Method of processing communication signals and vehicular radar signals for use in radar sensing, and apparatus configured to excute the method

Abstract

A method of processing communication and vehicular radar signals for radar sensing comprises receiving a reflected communication signal, reflected off of one or more physical objects, and an undisturbed copy of the corresponding previously transmitted communication signal, and determining one or more first values therefrom, each of which being respectively associated with an assumed velocity and / or distance, relative to the transmitter / receiver, of corresponding first respective candidate objects. The method further comprises receiving a reflected vehicular radar signal, reflected off of the one or more physical objects, and a corresponding undisturbed reference vehicular radar signal, and determining at least one second value therefrom, which is associated with an assumed velocity and / or distance, relative to the radar transmitter / receiver, of corresponding second respective candidate objects. A target detection is performed on the second and at least one first values, for identifying and outputting probable physical objects, their velocity and / or distance.

Description

[0001] METHOD OF PROCESSING COMMUNICATION SIGNALS AND VEHICULAR

[0002] RADAR SIGNALS FOR USE IN RADAR SENSING, AND APPARATUS CONFIGURED TO EXCUTE THE METHOD

[0003] FIELD OF THE INVENTION

[0004] The present invention relates to methods of processing communication and vehicular signals for use in radar sensing, in particular to signal processing targeted to suppress peaks in the radar sensing output, due to interference. The invention further relates to a computer program product implementing the method, to a computer- readable storage medium storing the computer program product, to an apparatus configured to execute the method, and to a vehicle comprising such apparatus.

[0005] BACKGROUND

[0006] Recent years have witnessed a proliferation of vehicular radars due to their crucial and weather-independent role in safety and functional applications for intelligent transport systems (ITS) and advanced driver assisted systems (ADAS). Besides the vehicular radar, vehicle-to-vehicle (V2V) communications have also become very important for ADAS, especially for automated or autonomous driving and teleoperations. However, with this proliferation of vehicular radars, interference among the vehicular radars has become a major impediment for proper operation of radio systems, such as automotive radar and Vehicle-to-everything (V2X) communications.

[0007] Modern vehicular radars use frequency modulated continuous wave (FMCW) chirps for sensing the environment. Furthermore, all the vehicles transmit more or less the same FMCW chirp waveform, which may cause interference to automotive radar in the ego vehicle, as shown in Fig. 1 . Here, a first vehicle E, also referred to herein as ego-vehicle, transmits a first FMCW vehicular radar signal, represented by the solid saw-tooth line, for sensing its environment. The FMCW vehicular radar signal travels towards a second vehicle T, also referred to herein as target vehicle. The target vehicle T emits a second FMCW vehicular radar signal for sensing its own environment, represented by the dash-dotted saw-tooth line, which travels towards the ego vehicle E. In addition, the first FMCW radar signal is reflected off of the target vehicle T and travels back towards the ego-vehicle E as a third FMCW radar signal, represented by the dotted saw-tooth line. While the third radar signal is the signal of interest for the ego-vehicle E, the superposition of the second and third radar signals may cause interference at the radar receiver of the ego-vehicle E.

[0008] Interference causes appearance of spurious ghost peaks in the radar sensing output of the ego vehicle as shown in Fig. 2. The presence of these ghost peaks causes an increase in the detection of spurious targets and thus degrades the target detection performance, which may ultimately lead to undesired consequences. The degradation of the target performance has been shown, e.g., by G. M. Brooker in "Mutual interference of millimeter-wave radar systems," IEEE Transactions on Electromagnetic Compatibility, vol. 49, no. 1 , pp. 170-181 , 2007, which discusses the nature and magnitude of the interference for millimeter-wave radar systems and other sensor systems by considering spatial, temporal, and operational frequency- related overlaps under different operating conditions. The suppression of these ghost peaks is, therefore, one of the most important challenges for vehicular radars.

[0009] A significant part of the prior art solutions is directed to discovering interference between millimeter-wave radars and then cancelling the interference passively at the receiver, e.g., as discussed by Martin Kunert in "More Safety for All by Radar Interference Mitigation," Dec 2012, PROJECT FINAL REPORT: MOSARIM project. https: / / cordis.europa.eu / docs / projects / cnect / 1 / 248231 / 080 / deliverables / 001 - D611 finalreportfinal, pdf.

[0010] The general approach for interference mitigation relies on the randomization of radar parameters in time, frequency, space, code and polarization domains followed by detection and avoidance. However, existing passive interference mitigation techniques have limited effects and high complexity for densely deployed vehicular radars, e.g., as discussed by F. Liu, A. Garcia-Rodriguez, C. Masouros, and G. Geraci in "Interfering channel estimation in radar-cellular coexistence: How much information do we need?" IEEE Transactions on Wireless Communications, vol. 18, no. 9, pp. 4238-4253, 2019.

[0011] Recent advances in mmWave communications, which operates in wavelengths similar or close to vehicular radar, and the wireless standardization efforts in the current fifth generation (5G) standard is opening up the possibility of integrating the radar and communication functions in the deployment of future V2X communications, smart cities and smart roads. The broad paradigm of integrating the radar and communication functions is referred to as joint radar and communications (JRC), and is widely recognized as a novel paradigm for 6G.

[0012] JRC based solutions are effective, simple and least demanding in terms of hardware. Exemplary approaches to interference mitigation based on JRC are presented by B. Li and A. P. Petropulu in "Joint transmit designs for coexistence of MIMO wireless communications and sparse sensing radars in clutter,” IEEE Transactions on Aerospace and Electronic Systems, vol. 53, no. 6, pp. 2846-2864, 2017, by Y. Cui, V. Koivunen, and X. Jing in "Interference alignment based spectrum sharing for mimo radar and communication systems," in 2018 IEEE 19th International Workshop on Signal Processing Advances in Wireless Communications (SPAWC), 2018, pp. 1 -5, and by F. Liu, A. Garcia-Rodriguez, C. Masouros, and G. Geraci, "Interfering channel estimation in radar-cellular coexistence: How much information do we need?" IEEE

[0013] Transactions on Wireless Communications, vol. 18, no. 9, pp. 4238-4253, 2019.

[0014] These approaches try to achieve interference mitigation by using pre-coding, beamforming and resource allocation techniques, which may require modifications on the transmitter-side and may not be feasible for legacy equipment that is already deployed in large numbers and which, therefore, will not immediately have significant beneficial impact.

[0015] It is, therefore, an object of the present invention to provide an improved method of processing vehicular radar signals and communication signals for use in radar sensing. It is a further object of the invention to provide an apparatus that is configured to implement and execute the method.

[0016] SUMMARY OF THE INVENTION

[0017] These objects are achieved by the method of claim 1 and the apparatus of claim 12. Advantageous embodiments and developments of the method and the apparatus are provided in the respective dependent claims. The present invention will be described in the following using an exemplary scenario as shown in figure 3 a). The basic setting shown here is similar to the one discussed further above with reference to figure 1 . However, in addition to the radar signals, the V2V or V2X system in the ego-vehicle E also transmits a separate communication signal after its FMCW chirp transmission, represented by the idealized transmission frame comprising a random data part and a preamble part. The random data part is represented by the random pattern and the preamble part is represented by the regular checkerboard pattern. At around the same time, the vehicular radar in the target vehicle T sends out its own FMCW chirp signal for sensing its environment. It is also assumed that there is no communication transmission from the target vehicle T for the considered time period of interest. From the perspective of the ego-vehicle E, the two transmitted signals are reflected by the target vehicle T, producing an FMCW radar echo and a communication echo, respectively. The FMCW radar echo is represented in the figure like it was represented in figure 1 . The communication echo is represented by the horizontally mirrored idealized transmission frame. Like in the scenario discussed with reference to figure 1 , the second FMCW radar signal transmitted from the target vehicle T superimposes with the echo of the first FMCW radar signal, i.e. , the signal reflected by the target vehicle T, in the ego-vehicle’s radar receiver. The superimposed radar signals and the reflected communication signal are received by respective receivers at the ego-vehicle E for further processing. As initially mentioned, the communication signal and the radar signal are transmitted at similar frequency ranges, e.g., 77 GHz for the FMCW radar signal and 60 GHz for the communication signal. Figure 3 b) shows a representation of the signals over the time as transmitted and received by the ego vehicle. The left graph shows the transmission of the communication signal next to the x-axis, and the reflected echo arriving at a later time. Note that the echo is inverted due to the reflection. The right graph shows the transmission of the FCMW radar signal next to the x-axis, and the reflected echo arriving at a later time. Likewise, the FCMW radar signal transmitted by the target vehicle T arrives at some time after the initial transmission of the FMCW radar signal by the ego vehicle E. It is obvious that the reflected echo and the FCMW radar signal transmitted by the target vehicle T will superimpose at the radar receiver during the time period tswhile they are both nonzero. The transmitted communication signal usually complies with a communication standard such as IEEE 802.11 ad. Hence, a typical communication signal transmission comprises a preamble part and a data part. The preamble signal up (r) is assumed to be transmitted in single-carrier (SC) modulation format, up (f) is given by where s(k),k = 0,1, -,NP- 1 are the preamble symbols to be transmitted and Npis the total number of preamble symbols. The pulse shape is assumed to be a rectangular pulse of duration Tc, and fcis the carrier frequency of the system.

[0018] In contrast, the data signal ud(t) is assumed to be transmitted in OFDM format. ud(t) is given by where S(kL + Z) represents the data symbol to be transmitted by the fcthOFDM block (symbol) over the Zthsub-carrier. To= T + Tcpis the OFDM block duration, including the elementary symbol duration T and the cyclic prefix (CP) duration Tcp. It is assumed that there are K OFDM blocks to be transmitted in total, each block being transmitted over L sub-carriers.

[0019] The transmitted FMCW radar signal, also referred to herein as chirp signal or chirp transmission, from both the ego-vehicle E and the target vehicle T is an FMCW waveform with Lcchirps and is given by where Tpis the chirp duration, Br= B / Tpis the modulation rate, B is the sweep bandwidth. It is assumed that the sweep bandwidth of the chirp transmission is same as the operating bandwidth of the communication transmission. All the transmitted signals are assumed to be reflected by M targets to produce M corresponding echoes. The receivers at the ego-vehicle E receive a superposition of these M echo signals. At the communication receiver, after down-conversion, the final baseband received signal for the SC preamble part can be expressed as, where amis the attenuation experienced by target m,dmis the distance of target m from the ego-vehicle E, vmis the relative velocity of target m with regard to the egovehicle E and c is the speed of light in vacuum.

[0020] Similarly, the final baseband received signal for the block-OFDM data payload part can be expressed as,

[0021] In the FMCW chirp receiver, the final de-chirped received signal is given by,

[0022] At the communication receiver, the preamble and data payload parts of the received signal, i.e., yp(t),yd(t) are processed separately using respective matched filters to obtain the corresponding radar ambiguity functions (AF) for the preamble part and the data part. Similarly, at the FMCW chirp receiver, yc(t) is processed to obtain the radar ambiguity function corresponding to the chirp signal.

[0023] In the following section the evaluation of these radar ambiguity functions will be discussed. The ego-vehicle E receives the communication echo signal and the chirp echo signal which are processed in the communication receiver and the FMCW radar receiver, respectively. In the first step, both receivers sample the respective signals, after down-converting and dechirping, as respectively required.

[0024] The samples obtained by sampling the preamble part of the communication echo in accordance with equation (4) are denoted by yp(n),n = 1,2, ..., Np. Similarly, the samples obtained by sampling the data part of the communication echo in accordance with equation (5) are denoted by yd(n),n = 1,2, ..., Nd. The samples obtained by sampling the chirp signal are denoted by the matrix Yc(n,Z), with n = 0, 1, 2,..., 7Vc and Z = 1, 2, ..., LC.

[0025] Furthermore, the clean reference samples for the preamble part, which is a copy of the preamble part of the baseband transmit communication signal are given by xp(n),n = 1,2, ... , Np. The clean reference samples for the data part, which is a copy of the data part of the baseband transmit communication signal in which the cyclic prefix (CP) is replaced by a zero prefix (ZP) are denoted by xd(n),n = 1,2, ..., Nd.

[0026] The AFs of the preamble and the data parts of the communication signal, respectively, are evaluated by the following steps:

[0027] 1. Divide the preamble part of the echo signal yp(n),n = 1,2, ..., NPand the corresponding clean reference signal, xp(n),n = 1,2, ...,Npinto Ipsegments, each of length Nc. yp(iNc+ ;) denotes the preamble sample j from segment i. Similarly, xp(iNc+ 7) denotes the sample j from segment i of the clean reference preamble samples.

[0028] 2. Divide the data part of the echo signal yd(n),n = 1,2, ... , Ndand the data part of the clean reference signal, xd(n), n = 1,2, ... , Ndinto Idsegments, each of length Nc. y(iNc+ j) denotes the data sample 7 from segment t. Similarly, yc(iNc+ 7) denotes the sample j from segment i of the clean reference data samples.

[0029] 3. The AF for the preamble part is given by

[0030] 4. Similarly, the AF for the data part is given by

[0031] The chirp AF can be evaluated from the chirp echo samples, Yc(n, 0 as follows:

[0032] 1 . Evaluate the Ncpoint range FFT over the columns of Yc(n, I) as,

[0033] 2. Evaluate the Icpoint Doppler FFT over the rows q = 1,2, .... Ncof Yc(q, 1) as,

[0034] It is to be noted that the sampling rates at the communication receiver and the chirp receiver are usually very different. For example, the communication receiver typically samples the communication echo signal at Fs= 1.76GHz, whereas the chirp radar samples the chirp echo signal at Fs= 128MHz. Hence, in order to enable combining of the preamble AF, the data AF and the chirp AF obtained in equations (7), (8) and (11 ) respectively, the corresponding echo sequences need to be zero-padded so that

[0035] Ip ~ ~ ^c-

[0036] Next, a method of processing communication signals and vehicular radar signals for use in radar sensing, in particular for suppressing the ghost peaks in a radar sensing image of vehicular radars (chirp AF), in accordance with the invention is discussed. A basic flow scheme of the proposed method is illustrated in Fig. 4. Labels A and B point to signals output by the vehicular radar and the method in accordance with the present invention, respectively, and will be referred to in figures 5 to 8.

[0037] As per the previous section, Apre(<7, fc), Adata(q,k) and Ac(q,k) denote the respective

[0038] AFs of the SC preamble part, block-OFDM data part and the chirp AF. In accordance with the present invention, a combined AF is generated by performing point-wise minimum selection of the square of the absolute values of the preamble AF

[0039] Apre (q> k), the square of the absolute values of data AF Adata(q, k) and the square of the absolute values of the chirp A¥Ac(q,k) at each delay and Doppler bin, as per

[0040] As noted previously, if the sampling frequencies of the communication receiver and the vehicular radar receiver differ, appropriate zero-padding may be invoked prior to the point-wise minimum selection of equation (12) to generate the combined AF

[0041] Under the assumption that the target vehicle T does not transmit an interfering communication signal in the duration of interest, any ghost peak that may appear in the preamble AF and the data AF will not be caused by such signal. However, even if the target vehicle T is transmitting a communication signal in the duration of interest, it can be assumed that the data part of the communication transmission from the target vehicle T is uncorrelated with the data part of the communication transmission from the ego-vehicle E. Also, as the communication transmission is regulated, it is also possible to ensure that different vehicles transmit different, uncorrelated preambles. In this case, the ghost peaks in the preamble AF or the data AF are negligibly small.

[0042] In order to properly detect targets while suppressing ghost peaks, proper setting of a threshold for the target detection in the combined AF is required. In the following section, the theoretical basis for such threshold-setting will be presented.

[0043] As described in the previous section, the final AF Amp(q,k) is a two dimensional function of time delay and Doppler frequency. For practical reasons the two- dimensional representation can be divided into discrete bins, i.e. , velocity ranges for the Doppler axis and distance ranges for the delay axis. In this section, the decisionmaking on whether or not a bin A.mp(q, k) contains a target will be presented. Here, the decision is made based on measuring the respective bin’s energy Emp(q, k) = fc)|2. As every bin will have some energy floor due to noise and ghost signals and the like, determining a proper threshold q is essential, above which it can be assumed that the energy in a bin is due to a target, i.e., Amp(q, k) corresponds to a target if Emp(q, k) > q. Hence, for each delay-Doppler bin of Amp(q, k), the decision on the presence or the absence of a target reduces to a binary hypothesis testing problem, i.e., does not contain target,

[0044] Mi: ,mp(q, k) contains target.

[0045] To this end, the following performance metrics are defined:

[0046] 1 . Probability of false alarm, PFA,

[0047] 2. Probability of detection, PD,

[0048] 3. Probability of interference rejection, PT, Given the above performance metrics, the goal now mathematically reduces to:

[0049] 1 . Maximizing PDfor the bins corresponding to the true target, and

[0050] 2. Maximizing PTfor the bins corresponding to the ghost peaks (interferer), so that the interference suppression is high, while keeping PFAlow. PFAmay be set by the user and may typically lie in a range between IO-4, IO-6.

[0051] The detection threshold q that maximizes the probability of correct detection (PD) for a predetermined probability of false alarm (PFA) is given by and the corresponding PDis given by where the Marcum Q-function and a2is the noise variance. 2pre, ■^-dat and Xcare given by

[0052] Hence, to make the decision regarding a target, the energy in each bin in the final AF, A.mp(q,k), is compared with the detection threshold q and the presence of the target in the bin is declared, if the energy, Emp(q,k) > q.

[0053] The combined final AF Amp(q,k) facilitates suppressing the interference for different reasons. First, it is assumed that the target vehicle T does not transmit an interfering communication signal in the duration of interest. Hence, any ghost peak in the preamble AF and the data AF cannot be caused by an interference of a communication signal. Second, even if the target vehicle T would transmit a communication signal in the duration of interest, it can be assumed that the data part of the communication signal transmission from target vehicle T is uncorrelated with the data part of the communication signal transmission from the ego-vehicle E. Also, as the communication signal transmission is regulated, it is also possible to ensure that different vehicles transmit different preambles which are uncorrelated. Hence in this case, the any ghost peaks that may appear in the preamble AF or the data AF will be negligibly small. As each bin in the final AF is obtained by taking the pointwise minimum of the square of the absolute values of the corresponding bins in the preamble AF, data AF and the chirp AF, negligibly small interference values in the data AF and the preamble AF, which will populate the interference bins in the final AF, will result in negligibly small values in the bins corresponding to the interferer.

[0054] Thus, in accordance with the first aspect of the invention, a method of processing communication signals and vehicular radar signals for use in radar sensing comprises receiving, at an apparatus implementing the method, representations of a reflected communication signal received at a communication antenna, which communication signal was previously transmitted via said communication antenna and which was reflected off of one or more physical objects, and of an undisturbed copy of the corresponding previously transmitted communication signal. From the received representations of communication signals one or more first values are determined, each of which being respectively associated with an assumed velocity and / or an assumed distance, relative to said communication antenna, of one of one or more corresponding first respective candidate objects. Alternatively, the method may comprise receiving one or more said first values at said apparatus. The method further comprises receiving, at the apparatus implementing the method, representations of a reflected vehicular radar signal received at a radar antenna, which vehicular radar signal was previously transmitted via said radar antenna and which was reflected off of the one or more physical objects, and of a corresponding undisturbed reference vehicular radar signal. From the received representations of vehicular radar signals, at least one second value is determined that is associated with an assumed velocity and / or an assumed distance, relative to the radar antenna, of one of one or more corresponding second respective candidate objects. Alternatively, the method may comprise receiving at least one such second value. The alternatives may apply, e.g., in case a communication apparatus and a radar apparatus operate independently from said apparatus implementing the method, and the apparatus implementing the method acts as a supervisor for suppressing false positives, i.e., ghost radar objects.

[0055] At least the representations of a reflected vehicular radar signal may comprise superimposed signals or signal components transmitted from other radar antennae, although the representations of the communication signals may likewise comprise superimposed signals or signal components transmitted from other communication antennae.

[0056] In a further method step a target detection is performed on the determined or received second and at least one first values, for identifying one or more probable physical objects in an environment of the communication antenna and / or the radar antenna. Finally, the result of the target detection is output, including information relating to objects in an environment of the communication and / or radar antenna, e.g., a velocity and / or a distance of at least one probable physical object is output, if any is identified. The output result may be used as is, i.e. , as a true representation of the environment, or may be used for complementing or correcting of data provided by a radar apparatus.

[0057] In one or more embodiments of the method determining one or more first values comprises determining one or more first ambiguity functions from said communication signals. Alternatively, one or more such first ambiguity functions, determined based on said communication signals may be received at the apparatus implementing the method.

[0058] In one or more embodiments of the method determining one or more first ambiguity functions Apre, Adatacomprises determining respective ambiguity functions based on a preamble of the communication signal and based on the data part thereof. Alternatively, one or more first ambiguity functions Apre, Adatadetermined based on a preamble of the communication signal and based on the data part thereof may be received. In one or more embodiments of the method determining at least one second value comprises determining a second ambiguity function Acfrom said vehicular radar signals. Alternatively, at least one second ambiguity function Acdetermined based on said vehicular radar signals may be received at the apparatus implementing the method.

[0059] In one or more embodiments of the method performing a target detection comprises pairing at least the one or more first and second candidate objects based on their respective assumed velocity and assumed distance, and eliminating improbable candidate objects based on the respective associated second and at least one first values. This may include dividing the feasible velocity and distance ranges covered by processing the communication signals and the radar signals, respectively, into corresponding discrete bins, and pairing a velocity-distance bin associated with a first candidate object with a corresponding velocity-distance bin available from processing the radar signals, irrespective of the second value assigned to the respective bin. Likewise, a velocity-distance bin associated with a second candidate object is paired with a corresponding velocity-distance bin available from processing the communication signals, irrespective of the first value assigned to the respective bin. In embodiments of the method, the pairing may be performed only for candidate objects whose first and second value, respectively, exceed respective thresholds. This may, inter alia, reduce the computational load.

[0060] In one or more embodiments of the method eliminating improbable candidate objects comprises determining a threshold value for comparing the second and at least one first values, or a third value derived therefrom, against. The comparison may be effected bin-wise for each velocity-distance bin.

[0061] When the second and the at least one first values are comprised in corresponding ambiguity functions, eliminating improbable candidate objects may further comprise combining the received or previously determined one or more first ambiguity function and the second ambiguity function into a combined ambiguity function Amp. This may further comprise, prior to combining, dividing the received or previously determined one or more first ambiguity function and the second ambiguity function Acinto respective discrete bins for combinations of distance and velocity ranges, and performing the combination bin-wise, each bin of the combined ambiguity function Ampbeing associated with a third value derived from the second and at least one first values.

[0062] The bin-wise combination may comprise performing a bin-wise m in-point selection on each corresponding distance and velocity bin of the at least one first ambiguity function (Apre, Adata), and of the second ambiguity function (Ac).

[0063] In one or more embodiments of the method eliminating improbable candidate objects comprises determining a threshold value for comparing the third values against.

[0064] In one or more embodiments of the method first ambiguity functions based on a preamble of the communication signal and based on the data part thereof are received or determined.

[0065] In one or more embodiments the representations of the communication signals, and of the undisturbed copy of the corresponding previously transmitted communication signal and / or of the representations of the reflected radar signal, and of the corresponding undisturbed reference radar signal, are zero-padded prior to determining the respective ambiguity functions, if the sampling frequencies with which the respective signals are sampled differently from each other.

[0066] In one or more embodiments the target detection comprises comparing a value representing a signal energy of each delay and Doppler bin of the combined ambiguity function with a predetermined detection threshold value, wherein a target is considered found in a delay-Doppler bin when the signal energy of that delay- Doppler bin exceeds the detection threshold value.

[0067] In the following section simulation results are discussed to assess the performance of the proposed algorithm. The simulation setup considers the scenario shown in Figure 3 and the corresponding system model described above. The bandwidth B is assumed to be 5 = 1.76GHz. Furthermore, it is assumed that the target vehicle T is at a distance of r = 150dr m and moving with a velocity of v = 8dv m / s. Here, dr = c / 2B = 0.0852 m is the range resolution, i.e., the size of the range bin, and dv the velocity resolution, i.e., the size of the velocity bin, with c being the speed of light in vacuum, X = 5 mm corresponding to the wavelength and Tf= 0.16 ms being the transmission duration. The communication system of the ego-vehicle E transmits Nf= 8 consecutive communication frames. The echo of these 8 frames reflected by the target vehicle T are received at the ego-vehicle E. The preamble AF, A.pre(q,ky and the data AF, Adata(q,k), is then obtained by processing the communication echoes. Similarly, the vehicular radars from the egovehicle E and the target vehicle T both transmit Lc= 32 consecutive chirps. The duration of each chirp is Tp= 5 / zs. The chirp echo reflected by the target vehicle T is received at ego-vehicle E's receiver. Furthermore, the target vehicle T's chirp transmission is also received by the ego-vehicle E's receiver. The superimposed chirp signal received at the ego-vehicle E's receiver is processed to evaluate the chirp AF, Ac(qt,fc). The interference peak in the radar sensing image of ego-vehicle E's vehicular radar apparatus is assumed to appear at bin (r = 30dr, v = 8dv). Next, the final AF, Amp(q,k) is evaluated by combining the preamble AF, the data AF and the chirp AF, Apre (q,k) and Ac(q,k) as described further above. The threshold for target detection is set by fixing the probability of false alarm, PFA, to pFA= io-4. For a signal consisting of samples, %(n),n = 1,2, ENR and SNR are defined as, where a2is the noise variance.

[0068] Figures 5 and 6 permit a comparison of the radar sensing image output, labelled A, of the conventional vehicular radar and the proposed final AF, Amp, labelled B. The following observations can be made in the figure: • the target is detected at the correct location by both the conventional vehicular radar image as well as the proposed method, and

[0069] • the interference peak, i.e. , the ghost peak appears in the conventional vehicular radar image, and is suppressed by the proposed method.

[0070] Figures 7 and 8 show a comparison of the target detection performance and the interference rejection performance, respectively, of the proposed final AF, Amp, labelled B in the figure, with the conventional radar sensing image output, labelled A in the figure. The following observations can be made from the figures:

[0071] • the proposed method has a better target detection performance, with possible gain of up to 2 dB ENR over a wide range of PDcompared to the conventional vehicular radar image, and

[0072] • the proposed method has a better interference suppression performance, with a possible gain of 100 times compared to the conventional vehicular radar image. Figures 4 and 5 show a clear improvement in the target detection performance and the interference rejection performance. Hence, it can be concluded that the proposed method effectively suppresses the interference as well as improves in target detection compared to the conventional vehicular radar sensing imaging.

[0073] In the previous results, it was assumed that the chirp transmission and the communication transmission have the same duration. As mentioned before, the proposed method also works when the duration of the chirp transmission is different from the communication transmission duration. However, a loss of detection performance may be noticed when the duration of the data transmission is made smaller. The final simulation result in figure 9 shows that in order to obtain the same PDthe required signal to noise ratio (SNR) is 5 dB higher, when the transmission duration of the chirp signal is eight (8) times the transmission duration of the communication signal, i.e., Tj = 8Tfomm. This is typically the case in practical systems, wherein the communication transmission happens in bursts of very short durations. Hence, it can be concluded that the proposed algorithm is applicable in a practical setting. In accordance with a second aspect of the present invention an apparatus configured for processing communication signals and vehicular radar signals for use in radar sensing is presented. The apparatus comprises one or more input interfaces configured for receiving representations of a reflected communication signal received at a communication antenna, which was reflected off of one or more physical objects, of an undisturbed copy of the corresponding previously transmitted communication signal, of a reflected radar signal received at a radar antenna, which was reflected off of the one or more physical objects, and of a corresponding undisturbed reference radar signal. The undisturbed copy of the corresponding previously transmitted communication signal may be received from a memory, where it was stored prior to transmitting. The reference radar signal may be obtained from a memory and may be complemented by information indicating a relative or absolute time instant of the transmission of the radar signal whose reflection, or echo, is being processed. The representations of the received signals may be provided as sampled versions of the signals or in pre-processed form, e.g., in the form of signals representing one or more ambiguity functions determined based on the corresponding signals. The one or more input interfaces may be implemented by one or more physical interfaces that receive the signals, sampled versions of the signals, or corresponding data representing the signals, and may comprise one or more logical interfaces that store and forward the data to corresponding processing modules. Thus, the various elements of the apparatus may be provided in hardware, software or a combination thereof. The apparatus may further comprise one or more microprocessors, and associated volatile and non-volatile memory. The non-volatile memory stores computer program instructions which, when executed by the one or more microprocessors, configure the apparatus to execute one or more embodiments of the method in accordance with the first aspect of the invention as presented further above.

[0074] The method described hereinbefore may be represented by computer program instructions. Accordingly, a computer program product comprises computer program instructions which, when executed by a microprocessor of a computer or an apparatus in accordance with the second aspect of the invention, cause the computer to execute or carry out one or more embodiments of a method in accordance with the first aspect of the present invention, and / or to accordingly control hardware and / or software blocks or modules of said apparatus in accordance with the second aspect of the invention as presented above.

[0075] The computer program instructions may be retrievably stored or transmitted on a computer-readable medium or data carrier. The medium or the data carrier may by physically embodied, e.g., in the form of a hard disk, solid state disk, flash memory device or the like. However, the medium or the data carrier may also comprise a modulated electro-magnetic, electrical, or optical signal that is received by the computer by means of a corresponding receiver, and that is transferred to and stored in a memory of the computer.

[0076] The present invention proposes a novel method which post-processes the independent radar sensing outputs obtained by processing the FMCW echo from the vehicular radar and a communication signal echo of the same vehicle. In the proposed method, a radar-like sensing output obtained by appropriate processing of the echoes of communication signals reflected off of a target is used to suppress ghost interference peaks in the radar sensing image that may be present due to radar signals independently transmitted by the target, thereby improving the target detection performance.

[0077] The performance of interference suppression can be measured by evaluating the probability of rejection (Pr) for ghost targets. Theoretical analysis and simulations show that the proposed method, along with the derived optimized threshold, has a better target detection performance of up to 2 dB ENR gain in terms of PD, as well as an up to 100 times better interference suppression performance in terms of PTat high ENR. These improvements result in more reliable automotive radar / sensing systems as the probability of false positives is reduced.

[0078] The gain in interference suppression can be achieved at the cost of only very little additional processing complexity at the receiver, and no additional cost or complexity at the transmission side.

[0079] BRIEF DESCRIPTION OF THE DRAWING

[0080] The invention will now be described with reference to the drawing, in which Fig. 1 schematically illustrates a the cause of interference between two vehicles having radar equipment,

[0081] Fig. 2 exemplarily shows the result of interfering radar signals,

[0082] Fig. 3 schematically shows a scenario in accordance with the present invention permitting improved radar detection performance, and signal relations occurring therein,

[0083] Fig. 4 illustrates a basic flow scheme of the method in accordance with the invention for improving radar detection performance,

[0084] Fig. 5 shows an exemplary vehicular radar sensing output for a prior art radar system,

[0085] Fig. 6 shows an exemplary sensing output obtained by applying the method in accordance with the present invention,

[0086] Fig. 7 shows a comparison between the target detection performances of a known radar system and a system implementing the method in accordance with the present invention,

[0087] Fig. 8 shows a comparison between the interference suppression performances of a known radar system and a system implementing the method in accordance with the present invention,

[0088] Fig. 9 shows a comparison between the target detection performances of a system implementing the method in accordance with the present invention for same and different transmission durations of the FMCW radar signal and the communication signal,

[0089] Fig. 10 shows an exemplary flow diagram of the method in accordance with the present invention, and

[0090] Fig. 11 shows a basic block diagram of an apparatus configured for executing the method in accordance with the present invention.

[0091] In the figures identical or similar elements may be referenced using the same reference designators.

[0092] DETAILED DESCRIPTION OF EXEMPLARY EMBODIMENTS

[0093] Figures 1 to 9 have already been discussed further above and will not be addressed again. Figure 10 shows a flow diagram of an exemplary embodiment of the method 100 in accordance with the present invention. In step 110a representations of a reflected communication signal received at a communication antenna, which was previously transmitted via said communication antenna and which was reflected off of one or more physical objects, are received at an apparatus 400 implementing the method. Further, an undisturbed copy of the corresponding previously transmitted communication signal is received at the apparatus 400. In step 120 one or more first values, each of which being respectively associated with an assumed velocity and / or an assumed distance, relative to said communication antenna, of one of one or more corresponding first respective candidate objects are determined. Alternatively, one or more such first values are received at the apparatus 400 in step 110b, the alternativity being indicated by the dashed outline of the box. In embodiments of the method step 120 may comprise a step 122a of determining one or more first ambiguity functions (AF) or a step 122b of receiving one or more such first AFs. Again, the alternativity is indicated by the dashed outline of the box, as it is in the entire figure. The method further comprises, in step 130a, receiving, at the apparatus 400 implementing the method, representations of a reflected vehicular radar signal received at a radar antenna, which was previously transmitted via said radar antenna and which was reflected off of the one or more physical objects, and of a corresponding undisturbed reference vehicular radar signal. In step 140 at least one second value that is associated with an assumed velocity and / or an assumed distance, relative to the radar antenna, of one of one or more corresponding second respective candidate objects is determined from said received representations of vehicular radar signals. Alternatively, at least one such second value is received in step 130b. In embodiments of the method step 140 may comprise a step 142a of determining a second AF from said vehicular radar signals, or a step 142b of receiving a second ambiguity function determined based on said vehicular radar signals. In step 150 a target detection is performed on the determined or received second and at least one first values, for identifying one or more probable physical objects. If at least one probable physical object is determined, “yes”-branch of step 158, a respective associated velocity and / or a distance is output in step 160.

[0094] Otherwise, “no”-branch of step 158, execution of the method is repeated beginning with the receiving steps 110a and 130a or 110b and 130b, respectively. In embodiments of the method step 150 may comprise a step 152 of pairing at least the one or more first and second candidate objects based on their respective assumed velocity and assumed distance, and a step 156 of eliminating improbable candidate objects based on the respective associated second and at least one first values. The eliminating step 156 may be preceded by a step 154 of determining a threshold value for comparing the second and the at least one first values, or a third value derived therefrom, against. In embodiments of the method step 156 may comprise a step 156-1 of dividing the received or previously determined one or more first ambiguity functions and the second ambiguity function into respective discrete bins for combinations of distance and velocity ranges, and a step 156-2 of bin-wise combining the received or previously determined one or more first ambiguity function and the second ambiguity function into a combined ambiguity function, each respective discrete bin for combinations of distance and velocity ranges being associated with a third value derived from the second and at least one third values.

[0095] Figure 11 shows a schematic block diagram of an exemplary apparatus 400 configured for executing the method in accordance with the first aspect of the invention. An input interface 402, a microprocessor 404, volatile memory 406 and non-volatile memory 408 are connected by one or more signal and / or data lines or buses 410. The non-volatile memory 408 stores computer program instructions which, when executed by the microprocessor 404 configure the apparatus 400 for performing the method in accordance with the first aspect of the invention.

[0096] LIST OF REFERENCE NUMERALS (PART OF THE DESCRIPTION) 100 method 156-1 divide AFs into bins

[0097] 110a receive representation of 156-2 bin-wise combining communication signals 156-3 comparing

[0098] 110b receive first values 158 probable physical object

[0099] 120 determine first values detected?

[0100] 122a determine first AF(s) 160 output

[0101] 122b receive first AF(s) 400 apparatus

[0102] 130a receive representation of radar 402 input interface signals 404 microprocessor

[0103] 130b receive second values 406 volatile memory

[0104] 140 determine second values 408 non-volatile memory

[0105] 150 perform target detection 410 data / signal line / bus

[0106] 152 pair candidate objects E ego-vehicle

[0107] 154 determine threshold T target vehicle

[0108] 156 eliminate improbable candidate objects

Claims

CLAIMS1. A method (100) of processing communication signals and vehicular radar signals for use in radar sensing, comprising:- receiving (110a), at an apparatus (400) implementing the method, representations of a reflected communication signal received at a communication antenna, which was previously transmitted via said communication antenna and which was reflected off of one or more physical objects, and of an undisturbed copy of the corresponding previously transmitted communication signal, and determining (120), from said received representations of communication signals, one or more first values, each of which being respectively associated with an assumed velocity and / or an assumed distance, relative to said communication antenna, of one of one or more corresponding first respective candidate objects, or receiving (110b) one or more such first values at said apparatus (400),- receiving (130a), at the apparatus (400) implementing the method, representations of a reflected vehicular radar signal received at a radar antenna, which was previously transmitted via said radar antenna and which was reflected off of the one or more physical objects, and of a corresponding undisturbed reference vehicular radar signal, and determining (140), from said received representations of vehicular radar signals, at least one second value that is associated with an assumed velocity and / or an assumed distance, relative to the radar antenna, of one of one or more corresponding second respective candidate objects, or receiving (130b) at least one such second value,- performing (150) a target detection on the determined or received second and at least one first values, for identifying one or more probable physical objects, and- outputting (160) a velocity and / or a distance of at least one probable physical object, if any is identified.

2. The method (100) of claim 1 , wherein determining (120) one or more first values comprises:- determining (122a) one or more first ambiguity functions from saidcommunication signals, or receiving (122b) one or more first ambiguity functions determined based on said communication signals.

3. The method (100) of claim 1 or 2, wherein determining (122a) one or more first ambiguity functions (Apre, Adata) comprises determining respective ambiguity functions based on a preamble of the communication signal and based on the data part thereof, or where receiving (122b) one or more first ambiguity functions (Apre, Adata) comprises receiving respective first ambiguity functions determined based on a preamble of the communication signal and based on the data part thereof.

4. The method (100) of one or more of claims 1 to 3, wherein determining (140) at least one second value comprises:- determining (142a) a second ambiguity function (Ac) from said vehicular radar signals, or receiving (142b) a second ambiguity function (Ac) determined based on said vehicular radar signals.

5. The method (100) of one or more of claims 1 to 4, wherein performing (150) a target detection comprises:- pairing (152) at least the one or more first and second candidate objects based on their respective assumed velocity and assumed distance, and- eliminating (156) improbable candidate objects based on the respective associated second and at least one first values.

6. The method (100) of claim 5, wherein eliminating (154) improbable candidate objects comprises determining (154) a threshold value for comparing the second and at least one first values, or a third value derived therefrom, against.

7. The method (100) of claim 5, when depending on claims 3 and 4, wherein eliminating (156) improbable candidate objects further comprises:- dividing (156-1 ) the received or previously determined one or more first ambiguity function (Apre, Adata) and the second ambiguity function (Ac) into respective discrete bins for combinations of distance and velocity ranges,- bin-wise combining (156-2) the received or previously determined one or morefirst ambiguity function (Apre, Adata) and the second (Ac) ambiguity function into a combined ambiguity function (Amp), each respective discrete bin for combinations of distance and velocity ranges being associated with a third value derived from the second and at least one third values.

8. The method (100) of claim 7, wherein bin-wise combining (156-2) comprises performing a bin-wise m in-point selection on each corresponding distance and velocity bin of the at least one first ambiguity function (Apre, Adata), and of the second ambiguity function (Ac).

9. The method (100) of claim 7 or 8, wherein eliminating (156) improbable candidate objects comprises determining (156-3) a threshold value for comparing the third values against.

10. The method (100) of any one or more of claims 3 to 9, further comprising zeropadding the representations of the communication signals and / or of the radar signals prior to determining the respective ambiguity functions, if the sampling frequencies with which the respective signals are sampled differ from each other.

11. The method (100) of any one or more of claims 3 to 10, wherein the target detection comprises comparing a value (Emp) representing a signal energy of each delay and Doppler bin of the combined ambiguity function (Amp) with a predetermined detection threshold value (77), wherein a target is considered found in a delay-Doppler bin when the signal energyof that delay-Doppler bin exceeds the detection threshold value (77).

12. Apparatus (400) configured for processing communication signals and vehicular radar signals for radar sensing, comprising one or more input interfaces (402) configured for receiving representations of a reflected communication signal received at a communication antenna, which was reflected off of one or more physical objects, of an undisturbed copy of the corresponding transmitted communication signal, of a reflected radar signal received at a radar antenna, which was reflected off of the one or more physical objects, and of acorresponding reference radar signal, and further comprising, one or more microprocessors (404) and associated volatile (406) and non-volatile (408) memory, the elements of the apparatus being communicatively connected via one or more signal and / or data lines and / or buses (410), wherein the nonvolatile memory (408) stores computer program instructions which, when executed by the one or more microprocessors (404) configure the apparatus (400) for performing the method (100) of any one of claims 1 to 11 .

13. A computer program product comprising instructions, which, when the program is executed by a microprocessor of a computer or an apparatus (400) in accordance with claim 12, cause the computer and / or control hardware blocks, modules or components of the apparatus (400) in accordance with claim 12 to execute or carry out a method (100) of any one of claims 1 to 11 .

14. Computer readable medium or data carrier retrievably transmitting or storing the computer program product of claim 13.

15. A vehicle, a road-side unit or an edge-computing unit comprising an apparatus (400) according to claim 12.

Images