a step of memorizing said at least one value of said representative parameter.

[0024] According to a characteristic, said at least one value selected at the selection step is stored in place of said at least one value obtained at or selected after the previous estimation.

[0025] This avoids storing all values ​​in memory. Only the last values ​​obtained or selected, corresponding to the values ​​whose estimate is the most accurate (i.e., closest to the actual position or movement of the target), are kept in memory.

[0026] According to one characteristic, as long as the number of objects remains unchanged, the process includes at least one iteration of the steps of estimation with ambiguity, comparison and selection.

[0027] This allows the unambiguous estimation to be performed only once. As long as the number of objects remains constant, any subsequent estimation is carried out while allowing the ambiguity to occur. This ambiguity is "cancelled" (by performing the comparison and selection steps) without adding complexity to the solution design.

[0028] According to a characteristic, if the number of objects changes, the unambiguous estimation step is repeated and then the other steps are executed.

[0029] When the number of objects changes, repeating the unambiguous estimation yields another accurate estimate. Continuing the process by executing the remaining steps, particularly the ambiguous estimation step, introduces ambiguity. This ambiguity is resolved through the comparison and selection steps.

[0030] According to one feature, the process includes a step of identifying a change in the number of objects for which an unambiguous estimation of said representative parameter has been carried out, said identification step being carried out after an ambiguous estimation step.

[0031] Depending on one characteristic, the process includes:

[0032] identification of periods between each value obtained during each estimation with ambiguity;

[0033] after each ambiguous estimation, comparison of the periods identified during the current ambiguous estimation with the periods identified during the previous ambiguous estimation.

[0034] As long as the periods remain unchanged, the process includes at least one iteration of the comparison and selection steps.

[0035] If a change in the periods is identified, the unambiguous estimation step is repeated and then the other steps are executed.

[0036] In other words, if the signal structure changes (changes in the periods between the signal's amplitude peaks), then the unambiguous estimation step is repeated to obtain an accurate estimate. As long as the signal structure remains unchanged, only ambiguous estimations followed by comparison and selection steps are performed.

[0037] A change in the signal structure can be representative of a change in the number of objects. Since the occurrence of ambiguity due to the comb structure of the reference signals is periodic, a change in the structure (i.e., the periodicity) allows us to identify a change in the number of objects.

[0038] A change in the signal structure can also be representative of an object that is moving relatively quickly.

[0039] A change in the signal structure can also occur when all objects leave and then return to the environment of the communication device or the detection zone.

[0040] According to one characteristic, reference signals are positioning reference signals.

[0041] Positioning reference signals (PRS) have longer time intervals compared to other reference signals, thus providing more resources.

[0042] According to another aspect, the invention relates to a communication device configured to implement the detection method having at least one of the characteristics described above.

[0043] According to another aspect, the invention relates to a computer program comprising instructions for implementing the detection method having at least one of the characteristics described above, when said program is executed by a computer.

[0044] According to another aspect, the invention relates to a computer-readable recording medium on which the computer program according to the invention is recorded.

[0045] 4. Presentation of the figures and the appendix

[0046] Other advantages and features of the invention will become apparent in the following description with reference to the drawings and the appendix, given by way of non-limiting examples:

[0047] Figure 1 shows examples of reference signal patterns of the positioning reference signal type;

[0048] Figure 2 represents in a flowchart the steps of the detection process according to one embodiment of the invention;

[0049] Figure 3 represents an example of unambiguous estimation using PRS reference signals and DMRS reference signals; Figure 4 represents an example of ambiguous estimation using PRS reference signals;

[0050] Figure 5 represents a simplified architecture of a communication device according to one embodiment of the invention; and

[0051] The Annex presents an example of pseudo-code that can be used by the communication device.

[0052] 5. Detailed description of at least one embodiment of the invention

[0053] The invention relates to communication systems integrating a radio sensing function (Integrated Sensing and Communication, or ISAC). These are systems in which radio waves are not only configured to establish communication between a transmitting device and at least one other device, but also to obtain more general information about the environment (for example, the environment of this device). The present invention falls within the field of communications based on multi-carrier modulation and can be implemented, in particular, in the context of data transmissions based on orthogonal frequency-division multiple access (OFDM) techniques.

[0054] The transmitting equipment can be a base station, an access point, a relay node, or any other communication device. The transmitted waveform interacts with objects (also referred to here as targets) in the environment. The corresponding reflected and / or scattered waves are observed by a receiving device. The receiving device can be located, for example, at the transmitting equipment (typically at the transmitting base station), at another device (typically at another base station), at another network element, or at a terminal (such as a mobile terminal). The receiving device estimates a parameter representing the position or movement of the object. This parameter can be, for example, the object's distance from the transmitting or receiving equipment, or the object's velocity.

[0055] The invention uses reference signals for detection purposes. The reference signals used have a comb structure. The comb structure refers to a specific transmission scheme used to arrange the reference signals in time and frequency. The comb structure is characterized by a regular, comb-like arrangement of the signals, where the signals are uniformly spaced in the time or frequency domain.

[0056] The waveform is constructed on a time-frequency resource grid, with the smallest allocable time-frequency resource, called a physical resource block (PRB), in which a number of contiguous subcarriers are allocated for the transmission of a certain number of consecutive symbols. As illustrated in Figure 1 (from Khosroshahi 2024), each PRB can comprise twelve contiguous subcarriers, on each of which fourteen consecutive symbols can be transmitted.

[0057] The frequency density, hereafter denoted by K CO mb is determined by the size of the comb structure in the subcarrier allocation. As an example, the frequency density K COmb can take the values ​​{2, 4, 6, 12} for positioning reference signals (PRS). This means that samples of the reference signals are placed in each Kcomb-th subcarrier, thus building a non-contiguous but regular pattern in the frequency bands.

[0058] Figure 1 shows examples of reference signal patterns of the type known as positioning reference signals, called PRS reference signals, having a comb structure. The figure shows a first example a) in which the frequency density K CO mb is equal to 4, and a second example b) in which the frequency density K CO mb is equal to 12.

[0059] The reference signals used in the invention may be of the positioning reference signal type or any other type of reference signals allocated according to a comb structure, for example demodulation reference signals ("Demodulation Reference Signal or DMRS").

[0060] Figure 2 shows in a flowchart the steps of the detection process according to an example implementation. The detection process can be implemented by a communication device.

[0061] The detection method includes an unambiguous estimation step (EO) of a parameter representative of the position or movement of at least one object. This yields values ​​r h history of said representative parameter,

[0062]

[0063] i.e. {1, x being the number of objects or targets present in the environment of the communication device. One or more values ​​are obtained depending on the number of objects.

[0064] An unambiguous estimate is one in which the risks of confusion that can arise during radio detection are reduced. Specifically, an unambiguous estimate is one in which the ambiguity due to the reference structure containing empty subcarriers (i.e., the comb structure) is eliminated. In particular, the risks of falsely detecting multiple objects in the vicinity of a device when only one object is actually present, or of falsely detecting only one object in the vicinity of a device when multiple objects are actually present (for example, because stronger signals mask weaker signals), are eliminated.

[0065] The parameter representing the position or movement of the object can be, for example, a distance of the object from the transmitting equipment or the receiving equipment ("range estimation"), or a speed of movement of the object ("Doppler estimation").

[0066] Unambiguous estimation can be performed using several types of reference signals, or in other words, a combination of several types of reference signals. For example, unambiguous estimation can be performed using PRS reference signals and DMRS reference signals, as described in [Khosroshahi 2024]. According to another embodiment, unambiguous estimation can be performed using reference signals having an irregular structure, as described in [Golzadeh 2024]. Such estimation techniques allow for the suppression of false amplitude peaks in the signal corresponding to ghost targets.

[0067] The detection process can then include a step E1 of memorizing the r values h history of the representative parameter.

[0068] The detection process then includes an E2 estimation step with ambiguity, using a single type of reference signal. The reference signals here have a comb structure, for example, PRS reference signals. The E2 estimation step allows ambiguity to occur. Indeed, by using only one type of reference signal with a comb structure, false amplitude peaks corresponding to false targets are allowed to appear.

[0069] The detection method further includes a comparison step E3 between at least one value of the representative parameter obtained by the ambiguous estimation and at least one value of the representative parameter obtained by a previous estimation of the representative parameter.

[0070] The detection process then includes a selection step E4 of a value obtained in the estimation step with ambiguity from the comparison carried out in the comparison step E3.

[0071] The selection process involves choosing a value obtained from the ambiguous estimation. If the ambiguous estimate is the first ambiguous estimate, the chosen or selected value is the closest to at least one value obtained from the unambiguous estimation. If the ambiguous estimate is a subsequent ambiguous estimate, the chosen or selected value is the closest to at least one value selected following the previous selection step, or to at least one value obtained from the previous unambiguous estimation in cases where the unambiguous estimation is repeated as explained below. Values ​​obtained from the ambiguous estimation that are too far removed from the values ​​obtained from the unambiguous estimation or from the values ​​selected in the previous selection step are ignored.These values, which are too far apart and therefore ignored, correspond to the values ​​of false objects or phantom targets. After the selection step E4, the process can continue by storing the selected values. In other words, the process can continue by repeating the storage step E1. The value(s) obtained in each ambiguous estimation can be stored, possibly replacing the value(s) obtained in the previous estimation.

[0072] The process may include a step to identify a change in the number NbO of objects or targets. This identification step may be performed after the ambiguous estimation step E2. Preferably, the identification step is performed after each ambiguous estimation step E2.

[0073] After each ambiguous estimation (carried out in step E2), as long as the number NbO of objects or targets remains unchanged, the process continues with the comparison steps E3, selection E4 and possibly memorization E1.

[0074] The detection process allows for several successive estimations with ambiguity. This makes it possible to gather information on the actual targets and to match the estimated values ​​with the actual values.

[0075] When the number NbO of objects changes, the unambiguous estimation step EO is repeated. The process then continues with the ambiguous estimation step E2, the comparison step E3, and the selection step E4. Optionally, the storage step E1 is executed. For example, the storage step E1 is executed after the unambiguous estimation step EO, and then again after the selection step E4.

[0076] Thus, in the first ambiguous estimation, the previous estimation in the comparison step E3 corresponds to the first unambiguous estimate. In subsequent ambiguous estimations, as long as the number NbO of objects remains unchanged, the previous estimate corresponds to another ambiguous estimate, and when the number NbO changes, the previous estimate corresponds to an unambiguous estimate.

[0077] A change in the number of objects (NbO) is identified through the periodicity of occurrence of the ambiguity due to the comb structure of the reference signals. False amplitude peaks of the signal appear periodically. When the period of occurrence of the amplitude peaks changes, it means that the number of objects (NbO) has changed.

[0078] The distances at which false amplitude peaks or ghost targets appear for reference signals are defined by the following expression [Golzadeh 2024]:

[0079] c

[0080] dghost =

[0081]

[0082] ^•^comb^J

[0083] where c is the speed of light, K comb frequency density, and Af is the subcarrier spacing.

[0084] This expression applies to any reference signal having a comb structure, and in particular to PRS reference signals.

[0085] The following table presents the values ​​(in meters) of such periodic distances d ghost , between the false amplitude peaks for PRS reference signals:

[0086] | 2 s - A,-„ 2 4 (» 12

[0087] nvn: I i 4Wn 2-W Jfi6' H î2

[0088] 2-;us 1 M2 -Ifi

[0089] 125ii 1.24 41 h 2(IS

[0090] fC J 312 2(b> 1(14

[0091] 312 l.-'h 1G I?2

[0092] :?(i.'s?2 2d

[0093] 7N. V; 26 13

[0094]

[0095] When a peak in amplitude appears at a distance that is not equal to or close to the distance d ghost This means that the number NbO of targets or objects has changed. A change in the number NbO of objects is identified, in particular, when an amplitude peak appears at a distance greater than d ghost + s, where s is a predefined distance.

[0096] Thus, the process may include a step of identifying periods between each value obtained of the representative parameter during each estimation with ambiguity.

[0097] The process can then include, after each ambiguous estimation, a comparison of the periods identified during the current ambiguous estimation with the periods identified during the previous ambiguous estimation. If a change in the periods is detected, the process can include identifying a change in the number of objects (NbO). If a change in the number of objects (NbO) is identified, the comparison step (E3) and the selection step (E4) are not performed. The unambiguous estimation step (EO) is repeated to obtain the values ​​of the parameter representing the position or motion of the true objects or targets. Once the unambiguous estimation is completed, the remaining steps of the process can be executed.

[0098] More generally, the process may include a step of identifying periods between each value obtained for the representative parameter during each ambiguous estimation. The process may then include, after each ambiguous estimation, a comparison of the periods identified during the current ambiguous estimation with the periods identified during the previous ambiguous estimation. As long as the identified periods remain unchanged, the process includes at least one iteration of the comparison step E3 and the selection step E4. If a change in the periods is detected, the comparison step E3 and the selection step E4 are not executed. The unambiguous estimation step EO is repeated to obtain the values ​​of the parameter representing the position or motion of the actual objects or targets. Once the unambiguous estimation is completed, the remaining steps of the process can be executed.Figure 2 represents the determination of whether or not to modify the periods by the decision "CHG?".

[0099] Changes in signal periods (i.e., changes in signal structure) can be caused by, or representative of, several events. A change in signal structure can represent a change in the number of objects (as explained above).

[0100] A change in the signal structure can also be indicative of an object or target moving relatively quickly. This is particularly true when the distance traveled by the target between two successive estimates is on the order of d ghost / 2. In this case, when a change in the periods is detected, the unambiguous estimation step EO is executed. The rest of the process continues as for the other cases described above, for example by executing steps E1, E2, E3, E4.

[0101] A change in the signal structure can also occur when all objects leave and then return to the communication device's environment or the detection zone. In this case, even if the number of objects is the same before and after leaving the environment, the unambiguous estimation step EO is executed. The rest of the process continues as for the other cases described above, for example, by executing steps E1, E2, E3, and E4.

[0102] The equation above defining d ghost shows that the distance at which false amplitude peaks or false targets appear depends on the frequency density K comb and the spacing of the Af subcarriers. Preferably, the choice of these parameters K comb and Af takes into account the mobility of targets. If a target is not moving or is moving slowly, large values ​​for the frequency density K combare preferably chosen. Fewer reference signals are thus used. Conversely, if a target is moving quickly, smaller values ​​for the frequency density K are used. comb are preferably chosen. This allows for a more accurate estimation of the positions or speeds of movement of the targets.

[0103] In a specific example, the position of two targets (target 1 and target 2) is estimated. Target 1 is located at a distance of 10 m and target 2 at a distance of 20 m. The method according to the invention is carried out to obtain an estimate of the position of the two targets. The unambiguous estimation step EO provides, for example, two rhi values. S tor yi=9m and rhistory2=21m. The ambiguous estimation step E2, for example, provides four values: ri=9m, r2=21m, r3=113m, r4=134m. Since there are only two targets, it is deduced that two of these four values, n, r2, r3, r4, correspond to false targets. The values ​​n, r2, r3, r4 are compared to the values ​​rhistory2 and r. h istory2 (comparison step E3). The values ​​n, r2 are selected (selection step E4) as the true target values ​​because they are the closest to the rhistoryi and rhistor values. y 2- The rhistoryi and rhistor values y 2 are updated by fixing r h history yi = n and r h history y 2 = r2(memorization step E1).

[0104] This example can be generalized by considering x real targets. The process is executed to estimate the position of these x targets. The unambiguous estimation step EO provides position values ​​r history., i.e. {1,

[0105]

[0106] The ambiguous estimation step E2 provides positional values ​​r y ,je {1,...,y} where y corresponds to the total number of targets (sum of true and false targets). The values ​​r history . and r7 are compared (comparison step E3). The r values rea are selected from the y values ​​estimated in the ambiguous estimation step E2 as follows:

[0107] C

[0108]

[0109] real = r } where j = min jE{1 y} \r history . - r y | Life {1,...,%} According to this equation, for each real target ie {1,

[0110]

[0111] The position value is associated with the closest value with the index j corresponding to the closest value among the values ​​(i.e., the value providing the smallest difference with the values ​​^history j) •

[0112] The r values history are updated by fixing r history . = r it Life {1,

[0113]

[0114] Figure 3 (from [Khosroshahi 2024]) shows an example of unambiguous estimation using PRS and DMRS reference signals. The frequency density K CO mb is equal to 4. An estimation of the position of two targets is performed (target 1, also designated C1, and target 2, also designated C2). It should be noted that the curve in Figure 3 is normalized.

[0115] Combining two types of reference signals eliminates false amplitude peaks. The estimation curve only shows two signal amplitude peaks at distances corresponding to the two actual targets (target 1 and target 2). Therefore, the unambiguous estimation step E0 can be performed using this combination of two types of reference signals.

[0116] Figure 4 (from [Khosroshahi 2024]) shows an example of an ambiguous estimation using PRS reference signals. An estimation of the position of the same two targets (target 1 and target 2) as in Figure 3 is performed. Note that the curve in Figure 4 is normalized. As shown in Figure 4, the estimation results in false amplitude peaks, in this case six false peaks.

[0117] The method according to the invention allows the values ​​of the true targets to be selected and the values ​​of the false targets to be ignored. In the example of Figures 3 and 4, the method allows the values ​​of the two true targets, target 1 and target 2, to be selected and the six false peaks to be ignored.

[0118] The Annex presents an example of pseudocode that can be used by the communication device to implement the detection method. The preceding description was made for position estimates ("range estimations"). The method according to the invention also applies to motion estimations ("Doppler estimations").

[0119] Figure 5 schematically and in a simplified manner represents the structure of the communication device in a particular embodiment. The communication device can typically be a transmitting device. The communication device comprises, for example, a memory 1, a processing unit 2, equipped, for example, with at least one processor pP, and controlled by a computer program Pg 3, implementing steps of the detection process, according to at least one embodiment of the invention.

[0120] The invention simplifies radio detection in communication systems while maintaining the accuracy of the estimates. Furthermore, by using a series of ambiguous estimates, the simultaneous use of multiple reference signals is reduced.

[0121] 5. Bibliography

[0122] [Khosroshahi 2024]: Khosroshahi, Keivan, Philippe Sehier, and Sami Mekki. " Leveraging PRS and PDSCH for integrated sensing and communication systems." arXiv preprint arXiv:2408.00667 (2024).

[0123] [Golzadeh 2024]: Golzadeh, M., Tiirola, E., Talvitie, J., Anttila, L., Hooli, K., Tervo, O., & Valkama, M. (2024, May). Joint Sensing and UE Positioning in 5G-6G: PRS Range Estimation with Suppressed Ambiguity. In 2024- IEEE Radar Conference (RadarConf24) (pp. 1-6). IEEE.Annexe

[0124] While (True) Exécuter indéfiniment

[0125] Modified = o inMafealiiion Perform range estimation via ambiguity-free methods Estimation sans Compote range r t for each detected target i e (1,..., x). ambiguïté (étape EO) Set thistory, = ri for all ie (1,..., x}, ie. update the history of ranges. Storage or update (step E1) While (Modified = 0) As long as the signal structure remains unchanged Re-estimate H using a reference signal: with a comb structure (e.g., PRS only) If (modification_in_patternff) If the signal structure changes | Modified = 1 Set "Modified" to 1 Else Match the x real ranges using the history values ​​via closest Estimation with ambiguity (step E2), matching method for all r t , i E (1,..., x}. comparison (step E3), selection (step E4) Set f history, = Tj for all i E {1,...,x}, ie. update the ranges' Storage or update (step E1)

[0126]

[0127] history.

[0128] Note: The 'modification n_pattern()' function is a boolean function that returns 1 (true) when the false peaks are not within a certain distance d ghost real targets.

Claims

DEMANDS 1. A method for detection by a communication device comprising an unambiguous estimation (EO) of a parameter representative of a position or movement of at least one object, the method being characterized in that it further comprises the following steps: - ambiguous estimation (E2) of said representative parameter using at least one reference signal having a comb structure; - comparison (E3) between at least one value of the representative parameter obtained by said ambiguous estimation and at least one value of the representative parameter obtained by a previous estimation of said representative parameter; selection (E4) of at least one value obtained in said estimation with ambiguity from said comparison.

2. A detection method according to claim 1, wherein the ambiguous estimation (E2) is carried out using only one type of reference signal, the reference signals having a comb structure.

3. A detection method according to claim 1 or claim 2, wherein the selection (E4) comprises a choice of a value of the representative parameter obtained during said estimation with ambiguity (E2) corresponding to the value closest to at least one value of the representative parameter obtained at or selected after the previous estimation.

4. A detection method according to any one of claims 1 to 3, comprising a storage step (E1) of said at least one value of said representative parameter.

5. Detection method according to claim 4, wherein at least one value selected at the selection step (E4) is stored in place of said at least one value obtained at or selected after the previous estimation.

6. A detection method according to any one of claims 1 to 5, wherein as long as the number of objects remains unchanged, the method comprises at least one iteration of the steps of estimation with ambiguity (E2), comparison (E3) and selection (E4).

7. A detection method according to any one of claims 1 to 6, wherein if the number of objects changes, the unambiguous estimation (EO) step is repeated and then the other steps are executed.

8. A detection method according to any one of claims 1 to 7, comprising a step of identifying a change in the number of objects for which an unambiguous estimation (EO) of said representative parameter has been carried out, said identification step being carried out after an ambiguous estimation step (E2).

9. A detection method according to any one of claims 1 to 8, comprising: - identification of periods between each value obtained during each estimation with ambiguity; - after each ambiguous estimation, a comparison is made between the periods identified during the current ambiguous estimation and the periods identified during the previous ambiguous estimation. a process in which, as long as the periods remain unchanged, the process includes at least one iteration of the comparison (E3) and selection (E4) steps, and If a change in the periods is identified, the unambiguous estimation (EO) step is repeated and then the other steps are executed.

10. A detection method according to any one of claims 1 to 9, wherein the reference signals are positioning reference signals.

11. Communication device configured to implement the detection method according to any one of claims 1 to 10.

12. Computer program comprising instructions for carrying out the method according to any one of claims 1 to 10, when said program is executed by a computer.

13. Computer-readable recording medium on which the computer program according to claim 12 is recorded.

Images