Radar signal processing method, circuit, virtual antenna array, antenna array and device

Abstract

The invention discloses a radar signal processing method, a circuit, a virtual antenna array, an antenna array and equipment. The radar signal processing method comprises the following steps: acquiring a distance Doppler spectrum of a radar signal; for a range gate, acquiring K Doppler units with targets in the range Doppler spectrum corresponding to the range gate; for a Doppler unit with a target, determining corresponding frequency spectrum estimation by adopting an iterative adaptive IAA spectrum estimation method accelerated by fast Fourier transform (FFT); and selecting one frequency spectrum meeting a set condition from the frequency spectrum estimations corresponding to the K Doppler units with the targets as an angle spectrum of the range gate, wherein the fast Fourier transform FFT is carried out based on the K Doppler units with the target, and K is an integer greater than 0. According to the scheme, the radar signal angle spectrum is determined through FFT accelerated IAA spectrum estimation, the distance angle spectrum calculation speed is remarkably increased, and the real-time requirement of radar signal processing is met.

Description

Technical Field

[0001] This article relates to radar technology, particularly a radar signal processing method, integrated circuits, virtual antenna arrays, antenna arrays, and electronic devices. Background Technology

[0002] Radar is an electronic system that uses electromagnetic waves to detect targets and acquire information such as their distance, speed, and angle. Since its invention in the early 20th century, radar technology has been widely used in military, civilian, and scientific research fields, becoming one of the core tools of modern detection and sensing. As an important component of radar technology, radar target detection systems aim to extract target information from complex electromagnetic environments and achieve target detection, tracking, and identification.

[0003] The basic working principle of a radar target detection system is to emit electromagnetic waves and receive the echo signals reflected from the target, then use signal processing techniques to extract the target's characteristic information. Its core task is to separate the target signal from noise, clutter, and interference, and accurately estimate the target's position, velocity, and other characteristics. With the continuous development of radar technology, target detection systems have gradually evolved from early simple range measurements into multifunctional, high-precision integrated detection systems, capable of adapting to complex environments and diverse target requirements. With the rapid development of electronic technology, signal processing algorithms, and computing power, radar target detection systems are evolving towards higher resolution, multifunctionality, and intelligence. Summary of the Invention

[0004] This disclosure provides a radar signal processing method, an integrated circuit, a virtual antenna array, an antenna array configuration, and an electronic device. The radar signal processing method employs FFT-accelerated IAA spectrum estimation to determine the radar signal angle spectrum, significantly improving the calculation speed of the range angle spectrum and meeting the real-time requirements of radar signal processing.

[0005] This disclosure provides a radar signal processing method, including: Obtain the range Doppler spectrum of the radar signal; For a range gate, obtain the K Doppler cells with targets in the range Doppler spectrum corresponding to the range gate; For a Doppler cell with a target, an iterative adaptive IAA spectral estimation method accelerated by Fast Fourier Transform (FFT) is used to determine the corresponding spectral estimate; and Select one of the K Doppler units with targets that meets the set conditions as the angle spectrum of the range gate; The Fast Fourier Transform (FFT) is performed based on the K Doppler units with targets, where K is an integer greater than 0.

[0006] This disclosure also provides an integrated circuit including a digital signal processing module configured to perform digital signal processing according to a radar signal processing method as described in any embodiment of this disclosure.

[0007] This disclosure also provides a virtual antenna array, including: at least one main subarray, each main subarray including at least four array elements distributed sequentially along a first direction; at least a portion of the main subarrays include a uniform linear array ULA subarray.

[0008] This disclosure also provides an antenna array, including: Multiple receiving antennas and multiple transmitting antennas are provided. The multiple receiving antennas are arranged sequentially along a first direction to form a non-equally spaced receiving antenna array, wherein some receiving antennas in the receiving antenna array are distributed sequentially at equal intervals. The plurality of transmitting antennas are distributed in multiple rows, wherein some transmitting antennas are distributed at equal intervals along the first direction, and when projected as a row, the plurality of transmitting antennas are arranged at non-equal intervals in the first direction.

[0009] This disclosure also provides an electronic device, including: a carrier, an integrated circuit as described in any embodiment of this disclosure, and an antenna; wherein the integrated circuit is disposed on the carrier; the antenna is disposed on the carrier and is integrated with the integrated circuit as a single device or disposed separately; and is connected to the integrated circuit for transmitting radio frequency transmission signals and / or receiving radio frequency reception signals.

[0010] Optionally, the antenna may be an antenna array as described in any embodiment of this application, and / or may be configured to form a virtual antenna array as described in any embodiment of this application.

[0011] Other features and advantages of this application will be set forth in the following description, and will be apparent in part from the description, or may be learned by practicing the application. Other advantages of this application can be realized and obtained by means of the solutions described in the description and the accompanying drawings. Attached Figure Description

[0012] The accompanying drawings are used to provide an understanding of the technical solutions of this application and constitute a part of the specification. They are used together with the embodiments of this application to explain the technical solutions of this application and do not constitute a limitation on the technical solutions of this application.

[0013] Figure 1 This is a schematic diagram of a radar signal processing system structure provided in an embodiment of this application; Figure 2 A schematic diagram of an MBC waveform provided in an embodiment of this application; Figure 3 This is a schematic diagram illustrating the target distance resolution effect in the relevant technical solution; Figure 4 A schematic diagram of zero-padding for MBC wave sampling data provided in an embodiment of this application; Figure 5 A flowchart of a radar signal bandwidth synthesis method provided in this application embodiment; Figure 6 Another MBC waveform diagram provided in this application embodiment; Figure 7 This is a schematic diagram illustrating the parallel execution of batch distance gate frequency domain transformation and velocity dimension FFT operations provided in an embodiment of this application. Figure 8 This is a schematic diagram illustrating the target distance resolution effect provided in an embodiment of this application. Figure 9 A flowchart of another radar signal bandwidth synthesis method provided in this application embodiment; Figure 10 A flowchart of another radar signal bandwidth synthesis method provided in this application embodiment; Figure 11 A flowchart of an IAA spectrum estimation algorithm accelerated by FFT is provided for embodiments of this application; Figure 12 A radar signal processing flowchart provided in this application embodiment; Figure 13 Another radar signal processing flowchart provided in this application embodiment; Figure 14 A schematic diagram of an integrated circuit structure provided in an embodiment of this application; Figure 15a A schematic diagram of an antenna array provided in an embodiment of this application; Figure 15b A schematic diagram of a virtual array provided in an embodiment of this application; Figure 15c A schematic diagram of the DBF result of a ULA provided in an embodiment of this application; Figure 15d A schematic diagram of the orientation DBF result of a principal subarray provided in an embodiment of this application; Figure 15e A schematic diagram of pitch DBF results provided in an embodiment of this application; Figure 15f This is a schematic diagram illustrating the effect of antenna amplitude and phase error on sidelobes, provided in an embodiment of this application. Figure 15g This is a schematic diagram illustrating the effect of antenna position error on sidelobes, provided in an embodiment of this application. Figure 15h A schematic diagram of an alternative antenna array provided in an embodiment of this application; Figure 15iA schematic diagram of the MIMO antenna position provided in an embodiment of this application; Figure 15j A schematic diagram of another ULA DBF result provided for an embodiment of this application; Figure 15k A schematic diagram of the orientation DBF result of another principal subarray provided in an embodiment of this application; Figure 15 This application provides a schematic diagram of an antenna array as an embodiment. Figure 15m This is another schematic diagram of the MIMO antenna position provided in an embodiment of this application; Figure 15n A schematic diagram of another ULA DBF result provided for an embodiment of this application; Figure 15o A schematic diagram of the orientation DBF result of another principal subarray provided in an embodiment of this application; Figure 15p This is another schematic diagram of pitch DBF results provided in an embodiment of this application; Figure 15q This is a schematic diagram illustrating the effect of antenna amplitude and phase error on sidelobes, provided in an embodiment of this application. Figure 15r This is a schematic diagram illustrating the effect of another antenna position error on the sidelobe, provided in an embodiment of this application. Figure 15s1 A schematic diagram of a radar system chip provided in an embodiment of this application; Figure 15s2 A schematic diagram of a MIMO virtual array provided in an embodiment of this application; Figure 15s3 This is a schematic diagram of another principal subarray orientation DBF result provided in an embodiment of this application; Figure 15s4 This is another schematic diagram of pitch DBF results provided in an embodiment of this application; Figure 15s5 This is a schematic diagram of another antenna array provided in an embodiment of this application; Figure 15s6 This is a schematic diagram of another MIMO virtual array provided in an embodiment of this application; Figure 15s7 This is a schematic diagram of another principal subarray orientation DBF result provided in an embodiment of this application; Figure 15s8 This is another schematic diagram of pitch DBF results provided in an embodiment of this application; Figure 15s9 This is a schematic diagram of another antenna array provided in an embodiment of this application; Figure 15s10 This is a schematic diagram of another MIMO virtual array provided in an embodiment of this application; Figure 15s11 This is a schematic diagram of another principal subarray orientation DBF result provided in an embodiment of this application; Figure 15s12 This is another schematic diagram of pitch DBF results provided in an embodiment of this application; Figure 15s13 A schematic diagram of DBF results for the orientation of a coproton array provided in this application embodiment; Figure 15s14 This is a schematic diagram of another coproton array orientation DBF result provided in an embodiment of this application; Figure 15s15 This is a schematic diagram of another antenna array provided in an embodiment of this application; Figure 15s16 This is a schematic diagram of another MIMO virtual array provided in an embodiment of this application; Figure 15s17 A schematic diagram of a 28-element ULA orientation DBF result provided in an embodiment of this application; Figure 15s18 A schematic diagram of a 5-element ULA pitch DBF result provided in an embodiment of this application; Figure 15s19 This is a schematic diagram of another antenna array provided in an embodiment of this application; Figure 15s20 This is a schematic diagram of another MIMO virtual array provided in an embodiment of this application; Figure 15s21 A schematic diagram of a 24-element ULA orientation DBF result provided in an embodiment of this application; Figure 15s22 This is a schematic diagram of another principal subarray DBF result provided in an embodiment of this application; Figure 15s23 This is a schematic diagram of another antenna array provided in an embodiment of this application; Figure 15s24 This is a schematic diagram of another MIMO virtual array provided in an embodiment of this application; Figure 15s25 Another schematic diagram of 24-unit ULA orientation DBF results provided in this application embodiment; Figure 15s26 This is a schematic diagram of another principal subarray DBF result provided in an embodiment of this application; Figure 15s27 This is a schematic diagram of another antenna array provided in an embodiment of this application; Figure 15s28 This is a schematic diagram of another MIMO virtual array provided in an embodiment of this application; Figure 15s29 A schematic diagram of a 20-element ULA orientation DBF result provided in an embodiment of this application; Figure 15s30 This is a schematic diagram of another principal subarray DBF result provided in an embodiment of this application; Figure 15s31 A schematic diagram of a radar chip layout provided in an embodiment of this application; Figure 15s32 This application provides a schematic diagram of an antenna interface layout. Figure 15s33 This is a schematic diagram of a radar chip and antenna layout provided in an embodiment of this application; Figure 15s34 This is a schematic diagram of another antenna interface layout provided in an embodiment of this application; Figure 15s35 This is another schematic diagram of a virtual array provided in an embodiment of this application; Figure 15s36 This is a schematic diagram of another antenna interface layout provided in an embodiment of this application; Figure 15s37 This is another schematic diagram of chip layout provided in an embodiment of this application; Figure 16 This is a schematic diagram of an electronic device structure provided in an embodiment of this application. Detailed Implementation

[0014] This application describes several embodiments, but these descriptions are exemplary and not limiting, and it will be apparent to those skilled in the art that many more embodiments and implementations are possible within the scope of the embodiments described herein. Although many possible combinations of features are shown in the drawings and discussed in the detailed description, many other combinations of the disclosed features are also possible. Unless specifically limited, any feature or element of any embodiment may be used in combination with or in lieu of any other feature or element in any other embodiment.

[0015] This application includes and contemplates combinations of features and elements known to those skilled in the art. The embodiments, features, and elements disclosed in this application may also be combined with any conventional features or elements to form a unique inventive scheme as defined by the claims. Any feature or element of any embodiment may also be combined with features or elements from other inventive schemes to form another unique inventive scheme as defined by the claims. Therefore, it should be understood that any feature shown and / or discussed in this application may be implemented individually or in any suitable combination. Therefore, the embodiments are not limited except by the limitations imposed by the appended claims and their equivalents. Furthermore, various modifications and changes may be made within the scope of the appended claims.

[0016] Furthermore, in describing representative embodiments, the specification may have presented methods and / or processes as a specific sequence of steps. However, the method or process should not be limited to the specific order of steps described herein, to the extent that it does not depend on such a specific order. As will be understood by those skilled in the art, other sequences of steps are also possible. Therefore, the specific order of steps set forth in the specification should not be construed as a limitation of the claims. Moreover, the claims concerning the method and / or process should not be limited to the steps performed in the written order, and those skilled in the art will readily understand that these orders can be varied and still remain within the spirit and scope of the embodiments of this application.

[0017] The basic working principle of a radar target detection system is to emit electromagnetic waves and receive the echo signals reflected from the target, then use signal processing techniques to extract the target's characteristic information. Its core task is to separate the target signal from noise, clutter, and interference, and accurately estimate the target's position, velocity, and other characteristics. With the continuous development of radar technology, target detection systems have gradually evolved from early simple range measurements into multifunctional, high-precision integrated detection systems, capable of adapting to complex environments and diverse target requirements.

[0018] Range resolution, velocity resolution, and angular resolution are all extremely important indicators for radar, with high resolution being key to achieving automotive radar imaging. Improving range resolution typically requires increasing the radar's effective sampling bandwidth, but this also means more range sampling points and higher memory requirements. Limited by the number of radar signal receiving channels, radar imaging resolution is not high, and employing super-resolution imaging schemes often places higher demands on real-time processing.

[0019] This application provides a radar super-resolution imaging scheme, which transmits MBC (Multi-band Chirp) waveforms and performs radar signal processing based on the ADC (Analog-to-Digital Converter) data of the received echo signals, such as... Figure 1 As shown, the radar super-resolution imaging system includes: a bandwidth synthesis and range FFT module, which performs range-dimensional FFT (Fast Fourier Transform) processing based on the synthesized bandwidth; a velocity FFT module, which performs velocity-dimensional FFT processing; a target detection module, which performs target detection operations such as CFAR (Constant False Alarm Rate) detection based on the range-Doppler spectrum; an angle spectrum estimation module, which estimates the angle spectrum based on the target detection results; and a range-angle spectrum generation module, which generates range-angle images.

[0020] As can be seen, the radar transmits MBC waveforms, achieving a wider bandwidth through inter-pulse frequency hopping, thus obtaining a high-resolution range spectrum. Some exemplary MBC waveforms with frequency hopping steps are shown below. Figure 2 As shown, the radar transmits N chirps, also known as N pulses, with each chirp having a transmission bandwidth of B, and the frequency step between chirps being... Also known as frequency hopping step, it can be a positive or negative value.

[0021] Some feasible MBC bandwidth synthesis schemes use range-dimensional FFT and velocity-dimensional FFT for radar signal processing, and the representation of targets at the same speed but different distances in the range-Doppler spectrum (RD spectrum) is as follows: Figure 3 As shown, the frequency stepping between Chirps It is a positive value. The frequency hopping bandwidth obtained by bandwidth synthesis in this method is:

[0022] Its distance resolution is

[0023] Where c is the speed of light.

[0024] As can be seen, the synthesized bandwidth does not include the bandwidth B of a single chirp, resulting in limited improvement in range resolution. Based on this, targets at different distances with the same velocity in the resulting RD spectrum, such as... Figure 3 As shown, the target spectrum in this method is distinguished diagonally along the RD diagram. In other words, in this method, the range dimension spectrum of two targets, Target1 and Target2, with the same speed but different distances, has insufficient distinguishability.

[0025] This application provides a radar signal bandwidth synthesis method, wherein the extended frequency hopping bandwidth is: ; Its distance resolution is: ; The bandwidth of each chirp is sampled throughout the extended process, such as... Figure 4 As shown, normal sampling is performed within the effective bandwidth B of each chirp, as indicated by black dots. Outside the effective bandwidth, zero-padding is applied, as indicated by white dots; that is, the white dots are filled with zeros. This is equivalent to zero-padding the unsampled frequency points for each pulse. A distance-dimensional FFT is performed based on the zero-padding sampled data from all chirs, thereby achieving a larger bandwidth. The synthesis improved the distance resolution.

[0026] This application provides a radar signal bandwidth synthesis method through its embodiments, such as... Figure 5 As shown, it may include: Step 510: Obtain sampling point data for multiple chirps, and pad the sampling point data outside the effective bandwidth of each chirp with zeros to obtain the zero-padded sampling point data.

[0027] Step 520: Divide the data of at least two chirped and zero-padded sampling points into multiple batches according to the distance gate, and perform distance-dimensional FFT on each batch to obtain the distance gate spectrum corresponding to each batch.

[0028] Step 530: After performing phase compensation on the spectrum of each range gate, perform velocity-dimensional FFT to obtain the range Doppler spectrum.

[0029] Optionally, each of the plurality of chirps has a bandwidth of B, and the frequency steps between consecutive chirps can be set to be equal or unequal depending on the actual scenario requirements.

[0030] It should be noted that the multiple chirs constitute the MBC waveform. The bandwidth synthesis method provided in this application is applicable to radar signal processing of the MBC waveform. The MBC waveform can be a frequency-hopping waveform with synchronized progression between consecutive chirs, for example... Figure 2 As shown, it can also be a frequency-hopping waveform with different timings between consecutive chirs, such as... Figure 6 As shown.

[0031] In some exemplary embodiments, the starting signal frequency corresponding to the first chirped sampling point of the plurality of chirps is The starting signal frequencies corresponding to the other chirped sampling points are all greater than 100 Hz. The highest signal frequency is In some exemplary embodiments, in a uniformly stepped multi-chirp scheme, the starting signal frequency of the sampling point of the i-th chirp is increased compared to the starting signal frequency of the sampling point of the (i-1)-th chirp. , This is a preset positive value. In a non-uniform stepping multi-chirp scheme, the starting signal frequency of the sampling point of the i-th chirp increases or decreases compared to the starting signal frequency of the sampling point of the (i-1)-th chirp. The changed starting signal frequency of the sampling point... All are greater than the starting signal frequency of the sampling point of the first chirp among multiple chirps. The maximum signal frequency of multiple chirped sampling points is denoted as . .

[0032] In some exemplary embodiments, the starting signal frequency corresponding to the first chirped sampling point of the plurality of chirps is The highest signal frequency corresponding to the first chirped sampling point is The starting signal frequencies corresponding to the other chirped sampling points are all less than [a certain value]. The minimum starting signal frequency is In some exemplary embodiments, in a uniformly stepped multi-chirp scheme, the starting signal frequency of the sampling point of the i-th chirp is decreased compared to the starting signal frequency of the sampling point of the (i-1)-th chirp. This is a preset negative value. In a non-uniform stepping multi-chirp scheme, if the starting signal frequency of the sampling point of the i-th chirp increases or decreases compared to the starting signal frequency of the sampling point of the (i-1)-th chirp, the starting signal frequency of the changed sampling point... All are less than the starting signal frequency of the sampling point of the first chirp among multiple chirps. The minimum starting signal frequency of multiple chirped sampling points is denoted as... .

[0033] In some exemplary embodiments, for a uniformly stepped frequency hopping MBC waveform, after zero-padding the data at sampling points outside the bandwidth of each chirp, the signal bandwidth is... Wherein, the number of the plurality of chirps is N, and the frequency step between consecutive chirps is . , This is the default value.

[0034] In some exemplary embodiments, for a non-uniform step-hopping MBC waveform, the starting signal frequency at the first chirped sampling point... Given the lowest frequency among the starting signal frequencies of the sampling points in multiple chirs, after zero-padding the sampling points outside the bandwidth of each chirp, the bandwidth of each chirp after zero-padding is... ,in, The highest signal frequency corresponding to the sampling point among the plurality of chirps. The starting signal frequency is the sampling point corresponding to the first chirp among the plurality of chirps.

[0035] In some exemplary embodiments, for a non-uniform step-hopping MBC waveform, the starting signal frequency at the first chirped sampling point... In the case of the highest starting frequency among the starting signal frequencies of the sampling points in multiple chirps, The highest signal frequency corresponding to the sampling point of the first chirp among the plurality of chirs is used. For each chirp, zero-padding is applied to the sampling points outside the bandwidth of that chirp. The bandwidth of each chirp after zero-padding is... ,in, The lowest starting signal frequency corresponding to the sampling point among the plurality of chirps. The highest signal frequency corresponding to the first chirped sampling point among the plurality of chirps.

[0036] As can be seen, by padding each chirp with zeros at sampling points outside the effective frequency of that chirp, the range of the synthesized bandwidth includes not only the additional bandwidth added by the frequency hopping step. This also includes the bandwidth B of the chirp itself, thereby improving distance resolution. The bandwidth synthesis method provided in this application treats each pulse as a sparse sample of the entire distance bandwidth.

[0037] Assume the radar's ADC sampling frequency is The effective sampling time of a Chirp is The number of ADC sampling points of the original radar for: ; The number of sampling points padded with zeros after bandwidth aggregation for: ; It is evident that zero-padding increases the number of sampling points, which in radar signal processing means a greater memory requirement.

[0038] After padding the sampling points outside the effective bandwidth of each chirp with zeros, the number of sampling points is significantly increased. In order to reduce the memory usage of distance dimension FFT processing, some exemplary embodiments process the zero-padded sampling point data in batches, which can reduce memory usage.

[0039] In some exemplary implementations, dividing the zero-padded sampling point data into multiple batches according to distance gates may include: dividing the zero-padded sampling point data into M batches, with each batch containing K / M distance gates; where K is the total number of distance gates, K is greater than or equal to M, and both are integers greater than 0. In some exemplary implementations, K is an integer multiple of M.

[0040] As can be seen, the entire range gate can be divided uniformly or non-uniformly. In the case of non-uniform division, the amount of sampling data points processed in each batch is different.

[0041] In some exemplary implementations, the distance dimension FFT is performed for each batch in the following manner: that is, by performing linear frequency modulated Z-transform ChirpZ, refined fast Fourier transform ZoomFFT, or discrete Fourier transform DFT on the sampling point data corresponding to the K / M distance gates included in the current batch as a specific implementation of the distance dimension FFT.

[0042] As can be seen, spectral refinement algorithms such as ChirpZ transform, ZoomFFT, or DFT are introduced to support frequency domain transformation in batches along the distance dimension.

[0043] For example, with K=512 and M=8, the radar system divides all 512 range gates into 8 batches. Each batch performs frequency domain transformation on the sampling point data corresponding to 512 / 8=64 range gates. Compared with the processing method of performing frequency domain transformation on all sampling point data, executing sampling point data of 64 range gates each time can greatly reduce the memory space required for performing frequency domain transformation.

[0044] In some exemplary implementations, the step of performing a distance-dimensional FFT to obtain the distance-gate spectrum corresponding to each batch includes: sequentially performing a frequency domain transformation on the M batches to obtain the distance-gate spectrum corresponding to each batch. That is, performing a distance-dimensional FFT serially on the sampling point data corresponding to each distance-gate batch.

[0045] In some exemplary implementations, the step of performing a distance-dimensional FFT to obtain the distance gate spectrum corresponding to each batch includes: performing frequency domain transformation on M batches in parallel to obtain the distance gate spectrum corresponding to each batch. To improve the efficiency of frequency domain transformation processing, multiple batches can also be executed in parallel when memory and computing resources are sufficient.

[0046] It is understandable that performing a distance-dimensional FFT independently on each batch yields the distance-gate spectrum corresponding to that batch. For example, for the i-th batch, Perform Chirp Z-transform to obtain the distance gate. The range gate spectrum is obtained. Then, phase compensation is performed on the range gate spectrum of this batch, followed by velocity-dimensional FFT processing to complete bandwidth synthesis, yielding the range-Doppler spectrum corresponding to the range gates of this batch. After performing the above processing on all batches, the complete range-Doppler spectrum is obtained.

[0047] Research has found that the frequency hopping frequency of a radar may not fall on integer frequency sampling points, making zero-padded FFT calculations impossible. Therefore, compensation at the range frequency is needed to achieve bandwidth synthesis. Assume the effective sampling time of one radar chirp is... Using Chirp1 as the reference, the frequency hopping frequency of Chirp2 relative to Chirp1 is... This is equivalent to the sampling delay of Chirp2 relative to Chirp1. for:

[0048] Based on the properties of FFT processing, the time delay is equal to the frequency shift, therefore, compensation needs to be performed at the frequency.

[0049] In some exemplary embodiments, phase compensation is performed for each range gate spectrum according to the following formula:

[0050] Where G(f) is the phase compensation function, and f is the frequency of the signal to be compensated. For the effective sampling time of a chirp, The frequency difference between the chirp containing the signal to be compensated and the first chirp among the plurality of chirps is denoted as G(f). It can be seen that after phase compensation using the phase compensation function G(f), the phase-compensated range gate spectrum is obtained.

[0051] In some exemplary embodiments, The frequency difference between the starting signal frequency of the chirp containing the signal to be compensated and the starting signal frequency of the first chirp among the plurality of chirps; or equal to the frequency difference between the starting signal frequency corresponding to the sampling point of the chirp containing the signal to be compensated and the starting signal frequency corresponding to the sampling point of the first chirp among the plurality of chirps.

[0052] In some exemplary embodiments, with In the MBC frequency modulation scheme with uniform stepping, ; Where f is the frequency of the signal to be compensated. Let G(f) be the effective sampling time for one chirp, and G(f) be the compensated spectrum. For multiple chirps, there is a frequency hopping step, where i represents the i-th chirp corresponding to the frequency signal, i=1,…,N.

[0053] In some exemplary embodiments, in the non-uniform step MBC frequency modulation scheme, ; Where G(f) is the phase compensation function, and f is the frequency of the signal to be compensated. The effective sampling time for one chirp is given by , where i represents the i-th chirp corresponding to the frequency signal. Let be the starting frequency of the chirp containing the signal to be compensated. Let be the starting signal frequency of the first chirp among N chirps.

[0054] In some exemplary embodiments, M=5, and the radar signal bandwidth synthesis is as follows: Figure 7 As shown, after zero-padding the sampling data of multiple chirped ADCs outside the chirped bandwidth, the data is stored in SRAM (Static Random Access Memory). The zero-padding sampling data of all range gates is divided into M batches for ChirpZ spectrum refinement. Then, phase compensation is performed according to the phase compensation function G(f), followed by a velocity-dimensional FFT transform to obtain the range-Doppler spectrum of the multiple chirped data, also known as the range-velocity spectrum. This completes the two-dimensional FFT transform based on the synthetic bandwidth. In some exemplary embodiments, the obtained range-Doppler spectrum is as follows... Figure 8 As shown, compared to Figure 3 The range-Doppler spectrum shown, according to the two-dimensional FFT transformation based on the synthetic bandwidth proposed in the embodiment of this application, has better range dimension resolution, and two targets Target1 and Target2 with the same velocity but different distances have more accurate discrimination in the range-Doppler spectrum.

[0055] The bandwidth synthesis scheme provided in this disclosure not only utilizes the bandwidth of frequency hopping but also synthesizes the transmit bandwidth of the chirp itself, further expanding the bandwidth. Based on this, FFT processing can achieve a further improvement in range resolution. In some exemplary embodiments, velocity-dimensional FFT processing is performed in batches according to the range gate, which does not require excessive memory resources and meets the cost requirements of radar product implementation.

[0056] This application also provides a radar signal processing method, such as... Figure 9 As shown, it includes: Step 910: Obtain the range Doppler spectrum of the radar signal.

[0057] Step 920: For a range gate, obtain the K Doppler units with targets in the range Doppler spectrum corresponding to the range gate.

[0058] Step 930: For a Doppler cell with a target, the corresponding spectrum estimate is determined by using the FFT-accelerated IAA (Iterative Adaptive Approach) spectrum estimation method.

[0059] Step 940: Select one of the K Doppler units with targets that meets the set conditions from the spectral estimates as the angular spectrum of the range gate.

[0060] Optionally, the Fast Fourier Transform (FFT) is performed based on the K Doppler units with targets, where K is an integer greater than 0.

[0061] Among them, IAA spectral estimation is a method for estimating the spectrum of a signal. It optimizes the accuracy of the spectrum estimation by iteratively updating the weights. This method can significantly improve the spectral resolution and reduce sidelobes, and is also known as IAA super-resolution spectral estimation.

[0062] In some exemplary embodiments, the method further includes: step 950, setting the angular spectrum of a range gate to a noise value for a range gate in which no Doppler cell with a target exists in the range Doppler spectrum corresponding to the range gate.

[0063] In some exemplary embodiments, step 930 includes: performing an FFT operation on the signal data corresponding to the Doppler unit in the radar signal to obtain the energy spectrum. The initial value is determined; an IAA spectral estimation using FFT acceleration is performed based on the energy spectrum, and the energy spectrum is iteratively updated. The iteration continues until the exit condition is met, at which point the spectral estimate corresponding to the Doppler unit is determined.

[0064] In some exemplary embodiments, IAA spectrum estimation accelerated by FFT is performed for each Doppler cell with a target to obtain the corresponding spectrum estimate.

[0065] The iteration exit condition (also known as the iteration algorithm convergence condition) includes any one of the following: the number of iterations reaches the set maximum iteration threshold, or the change in the updated energy spectrum compared to the energy spectrum before the update is less than the threshold value.

[0066] In some exemplary embodiments, the step of performing an IAA spectral estimation iteration update based on the energy spectrum using FFT acceleration may include: updating the energy spectrum based on the energy spectrum. The result of the FFT operation yields the root r of the equation; the Levinson-Durbin (LD) recursive algorithm is then applied to the root r to optimize the covariance matrix R and obtain the optimized signal Y; based on the optimized signal Y, an FFT operation is performed to update the energy spectrum. .

[0067] In some exemplary embodiments, the method based on the energy spectrum The result of performing the FFT operation, yielding the root r of the equation, may include: the energy spectrum. Perform a P-point FFT operation to obtain the initial roots of the equation. Take the roots of the initial equation. The first Q data points are the roots r of the equation; where P is the number of angle units and Q is the number of radar signal receiving channels.

[0068] Wherein, the inverse matrix of the optimized covariance matrix R ; Optimized signal ; , It is a strictly lower triangular matrix of size P×P, where all elements on the first subdiagonal are 1 and all other elements are 0. ; ; ; ; ; Where X is the signal data corresponding to the Doppler unit in the radar signal, and x is the signal data of each radar signal receiving channel in X; The root of the equation The i-th element, H is the vector conjugate transpose operation, * is the vector conjugate operation, and LD() is the LD recursive operation.

[0069] In some exemplary embodiments, the step of performing an FFT operation based on the optimized signal Y to update the energy spectrum is... This may include: performing an FFT operation based on the optimized signal Y, and obtaining the angle spectrum estimate using the following method: ; Update the energy spectrum ; in, ; ; ; ; ; C = [ ]; ; ; ; in, Let i be the i-th element of vector c. Let be the conjugate of the i-th element of vector c. This is the reversal operation on vector c. F ( ) represents the FFT operation; is the inversion operation for vector c, * is the conjugate operation for vectors, and T is the transpose operation for vectors; Let be the i-th element of vector t. It is the conjugate of the i-th element of vector t; Let be the i-th element of vector s. Let be the conjugate of the i-th element of vector s. For the P angle units, the first p indivual.

[0070] In some exemplary embodiments, selecting one of the spectrum estimates that meets the set conditions from the spectrum estimates corresponding to the K Doppler cells with the target as the angle spectrum of the range gate includes: selecting the spectrum estimate corresponding to the maximum energy value of the spectrum estimates of the K Doppler cells with the target as the angle spectrum of the range gate.

[0071] In some exemplary embodiments, in step 920, target detection is performed according to the following method to obtain a Doppler cell with a target: CFAR detection is performed based on the range Doppler spectrum to obtain a target point whose distance and velocity exceed a set threshold.

[0072] In some exemplary embodiments, the number of channels in the radar array is Q, the number of angles for spectral estimation is P, and the radar antenna array is a ULA (Uniform Linear Array) array, such as... Figure 10 As shown, in step 930, for each Doppler cell with a target, IAA spectrum estimation accelerated by FFT is performed according to the following steps.

[0073] Step 931: Perform FFT operation on the signal data corresponding to the Doppler element in the radar signal to obtain the energy spectrum. Initial value: ; in, For FFT operation; the signal X is the signal corresponding to the current range gate and the current Doppler cell in the radar array received signal. That is, a P-point FFT operation is performed on the signal data X corresponding to the range gate and the Doppler cell in the radar signal to obtain the energy spectrum. The initial value.

[0074] Step 932, based on the energy spectrum The result of performing the FFT operation Thus, the root r of the equation is obtained; ; ; That is, perform a P-point FFT operation on the current energy spectrum to obtain ,Pick The first Q data points yield the equation root r. During the first iteration, a P-point FFT operation is performed on the initial values ​​of the energy spectrum to obtain... In non-first iterations, a P-point FFT operation is performed on the energy spectrum obtained from the previous calculation to obtain... .

[0075] Step 933: Execute the Levinson-Durbin recursive algorithm LD() to calculate the vector based on the root r of the equation. and generate vectors and ,in, ; ; ; ; ; Step 934, based on vector and The inverse matrix of the covariance matrix R is calculated. and received the signal ,right Perform FFT operation to obtain the molecule ,in, ; ; ; in, , It is a strictly lower triangular matrix of size P×P, where all elements on the first subdiagonal are 1 and all other elements are 0; that is, each iteration can continuously optimize the covariance matrix R and the corresponding inverse matrix. This leads to the optimized signal Y.

[0076] Step 935: Generate vector c, and perform FFT operation on vector c to obtain the denominator. ;in, ; ; ; C = [ ]; ; ; ; ; in, Let i be the i-th element of vector c. Let be the conjugate of the i-th element of vector c. This is the reversal operation on vector c. F ( ) represents the FFT operation; is the inversion operation for vector c, * is the conjugate operation for vectors, and T is the transpose operation for vectors; Let be the i-th element of vector t. It is the conjugate of the i-th element of vector t; Let be the i-th element of vector s. Let be the conjugate of the i-th element of vector s. For the P angle units, the first p indivual.

[0077] Step 936: Generate the IAA angle spectrum estimate for this iteration. And update the energy spectrum , ; Update the energy spectrum 。

[0078] Step 937: Determine whether the iteration exit condition is met. If it is met, exit the iteration; otherwise, return to step 932 to execute the next iteration.

[0079] The IAA spectrum estimation method accelerated by FFT provided in this application accelerates processing efficiency by first performing FFT on the energy spectrum to determine the root r of the equation during the iteration process. Then, FFT is performed on the optimized signal Y before inverse matrix multiplication, reducing computational complexity compared to direct matrix multiplication. It can be seen that the use of FFT for signal or data processing during the iteration process of the IAA spectrum estimation algorithm significantly accelerates the execution of the algorithm and meets the real-time requirements of radar signal processing.

[0080] Using P-point FFT operations, Q radar received signal channels, and V as the maximum number of iterations, the IAA spectrum estimation algorithm without FFT acceleration has a computational complexity of... The IAA spectrum estimation algorithm provided in this application, which uses FFT operations for acceleration, reduces the computational load to [missing information]. For a radar system with Q=32, P=512, and V=10, the accelerated IAA spectrum estimation method provided in this application reduces the computational load to 1.63% of the original value.

[0081] Therefore, the angular spectrum for each range is obtained based on the range-Doppler spectrum of the radar signal, thus forming the range-angle spectrum of the entire radar signal. The IAA spectrum estimation algorithm uses FFT operations to accelerate the calculation process, achieving high resolution while maintaining processing efficiency, meeting the application requirements of high real-time performance.

[0082] In some exemplary embodiments, such as Figure 11 As shown, the radar signal processing method includes: Step 1100: Obtain the range Doppler spectrum of the radar signal.

[0083] Step 1110: Target point detection is performed based on the distance Doppler spectrum.

[0084] Step 1120: For each distance gate, perform the following steps to determine the angle spectrum of that distance gate.

[0085] Step 1121: Obtain the Doppler unit in the range Doppler spectrum corresponding to the range gate.

[0086] Step 1122: Perform IAA spectrum estimation using FFT acceleration for each Doppler unit to determine the corresponding spectrum estimate.

[0087] Step 1123: Select one of the spectrum estimates corresponding to all the Doppler units that meets the set conditions as the angle spectrum of the range gate.

[0088] Step 1124: If no target point is detected in the distance gate, the set noise value is used as the angle spectrum of the distance gate.

[0089] In some exemplary embodiments, the radar signal includes: an MBC waveform radar signal, each frame of the radar signal including multiple chirps, such as... Figure 12 As shown, the radar signal processing method may further include: Step 900: Generate the range Doppler spectrum by performing bandwidth synthesis based on the radar signal; wherein, step 900 performs bandwidth synthesis according to the radar signal bandwidth synthesis method described in any embodiment of this application, and generates the range Doppler spectrum.

[0090] In some exemplary embodiments, the radar signal processing method is applied to a radar system comprising a ULA (Uniform Linear Array) array. The number of array channels in the radar system equals the number of transmitting antennas multiplied by the number of receiving antennas.

[0091] It is understood that the radar signal processing method provided in this application embodiment achieves high-resolution range profiles of the radar through bandwidth synthesis, and by using FFT to accelerate the IAA spectrum estimation algorithm, it can achieve high resolution while taking into account the real-time requirements of the radar.

[0092] In some exemplary embodiments, selecting one of the spectrum estimates corresponding to the K Doppler units that meets the set conditions as the angle spectrum of the range gate includes: selecting the spectrum estimate corresponding to the maximum energy value of the spectrum estimates of the K Doppler units as the angle spectrum of the range gate. That is, for each range gate, the maximum energy value of the spectrum estimate is taken along the Doppler dimension to obtain the angle spectrum of the range gate.

[0093] In some exemplary embodiments, the radar signal processing method is applied to an FMCW (Frequency-Modulated Continuous Wave) radar with a MIMO (Multiple-Input Multiple-Output) antenna array, wherein the MIMO antenna array includes at least one main subarray, and at least a portion of the main subarray includes a uniform linear array ULA subarray; and the angle spectrum of the range gate is obtained based on the radar signal corresponding to the ULA subarray.

[0094] In some exemplary embodiments, the radar signal corresponding to the main subarray to which the ULA subarray belongs is configured to perform a deambiguation operation.

[0095] In some exemplary embodiments, the radar signal is obtained based on a virtual antenna array; the virtual antenna array includes at least one main subarray, at least a portion of the main subarray comprising a uniform linear array ULA subarray; the radar signal corresponding to the ULA subarray is used to determine the angular spectrum of each range gate.

[0096] In some exemplary embodiments, each of the main subarrays may include at least four array elements distributed sequentially along a first direction; at least a portion of the main subarrays include ULA subarrays, that is, for any main subarray that includes ULA subarrays, the number of array elements contained in the main subarray is greater than the number of array elements contained in the ULA subarrays it contains.

[0097] Specifically, for any main subarray containing a ULA subarray, the ULA subarray can be configured to perform angle determination operation in the first direction during radar signal processing, and the corresponding main subarray can be used to perform de-ambiguity operation on the aforementioned angle determination results, thereby effectively improving the resolution of angle measurement while retaining the advantage of low ULA sidelobes.

[0098] In some exemplary embodiments, such as Figure 13 As shown, the method further includes: Step 960: Perform angle resolution operation based on the angle spectrum of each distance gate to obtain the angle resolution result, and perform defuzzification operation on the angle resolution result to determine the direction of arrival (DoA).

[0099] Step 960 can be executed after step 940 is completed for each distance gate, or after step 940 is completed for all distance gates, and is not limited to any specific aspect.

[0100] The radar signal is obtained based on a virtual antenna array; the virtual antenna array includes at least one main subarray, and at least a portion of the main subarray contains a uniform linear array ULA subarray; the radar signal corresponding to the ULA subarray is used to determine the angular spectrum of each range gate; the radar signal corresponding to the main subarray to which the ULA subarray belongs is used for deambiguation operation.

[0101] It can be understood that the radar signal corresponding to the ULA subarray is used to determine the angle spectrum of each range gate, which means that the radar signal used to determine the angle spectrum in step 940 is the radar signal corresponding to the ULA subarray; the radar signal corresponding to the main subarray to which the ULA subarray belongs is used for defuzzification operation, which means that the radar signal used for defuzzification operation in step 960 is the radar signal corresponding to the main subarray.

[0102] For example, a virtual antenna array can be constructed using a MIMO antenna array. This virtual antenna array may include at least one sparse master subarray, and at least a portion of the sparse master subarray contains a ULA subarray. For a sparse master subarray containing a ULA subarray, when the ULA subarray is configured for angle estimation, the corresponding sparse master subarray can be used to perform angle de-ambiguation on the angle estimation operation of the ULA subarray, thereby improving the accuracy of the ULA subarray angle estimation or achieving operations such as super-resolution angle estimation. That is, when configured for angle estimation, the accuracy of angle estimation using the sparse master subarray is higher than that using the ULA subarray.

[0103] As can be seen, since the number of array elements contained in the sparse master subarray is greater than the number of array elements contained in the ULA subarray it contains, when configured for target detection (such as ranging, velocity and / or angle measurement), the ULA subarray (such as an equally spaced ULA subarray) requires less processing resources and is more efficient than the sparse master subarray.

[0104] It should be noted that the angles involved in angle estimation and angle ambiguity resolution in this embodiment may include the horizontal angle (or azimuth angle) and / or the elevation angle. If it is necessary to improve the angle measurement accuracy of the horizontal and elevation angles, the virtual antenna array may include at least two sparse master subarrays, that is, based on at least two sparse master subarrays arranged in a cross-extension pattern (both containing ULA subarrays), to realize the resolution of azimuth and elevation angle ambiguity.

[0105] For example, when using FMCW millimeter-wave radar for angle estimation, a virtual array can be constructed using a MIMO antenna array. This virtual array includes a sparse master subarray consisting of 32 array elements, and the sparse master subarray contains equally spaced ULA subarrays with a spacing of one wavelength, such as 24 of the 32 array elements mentioned above. The spacing between adjacent array elements is one wavelength, thus forming a ULA subarray. When the ULA subarray is configured for angle estimation and angle ambiguity exists, during the DoA operation in radar signal processing, the corresponding sparse master subarray can be configured for solving the ambiguity of the ULA subarray. That is, in the DoA estimation process, the possible list of ambiguous angles can be obtained from the angle resolution results of the ULA subarray (i.e., the ULA part of the master subarray). Then, the ambiguity list is orthogonally projected (e.g., maximum likelihood estimation (DML, deterministic maximum likelihood)) in the sparse master subarray (i.e., the full master subarray) to obtain the combination corresponding to the maximum likelihood probability, which is the final azimuth angle estimation list. This makes it feasible to improve resolution and dynamic range while retaining the low sidelobe level of ULA and the ability to add traditional signal processing windows. Furthermore, the FIAA (Fast Iterative Adaptive Algorithm) algorithm can be used to accelerate the super-resolution algorithm to ensure the frame rate of the entire radar signal processing.

[0106] According to the radar signal processing scheme provided in the embodiments of this application, FFT processing is used to accelerate the IAA algorithm to achieve super-resolution spectral estimation of angles, which can fully meet the real-time requirements of radar signal processing. In some exemplary embodiments, the range-angle spectrum of radar signals synthesized based on the bandwidth synthesis scheme provided in the embodiments of this disclosure has a higher bandwidth, which can further improve the range resolution. The radar signal processing performance is improved in terms of both resolution enhancement and real-time processing, so as to fully meet the application requirements of radar systems such as intelligent driving and real-time monitoring.

[0107] This application embodiment also provides an integrated circuit 140, such as Figure 14 As shown, it includes a digital signal processing module 1410, which is configured to perform digital signal processing according to the radar signal processing method described in any embodiment of this application.

[0108] In some exemplary embodiments, the integrated circuit further includes: a radio frequency (RF) module 1420 and an analog signal processing module 1430. The RF module 1420 is used to generate an RF transmission signal and process an RF reception signal; the analog signal processing module is used to obtain an intermediate frequency (IF) signal based on the RF reception signal; and the digital signal processing module 1410 is further configured to perform analog-to-digital conversion on the IF signal to obtain the digital signal. The RF module 1420, the analog signal processing module 1430, and the digital signal processing module 1410 are connected sequentially.

[0109] In some exemplary embodiments, the integrated circuit may include a millimeter-wave radar chip, etc.

[0110] In some exemplary embodiments, the integrated circuit further includes: an antenna for radiating high-frequency electromagnetic waves generated by the radio frequency module into space, and for receiving electromagnetic waves reflected from a target and transmitting them to the radio frequency module. The antenna carrier can be independently mounted or integrated with other modules on the same carrier. In some exemplary embodiments, the antenna includes a ULA array antenna.

[0111] In some exemplary embodiments, the antenna includes: a multiple-input multiple-output (MIMO) antenna array; the MIMO antenna array constitutes a virtual antenna array, including: at least one main subarray; at least a portion of the main subarray comprises a uniform linear array (ULA) subarray.

[0112] In some exemplary embodiments, the radar signal corresponding to the ULA subarray is used to determine the angular spectrum of each range gate; the radar signal corresponding to the master subarray to which the ULA subarray belongs is used for deambiguation operation.

[0113] In some exemplary embodiments, the MIMO antenna array includes multiple receiving antennas and multiple transmitting antennas. The multiple receiving antennas are arranged sequentially along a first direction to form a row of non-equally spaced receiving antenna arrays, and some of the receiving antennas in the receiving antenna array are arranged at equal intervals. The multiple transmitting antennas are distributed in multiple rows, wherein some of the transmitting antennas are distributed at equal intervals along the first direction, and all the transmitting antennas as a whole are arranged at non-equal intervals in the first direction.

[0114] This disclosure also provides a virtual antenna array, including: at least one main subarray, each main subarray including at least four array elements distributed sequentially along a first direction; at least a portion of the main subarrays include a uniform linear array ULA subarray.

[0115] In other words, for any main subarray containing a ULA subarray, the number of array elements it contains is greater than the number of array elements contained in the ULA subarray it contains.

[0116] In some exemplary embodiments, for any main subarray containing a ULA subarray, the ULA subarray is configured to perform angle resolution operations in a first direction during radar signal processing, and the main subarray is configured to perform deblurring operations on the angle resolution results. This effectively improves the resolution of angle measurement while retaining the advantage of low ULA sidelobes.

[0117] This disclosure also provides an antenna array, including: multiple receiving antennas and multiple transmitting antennas, wherein the multiple receiving antennas are arranged sequentially along a first direction to form a row of non-equally spaced receiving antenna arrays, wherein some of the receiving antennas in the receiving antenna array are distributed sequentially at equal intervals; the multiple transmitting antennas are distributed in multiple rows, wherein some of the transmitting antennas are distributed at equal intervals along the first direction, but as a whole, they are arranged non-equally spaced in the first direction.

[0118] The antenna array in the embodiments of this application is described in detail below with reference to the accompanying drawings. The detailed description is based on a dual-chip cascade scheme. However, the actual antenna array can be applied to scenarios other than dual-chip cascade schemes, such as single-chip, three-chip cascade, and four-chip cascade. This application does not impose specific limitations on these scenarios.

[0119] For example, Figure 15a The antenna array shown includes 8 transmitting antennas and 8 receiving antennas. Two SoC chips are cascaded as Master and Slave. Since the transmitting port of the SoC chip is located on top of the chip, while the receiving part is symmetrically distributed on the bottom of the chip, and considering that the shorter the antenna feed line, the smaller the impact on performance, the 8 receiving antennas are arranged in a row according to the spacing shown in the figure. That is, the six antennas Rx0 to Rx5 are arranged at equal intervals of 2λ between adjacent antennas, while Rx5 to Rx7 are arranged at non-equal intervals, that is, the spacing between Rx5 and Rx6 is 2.5λ, and the spacing between Rx6 and Rx7 is 3λ; where λ is the wavelength of the signal transmitted by the transmitting antenna or the wavelength corresponding to the (equivalent) center frequency. The eight transmitting antennas Tx0 to Tx7 can be distributed below the chip in five rows. Along the first direction (as shown in the horizontal direction), the spacing between Tx0 and Tx1, and between Tx6 and Tx7, is λ. Tx1 to Tx6 are arranged at equal intervals of 1.5λ. Therefore, although Tx0 to Tx7 are not equally spaced in the first direction, some of the transmitting antennas are evenly spaced, thus forming... Figure 15b The virtual array shown, i.e. the formed virtual array, includes a sparse principal subarray, which contains a ULA subarray for subsequent angle estimation operations.

[0120] In some exemplary embodiments, the integrated circuit is integrated into a millimeter-wave radar receiver, or into a smart vehicle device. This is not limited to any particular aspect. The integrated circuit may be located externally or internally of the device body. The device body may be a component or product used in fields such as smart vehicle systems and autonomous driving.

[0121] refer to Figure 15a As shown, an 8-transmit, 8-receive antenna array can be arranged around two cascaded SoC chips (if the SoC chip supports 8 transmit and 8 receive, a single chip can also be used as a component of the radar system; the number of chips is not limited here). That is, the antenna array can include 8 transmitting antennas and 8 receiving antennas. Since the transmitting port of the SoC chip is located on top of the chip, while the receiving part is symmetrically distributed below the chip, and considering that the shorter the antenna feed line has a smaller impact on performance, the two chips can be arranged side by side in the first direction. At the same time, the 8 receiving antennas are arranged in a row according to the spacing shown in the figure. That is, the six adjacent antennas Rx0 to Rx5 are arranged at equal intervals of 2λ, while the antennas Rx5 to Rx7 are arranged at non-equal intervals, that is, the distance between Rx5 and Rx6 is 2.5λ, and the distance between Rx6 and Rx7 is 3λ; where λ is the wavelength of the transmitted signal or the wavelength corresponding to the (equivalent) center frequency point. The eight transmitting antennas Tx0 to Tx7 can be distributed below the chip in five rows. Along the first direction (as shown in the horizontal direction), the spacing between Tx0 and Tx1, and between Tx6 and Tx7, is λ. Tx1 to Tx6 are arranged at equal intervals of 1.5λ. Therefore, although Tx0 to Tx7 are not equally spaced in the first direction, some of the transmitting antennas are still equally spaced, thus forming... Figure 15b The virtual array shown contains a sparse principal subarray, which includes a ULA subarray for subsequent angle estimation operations.

[0122] In some alternative embodiments, unequal lengths of the Rx feed lines can be used to further reduce the feed line length (or loss), and the PCB layout can be based on different chip functions. For example, if the master chip in the cascaded chip is mainly configured for target detection, then... Figure 15a Based on the antenna array shown, an unequal length Rx feed line is used to further reduce the feed line length of the Master chip, thereby minimizing the feed line loss and improving the stability of the sensor system's core performance. Specifically, Figure 15a The antenna array positions are shown in the table below: Table 1 - Antenna Location Table 1

[0123] Right now Figure 15a The antenna array shown is based on MIMO formation. Figure 15b The virtual array shown (i.e., channel arrangement) has 5 rows (or columns) of antennas in the second direction (e.g., elevation). The main subarray contains 32 elements (i.e., the 32-element main subarray shown in the figure), which can be divided into two parts: Rx0~Rx5 and Tx0, Tx1, Tx6, Tx7 MIMO to form a 24-element ULA subarray (i.e., the 24-element ULA with an interval of λ shown in the figure). The parameters that can be achieved for angle measurement are: maximum unambiguous angle range of 60°, -3dB beamwidth of 2.1°, and sidelobe level of -13dB. With Taylor windowing, the -3dB beamwidth can be widened to 2.8°, but the sidelobe level decreases to -35dB (see details). Figure 15c If the ULA subarray has angular ambiguity, it can be resolved by using the main subarray (i.e., all 32 elements). The DBF results can be found in [reference needed]. Figures 15d-15e As shown, after angle de-ambiguity, the sensor no longer exhibits angle ambiguity within the entire FOV.

[0124] Additionally, see Figure 15f As shown, based on Figure 15a The antenna array shown, after first combining the azimuth signal, allows the MIMO virtual channel's elevation DBF to achieve a -3dB elevation beamwidth of 4°, while suppressing the elevation sidelobe level to -5dB. Simultaneously, the amplitude and phase errors of the ULA subarray affect the sidelobe level. Figure 15f As shown, the effect of the position error of the ULA subarray antenna on the sidelobe level can be found in [reference needed]. Figure 15g As shown.

[0125] In summary, see Figures 15a-15g As shown, for multipath point identification scenarios, DoA (Rx channel resolution) and DoD (Tx channel resolution) can be used for... Figure 15a The antenna array shown is a 2λ array with 8 Tx elements. To address the angular ambiguity, the DoD angle can be calculated using DBF (or phase comparison) across the 8 Tx channels. The main array of Rx elements (Rx0, Rx1, Rx6, Rx7) can be solved using phase comparison or DBF. Here, DOA (Direction of Arrival) refers to the direction the signal arrives at the antenna, i.e., the direction of arrival, while DoD (Direction of Departure) refers to the direction the signal is emitted from the antenna, i.e., the direction of beamforming.

[0126] In some other optional embodiments of this application, another antenna array configuration is also provided, such as... Figure 15h As shown, an antenna array based on a microstrip antenna on a PCB is illustrated, with reference to... Figure 15a The related explanations state that by arranging the transmitting antennas (Tx) in one row and distributing the receiving antennas (Rx) in five rows, and to minimize the feed line length, the placement of the SoC chips can be adjusted. Specifically, the master chip in the cascaded chips is placed upright, while the slave chip is placed at an angle. To further reduce the feed line length, the physical position and orientation of the two cascaded SoC chips can be adjusted based on the antenna array arrangement. Optionally, in... Figure 15h In the diagram, the port for the transmitting antenna Tx is located at the top of the chip (i.e., in the direction of the text in the reference diagram), while the port for the receiving antenna Rx is located at the bottom of the chip.

[0127] because Figure 15h Related content and Figure 15a It is quite similar; please refer to the following for details. Figure 15a The related explanations involve swapping the descriptions of the receiving and transmitting antennas. The transmitting antennas Tx0~Tx7 are arranged in one row, while the receiving antennas Rx0~Rx7 are arranged in five rows, thus forming... Figure 15i The MIMO virtual array shown is readily understood by those skilled in the art by referring to the above content, and therefore will not be described in detail here.

[0128] based on Figure 15h The antenna array layout and antenna positions are shown in the table below: Table 2 - Antenna Location Table 2

[0129] See also Figure 15h , Figure 15i As shown in the table above, the antenna has 5 rows (followed by 5 columns) in elevation. The resulting MIMO virtual array contains a sparse master subarray with 32 elements (i.e., the 32-element master subarray shown in the figure). This master subarray can also be divided into two parts: Rx0~Rx4 and Tx0, Tx1, Tx3, Tx4, Tx6, and Tx7. After MIMO, a ULA subarray is formed (i.e., the 24-element ULA with a spacing of λ shown in the figure). The DBF results for this ULA subarray can be found in [reference needed]. Figure 15jAs shown, its maximum angular unambiguous range is 60°, the -3dB beamwidth is 2.1°, and the sidelobe level is -13dB. With Taylor windowing, the -3dB beamwidth can be widened to 2.8°, but the sidelobe level decreases to -35dB. This indicates that the ULA subarray exhibits angular ambiguity. Therefore, angular deambiguity can be achieved using the main subarray (i.e., all 32 elements). The DBF results can be found in [reference needed]. Figure 15k As shown, there is no angular ambiguity issue within the entire FOV at this point. Meanwhile, the ULA array and the pitch DBF scheme in the above embodiments are similar; please refer to the relevant figures and descriptions for details, which will not be repeated here.

[0130] In some other optional embodiments of this application, another antenna array configuration is also provided, such as... Figure 15 As shown, an antenna array based on a RoP (Radiator on Package) antenna on a PCB is illustrated, referencing... Figure 15a Furthermore, the receiving antennas (Rx) can be arranged in a cross shape at the center of the wiring area on the PCB, while the transmitting antennas (Tx) are placed at the edges and corners of the wiring area. This maximizes the utilization of the wiring area and results in a larger antenna aperture. Since RoP antennas are used, their loss is minimal, so feeder loss is less of a concern, making the antenna array more flexible and maximizing the use of the wiring area. In another optional embodiment, the transmitting antenna can be placed in the central area, while the receiving antennas are placed at the edges and corners of the wiring area. Alternatively, they can be distributed according to requirements, as long as the resulting MIMO virtual array meets the requirements and maximizes the use of the wiring area space.

[0131] refer to Figure 15 In an optional antenna array configuration that simultaneously extends the azimuth and elevation apertures via MIMO, the corresponding antenna positions are shown in the table below: Table 3 - Antenna Location Table 3

[0132] based on Figure 15 The antenna array can form Figure 15mThe virtual array shown, i.e., the antenna array, forms 9 rows (or columns) in elevation. The sparse main subarray can include 24 elements (i.e., the row of 24 elements clearly marked in red in the figure). It can include two parts: a ULA subarray with a spacing of 1.5λ formed after MIMO of Rx0, Rx1, Rx2, Rx5, Rx7 and Tx2, Tx3, Tx4, Tx5, that is, this ULA subarray includes 20 equally spaced elements. The DBF results of the corresponding ULA subarray can be found in [reference needed]. Figure 15n As shown, its maximum unambiguous angular range is 41.8°, the -3dB beamwidth is 1.7°, and the sidelobe level is -13dB. Furthermore, with Taylor windowing, its -3dB beamwidth can be widened to 2.3°, but its sidelobe level decreases to -35dB. In this case, the ULA subarray exhibits angular ambiguity; therefore, angular deambiguity can be achieved using a 24-element main subarray. The DBF results are shown below. Figure 15o As shown, there is no angle blurring problem within the entire FOV at this point.

[0133] See Figure 15p As shown, after combining the azimuth signal, the elevation DBF results of the virtual channel show a -3dB beamwidth of 1.8° in the elevation direction and a sidelobe level of -5.6dB. Meanwhile, the effect of the amplitude and phase errors of the ULA subarray on the sidelobe level can be found in [reference needed]. Figure 15q As shown, the effect of the position error of the ULA subarray on the sidelobe level can be found in [reference needed]. Figure 15r As shown.

[0134] Additionally, in some optional embodiments, for multipath point identification scenarios, DoA (Rx channel resolution) and DoD (Tx channel resolution) can also be used. Specifically, for scenarios such as... Figure 15 The array configuration shown allows for the following calculations for azimuth multipath: DoA angles can be obtained using a Tx-based main array (Tx2, Tx3, Tx4, Tx5) via phase comparison or DBF; DoD angles can be calculated using DBF across 8 Rx channels. Multipath detection can be performed under different ambiguity factor assumptions. For elevation multipath, DoA angles can be obtained using DBF with TX0, TX1, and TX2; DoD angles can be obtained using DBF with RX2, RX3, and RX4.

[0135] A radar system composed of multiple radar chips can achieve a larger detection range and a wider detection angle. The design of the radar system's antenna array directly affects the angular resolution of the Direction of Arrival (DOA) algorithm. Typically, increasing the number of antennas directly enlarges the array aperture to improve angular resolution. However, the physical space available for antenna array placement is generally limited. Therefore, MIMO technology is needed to exponentially increase the number of virtual array elements with a relatively small number of antennas, forming a large-aperture virtual array and improving the angular resolution of the radar system. Optionally, the radar system of this invention includes at least two cascaded radar chips, such as two cascaded radar chips, three cascaded radar chips, four cascaded radar chips, five cascaded radar chips, or six cascaded radar chips.

[0136] This embodiment of the invention takes a radar system including a first radar chip and a second radar chip as an example. The first radar chip and the second radar chip respectively include a transmitting antenna array and a receiving antenna array. Figure 15s1 As shown, the transmitting antenna array of the first radar chip includes four transmitting antennas, namely Tx0, Tx1, Tx2, and Tx3, and the receiving antenna array of the first radar chip includes four receiving antennas, namely Rx0, Rx1, Rx2, and Rx3. The transmitting antenna array of the second radar chip includes four transmitting antennas, namely Tx4, Tx5, Tx6, and Tx7, and the receiving antenna array of the second radar chip includes four receiving antennas, namely Rx4, Rx5, Rx6, and Rx7.

[0137] like Figure 15s1 As shown, in the first direction, transmitting antennas Tx0, Tx1, and Tx2 are arranged at equal intervals, with a spacing of 2.5λ between adjacent transmitting antennas, and a spacing of 3λ between transmitting antennas Tx2 and Tx3; the spacing between transmitting antenna Tx3 of the first radar chip and transmitting antenna Tx4 of the second radar chip is 3.5λ; the spacing between transmitting antennas Tx4 and Tx5 is 3.5λ; transmitting antennas Tx5, Tx6, and Tx7 are arranged at equal intervals, with a spacing of 1.5λ between adjacent transmitting antennas.

[0138] like Figure 15s1 As shown, in the first direction, receiving antennas Rx0, Rx1, Rx2, and Rx3 are equally spaced, and the distance between adjacent receiving antennas is 2λ; the distance between the receiving antenna Rx3 of the first radar chip and the receiving antenna Rx4 of the second radar chip is 7λ; the distance between receiving antennas Rx4 and Rx5 is 4λ; receiving antennas Rx5, Rx6, and Rx7 are equally spaced, and the distance between adjacent receiving antennas is 2λ.

[0139] like Figure 15s1As shown, in the second direction perpendicular to the first direction, the transmitting antennas Tx0, Tx1, Tx2, Tx3, Tx4, Tx5, Tx6, and Tx7 are arranged with no gap between their phase centers; in the second direction, the receiving antennas Rx0, Rx1, Rx2, and Rx3 also have no gap between their phase centers.

[0140] like Figure 15s1 As shown, in the second direction, the phase center of receiving antenna Rx4 is 2λ away from the phase center of receiving antenna Rx3, the phase center of receiving antenna Rx5 is 4λ away from the phase center of receiving antenna Rx4, the phase center of receiving antenna Rx5 is 2λ away from the phase center of receiving antenna Rx3, the phase center of receiving antenna Rx6 is 2λ away from the phase center of receiving antenna Rx5, and the phase center of receiving antenna Rx6 is 2λ away from the phase center of receiving antenna Rx5. Here, λ represents the wavelength of the electromagnetic wave signal transmitted by the transmitting antenna and received by the receiving antenna.

[0141] Figure 15s2 The MIMO virtual array of the 8T8R radar system according to this embodiment is shown, such as Figure 15s2 As shown, it contains 64 virtual channels, and the position of each virtual channel is calculated by convolution operation between discrete antenna elements of the antenna array.

[0142] like Figure 15s2 As shown, the MIMO virtual array includes five elevation subarrays: array1, array2, array3, array4, and array5. The main subarray array4 contains 32 channels. The virtual channels in the main subarray array4 are obtained by convolution operations between the transmit antennas Tx0, Tx1, Tx2, Tx3, Tx4, Tx5, Tx6, and Tx7 and the receive antennas Rx0, Rx1, Rx2, and Rx3.

[0143] Figure 15s3 The azimuth DBF results of the main subarray array4 are shown, where the -3dB beamwidth is 2.2° and the sidelobe level is -12.6dB.

[0144] Figure 15s4 The DBF results for the elevation subarrays arrays 1-5 are shown. The elevation beamwidth is -3dB at 5.2°, and the elevation sidelobes are horizontally -12dB. In this embodiment, due to the presence of the elevation grating lobes, the sidelobes entering the target require back-end processing by the signal processing unit.

[0145] In this invention, all transmitting antennas Tx have the same elevation, and all receiving antennas of the first radar chip have the same elevation. Coherent accumulation of the channels is achieved through the 8T4R of the first radar chip, thereby improving the signal-to-noise ratio and reducing interference caused by signal frequency shift and Doppler effect.

[0146] This embodiment of the invention takes a radar system including a first radar chip and a second radar chip as an example. The first radar chip and the second radar chip respectively include a transmitting antenna array and a receiving antenna array. Figure 15s5 As shown, the transmitting antenna array of the first radar chip includes four transmitting antennas, namely Tx0, Tx1, Tx2, and Tx3, and the receiving antenna array of the first radar chip includes four receiving antennas, namely Rx0, Rx1, Rx2, and Rx3. The transmitting antenna array of the second radar chip includes four transmitting antennas, namely Tx4, Tx5, Tx6, and Tx7, and the receiving antenna array of the second radar chip includes four receiving antennas, namely Rx4, Rx5, Rx6, and Rx7.

[0147] like Figure 15s5 As shown, in the first direction, the spacing between transmitting antennas Tx0 and Tx1, Tx1 and Tx2, Tx2 and Tx3, and Tx3 and Tx4 is 3λ, 3.5λ, 3λ, and 3.5λ, respectively, and the spacing between Tx4 and Tx5, Tx5 and Tx6, and Tx6 and Tx7 is 1.5λ. In the second direction, transmitting antennas Tx0, Tx1, Tx2, Tx3, Tx4, Tx5, Tx6, and Tx7 are arranged with no spacing between their phase centers.

[0148] like Figure 15s5 As shown, in the first direction, receiving antennas Rx0, Rx1, Rx2, and Rx3 are arranged at equal intervals, and the distance between adjacent receiving antennas is 2λ; in the first direction, the intervals between receiving antennas Rx3 and Rx4, and between Rx4 and Rx5 are 5λ and 4λ, respectively, and receiving antennas Rx5, Rx6, and Rx7 are arranged at equal intervals, and the distance between adjacent receiving antennas is 2λ.

[0149] In the second direction, receiving antennas Rx0, Rx1, Rx2, and Rx3 are arranged with no gap between their phase centers. The phase centers of receiving antennas Rx3 and Rx4, and Rx3 and Rx5 are spaced 1.9λ apart. The phase centers of receiving antennas Rx4 and Rx5 are spaced 3.8λ apart. The phase centers of receiving antennas Rx5, Rx6, and Rx7 are spaced equally, and the distance between the phase centers of adjacent receiving antennas is 1.9λ.

[0150] Figure 15s6 The MIMO virtual array of the 8T8R radar system according to this embodiment is shown, such as Figure 15s6As shown, it contains 64 virtual channels, and the position of each virtual channel is calculated by convolution operation between discrete antenna elements of the antenna array.

[0151] like Figure 15s6 As shown, the MIMO virtual array includes five elevation subarrays: array1, array2, array3, array4, and array5. The main subarray array4 is a 32-element sparse array, while the elevation subarrays are 5-element uniform linear arrays (ULA). The virtual channels in the main subarray array4 are obtained by convolution operations between the transmit antennas Tx0, Tx1, Tx2, Tx3, Tx4, Tx5, Tx6, and Tx7 and the receive antennas Rx0, Rx1, Rx2, and Rx3.

[0152] Figure 15s7 The azimuth DBF results of the main subarray array4 are shown, where the -3dB beamwidth is 2.3° and the sidelobe level is -12.6dB.

[0153] Figure 15s8 The DBF results for the elevation subarrays array1-array5 are shown. The elevation beamwidth is 5.2° at -3dB and the elevation sidelobe horizontal beamwidth is -13dB.

[0154] like Figure 15s3 and Figure 15s7 As shown, this invention exhibits low DBF sidelobes and a small -3dB beamwidth, meeting the requirements for DOA resolution. However, when using deterministic maximum likelihood (DML) for DOA estimation to meet both large FOV (Field of View) and high resolution requirements, this invention leads to a surge in the number of steering vectors and / or FFT points, resulting in excessive memory and computational burden. This invention addresses this by reducing the number of coarse search points. For example, with an FOV requirement of ±60° and a resolution requirement of 1°, typically, to avoid getting trapped in local extrema, the number of coarse search points within the resolution must be greater than 2. Therefore, the number of DML coarse search points would be at least 240. This invention reduces the number of coarse search points, alleviating the memory and computational burden while maintaining a large FOV and high resolution requirements.

[0155] like Figure 15s2 and Figure 15s6As shown, the main subarray in this invention is a sparse array. Compared with traditional arrays, sparse arrays have significant advantages in improving estimation accuracy, enhancing angular resolution, and reducing physical costs. Typically, a larger element spacing can further expand the virtual aperture, thereby increasing the number of detectable sources and angular resolution. Increasing the element spacing can also effectively suppress element mutual coupling effects and receiver noise coherence, thus improving estimation accuracy. Lower element redundancy can effectively reduce resource waste and improve computation speed. At the same time, due to the presence of holes, sparse arrays cannot use some fast computation methods, such as spatial smoothing and matrix inversion (e.g., the IAA algorithm, Iterative Adaptive Approach). Moreover, its FFT computation efficiency is lower than that of ULA arrays, and the computation time is longer. In addition, the lower the sidelobe level, the better the signal processing unit performs in the initial source number estimation (Rough Number Estimate / Reflector Number Estimator, RNE).

[0156] In this embodiment of the invention, a co-prime array is used in the antenna array design. This embodiment still takes a radar system including a first radar chip and a second radar chip as an example. The first radar chip and the second radar chip respectively include a transmitting antenna array and a receiving antenna array. Figure 15s9 As shown, the transmitting antenna array of the first radar chip includes four transmitting antennas, namely Tx0, Tx1, Tx2, and Tx3, and the receiving antenna array of the first radar chip includes four receiving antennas, namely Rx0, Rx1, Rx2, and Rx3. The transmitting antenna array of the second radar chip includes four transmitting antennas, namely Tx4, Tx5, Tx6, and Tx7, and the receiving antenna array of the second radar chip includes four receiving antennas, namely Rx4, Rx5, Rx6, and Rx7.

[0157] like Figure 15s9 As shown, in the first direction, transmitting antennas Tx0, Tx1, and Tx2 are arranged at equal intervals, with a spacing of 2.5λ between adjacent transmitting antennas; transmitting antennas Tx3, Tx4, Tx5, Tx6, and Tx7 are arranged at equal intervals, with a spacing of 2λ between adjacent transmitting antennas; in the second direction, transmitting antennas Tx0, Tx1, Tx2, Tx3, Tx4, Tx5, Tx6, and Tx7 are arranged with no spacing between their phase centers.

[0158] like Figure 15s9 As shown, in the first direction, receiving antennas Rx0, Rx1, Rx2, and Rx3 are arranged at equal intervals, and the distance between adjacent receiving antennas is 2.5λ. In the second direction, receiving antennas Rx0, Rx1, Rx2, and Rx3 are arranged at equal intervals, and the distance between the phase centers of adjacent receiving antennas is 1.9λ.

[0159] like Figure 15s9 As shown, in the first direction, receiving antennas Rx0 and Rx4, and Rx4 and Rx7 are equally spaced with a spacing of 10λ; in the second direction, receiving antennas Rx4, Rx5, and Rx6 are equally spaced with a spacing of 3λ, and receiving antennas Rx6 and Rx7 are spaced 4λ apart. In the third direction, receiving antennas Rx0, Rx4, and Rx7 are arranged with no gap between their phase centers, the phase centers of receiving antennas Rx4 and Rx5 are spaced 9.5λ apart, and the phase centers of receiving antennas Rx4 and Rx6 are spaced 7.6λ apart.

[0160] Figure 15s10 The MIMO virtual array of the 8T8R radar system according to this embodiment is shown, such as Figure 15s10 As shown, it contains 51 virtual channels. The MIMO virtual array includes a master subarray, a coproton array 1, and a coproton array 2. The master subarray is a 24-element sparse array, the coproton array 1 is a 15-element ULA array with a spacing of 2λ, and the coproton array 2 is a 12-element ULA array with a spacing of 2.5λ.

[0161] Figure 15s11 The azimuth DBF results of the main subarray are shown, with a -3dB beamwidth of 1.6° and a sidelobe level of -6dB.

[0162] Figure 15s12 DBF results for a 6-element ULA elevation subarray (1.9λ) are shown, with a -3dB elevation beamwidth of 4.3° and a horizontal sidelobe elevation of -13dB.

[0163] at the same time, Figure 15s13 and Figure 15s14 The orientation DBF results for coproton array 1 and coproton array 2 are shown respectively. Figure 15s13 The unambiguous range of the intermediate proton array 1 is [-15°, 15°], the -3dB beamwidth is 1.7°, and 2.3° with Taylor windowing. The sidelobe horizontal width is -13dB, and can reach -35dB with Taylor windowing. Figure 15s14 The unambiguous range of the inter-proton array 2 is [-11.8°, 11.8°], with a sidelobe level of -13dB, which can reach -35dB with a Taylor window. The -3dB beamwidth is 1.7°, and 2.3° with a Taylor window.

[0164] In this embodiment, a coprime array is used for antenna array design. At the same time, the coprime relationship of the coprime subarrays is used to resolve angular ambiguity and improve the angular resolution of target recognition.

[0165] This embodiment takes a radar system including a first radar chip and a second radar chip as an example. The first radar chip and the second radar chip respectively include a transmitting antenna array and a receiving antenna array. Figure 15s15 As shown, the transmitting antenna array of the first radar chip includes four transmitting antennas, namely Tx0, Tx1, Tx2, and Tx3, and the receiving antenna array of the first radar chip includes four receiving antennas, namely Rx0, Rx1, Rx2, and Rx3. The transmitting antenna array of the second radar chip includes four transmitting antennas, namely Tx4, Tx5, Tx6, and Tx7, and the receiving antenna array of the second radar chip includes four receiving antennas, namely Rx4, Rx5, Rx6, and Rx7.

[0166] like Figure 15s15 As shown, in the first direction, transmitting antennas Tx0, Tx1, Tx2, Tx3, Tx4, Tx5, and Tx6 are arranged at equal intervals, with a spacing of 2λ between adjacent transmitting antennas, and transmitting antennas Tx6 and Tx7 are arranged at a spacing of 1.5λ; in the second direction, transmitting antennas Tx0, Tx1, Tx2, Tx3, Tx4, Tx5, Tx6, and Tx7 are arranged with no spacing between their phase centers.

[0167] like Figure 15s15 As shown, in the first direction, the spacing between receiving antennas Rx0 and Rx1, and Rx6 and Rx7 is λ. In the second direction, receiving antennas Rx0, Rx1, Rx6, and Rx7 are arranged with no spacing between their phase centers.

[0168] like Figure 15s15 As shown, in the first direction, receiving antennas Rx1, Rx2, and Rx3 are equally spaced with a spacing of 1.5λ; in the first direction, receiving antennas Rx4, Rx5, and Rx6 are equally spaced with a spacing of 1.5λ; in the first direction, receiving antennas Rx3 and Rx4 are spaced 7λ apart.

[0169] like Figure 15s15 As shown, in the second direction, the phase centers of receiving antennas Rx1 and Rx2, Rx2 and Rx3, Rx3 and Rx5, and Rx5 and Rx4 are all spaced 1.9λ apart.

[0170] Figure 15s16 The MIMO virtual array of the 8T8R radar system according to this embodiment is shown, such as Figure 15s16 As shown, it contains 64 virtual channels, the position of which is calculated by convolution operations between discrete antenna elements of the antenna array. Figure 15s16As shown, the MIMO virtual array includes 5 elevation subarrays, of which the main subarray is a 32-element sparse array. The virtual channels in the main subarray are obtained by convolution operation between the transmit antennas Tx0, Tx1, Tx2, Tx3, Tx4, Tx5, Tx6, Tx7 and the receive antennas Rx0, Rx1, Rx6, Rx7.

[0171] Hidden within the main subarray is a 28-element ULA array. The virtual channels of this ULA array are obtained through convolution operations between transmit antennas Tx0, Tx1, Tx2, Tx3, Tx4, Tx5, and Tx6 and receive antennas Rx0, Rx1, Rx6, and Rx7. In the DOA estimation process, the signal processing unit needs to obtain a list of possible ambiguous angles from the angle resolution results of the ULA array. Then, by performing orthogonal projection (DML) on this list across the entire main subarray, the combination corresponding to the highest likelihood probability is obtained, which is the final azimuth estimation list.

[0172] The array design in this embodiment retains the characteristics of ULA with low sidelobe levels and the ability to add traditional signal processing windows, while also making it feasible to improve resolution and dynamic range. Furthermore, the FIAA (Fast IAA, Fast Iterative Adaptive Algorithm) algorithm can be used to accelerate the super-resolution algorithm to ensure the frame rate of the entire radar signal processing.

[0173] Figure 15s17 The results of the 28-element ULA azimuth DBF are shown, with an unambiguous range of [-30, 30]°, a sidelobe level of -13dB (down to -35dB with Taylor windowing), a -3dB beamwidth of 1.8° (2.4° with Taylor windowing).

[0174] Figure 15s18 The DBF results for a 5-element ULA elevation subarray (1.9λ) are shown, with a -3dB elevation beamwidth of 5.2° and a horizontal sidelobe elevation of -13dB.

[0175] This embodiment takes a radar system including a first radar chip and a second radar chip as an example. The first radar chip and the second radar chip respectively include a transmitting antenna array and a receiving antenna array. Figure 15s19 As shown, the transmitting antenna array of the first radar chip includes four transmitting antennas, namely Tx0, Tx1, Tx2, and Tx3, and the receiving antenna array of the first radar chip includes four receiving antennas, namely Rx0, Rx1, Rx2, and Rx3. The transmitting antenna array of the second radar chip includes four transmitting antennas, namely Tx4, Tx5, Tx6, and Tx7, and the receiving antenna array of the second radar chip includes four receiving antennas, namely Rx4, Rx5, Rx6, and Rx7.

[0176] like Figure 15s19 As shown, in the first direction, receiving antennas Rx0, Rx1, Rx2, Rx3, Rx4, and Rx5 are arranged at equal intervals, the distance between adjacent transmitting antennas is 2λ, receiving antennas Rx5 and Rx6 are arranged at a distance of 2.5λ, and receiving antennas Rx6 and Rx7 are arranged at a distance of 3λ; in the second direction, receiving antennas Rx0, Rx1, Rx2, Rx3, Rx4, Rx5, Rx6, and Rx7 are arranged with no gap at the phase center.

[0177] like Figure 15s19 As shown, in the first direction, the spacing between transmitting antennas Tx0 and Tx1, and Tx6 and Tx7 is λ. In the second direction, transmitting antennas Tx0, Tx1, Tx6, and Tx7 are arranged with no spacing between their phase centers.

[0178] like Figure 15s19 As shown, in the first direction, transmitting antennas Tx1, Tx2, and Tx3 are equally spaced with a spacing of 1.5λ; in the first direction, transmitting antennas Tx4, Tx5, and Tx6 are equally spaced with a spacing of 1.5λ; in the first direction, transmitting antennas Tx3 and Tx4 are spaced 4λ apart.

[0179] like Figure 15s19 As shown, in the second direction, the phase centers of transmitting antennas Tx1 and Tx2, Tx2 and Tx3, Tx3 and Tx5, and Tx5 and Tx4 are spaced apart by λ, 3λ, 2λ, and 4λ respectively.

[0180] Figure 15s20 The MIMO virtual array of the 8T8R radar system according to this embodiment is shown, such as Figure 15s20 As shown, it contains 64 virtual channels, the position of which is calculated by convolution operations between discrete antenna elements of the antenna array. Figure 15s20 As shown, the MIMO virtual array includes five elevation subarrays, with the main subarray being a 32-element sparse array. The virtual channels in this main subarray are obtained by convolution operations between the transmit antennas Rx0, Rx1, Rx2, Rx3, Rx4, Rx5, Rx6, and Rx7 and the receive antennas Tx0, Tx1, Tx6, and Tx7. Hidden within the main subarray is a 24-element ULA array. The virtual channels of this ULA array are formed by convolution operations between the receive antennas Rx0, Rx1, Rx2, Rx3, Rx4, and Rx5 and the transmit antennas Tx0, Tx1, Tx6, and Tx7, resulting in a 24-element ULA with a spacing of λ.

[0181] Figure 15s21The azimuth DBF results for a 24-element ULA are shown. The maximum unambiguous angular range is 60°, with a horizontal sidelobe of -13dB, which can be reduced to -35dB with a Taylor window. The -3dB beamwidth is 1.8°, and 2.4° with a Taylor window. The -3dB beamwidth is 2.1°, with a horizontal sidelobe of -13dB. With a Taylor window, the -3dB beamwidth widens to 2.8°, but the horizontal sidelobe decreases to -35dB. Because angular ambiguity exists in the main subarray, we need to perform angular ambiguity resolution in the entire main subarray MIMO. The DBF results are shown below. Figure 15s22 As shown, there is no angular blurring within the entire field of view.

[0182] This embodiment takes a radar system including a first radar chip and a second radar chip as an example. The first radar chip and the second radar chip respectively include a transmitting antenna array and a receiving antenna array. Figure 15s23 As shown, the transmitting antenna array of the first radar chip includes four transmitting antennas, namely Tx0, Tx1, Tx2, and Tx3, and the receiving antenna array of the first radar chip includes four receiving antennas, namely Rx0, Rx1, Rx2, and Rx3. The transmitting antenna array of the second radar chip includes four transmitting antennas, namely Tx4, Tx5, Tx6, and Tx7, and the receiving antenna array of the second radar chip includes four receiving antennas, namely Rx4, Rx5, Rx6, and Rx7.

[0183] like Figure 15s23 As shown, in the first direction, the spacing between transmitting antennas Tx0 and Tx1, Tx1 and Tx2, Tx2 and Tx3, Tx3 and Tx4, Tx4 and Tx5, Tx5 and Tx6, and Tx6 and Tx7 is λ, 1.5λ, 5.5λ, λ, 2.5λ, 4.5λ, and λ, respectively; transmitting antennas Tx0, Tx1, Tx2, Tx3, Tx4, Tx5, Tx6, and Tx7 are arranged with no spacing between their phase centers.

[0184] like Figure 15s23 As shown, in the first direction, receiving antennas Rx0, Rx1, Rx2, and Rx3 are arranged at equal intervals, and the distance between adjacent receiving antennas is 2λ; in the first direction, the intervals between receiving antennas Rx3 and Rx4, Rx4 and Rx5, Rx5 and Rx6, and Rx6 and Rx7 are 7λ, 2λ, 4λ, and 2λ, respectively.

[0185] like Figure 15s23 As shown, in the second direction, receiving antennas Rx0, Rx1, Rx2, and Rx3 are arranged with no gap between their phase centers, while the phase centers of receiving antennas Rx3 and Rx4, Rx3 and Rx5, Rx5 and Rx6, and Rx6 and Rx7 are spaced apart by 3λ, 1.5λ, 4.5λ, and 4.5λ, respectively.

[0186] Figure 15s24 The MIMO virtual array of the 8T8R radar system according to this embodiment is shown, such as Figure 15s24 As shown, it contains 64 virtual channels, the position of which is calculated by convolution operations between discrete antenna elements of the antenna array. Figure 15s24 As shown, the MIMO virtual array includes five elevation subarrays, with the main subarray being a 32-element sparse array. The virtual channels in this main subarray are obtained by convolution operations between the transmit antennas Tx0, Tx1, Tx2, Tx3, Tx4, Tx5, Tx6, and Tx7 and the receive antennas Rx0, Rx1, Rx2, and Rx3. Hidden within the main subarray is a 24-element ULA array. The virtual channels of this ULA array are formed by convolution operations between the receive antennas Rx0, Rx1, Rx2, and Rx3 and the transmit antennas Tx0, Tx1, Tx3, Tx4, Tx6, and Tx7, resulting in a 24-element ULA with a spacing of λ.

[0187] Figure 15s25 The azimuth DBF results for the 24-element ULA are shown. Its maximum angular unambiguous range is 60°, the -3dB beamwidth is 2.1°, and the sidelobe level is -13dB. With Taylor windowing, the -3dB beamwidth widens to 2.8°, but the sidelobe level decreases to -35dB. Due to angular ambiguity in the main subarray, angular deambiguity needs to be performed on all main subarray MIMOs. The DBF results are shown below. Figure 15s26 As shown, there is no angular blurring within the entire field of view.

[0188] This embodiment takes a radar system including a first radar chip and a second radar chip as an example. The first radar chip and the second radar chip respectively include a transmitting antenna array and a receiving antenna array. Figure 15s27 As shown, the transmitting antenna array of the first radar chip includes four transmitting antennas, namely Tx0, Tx1, Tx2, and Tx3, and the receiving antenna array of the first radar chip includes four receiving antennas, namely Rx0, Rx1, Rx2, and Rx3. The transmitting antenna array of the second radar chip includes four transmitting antennas, namely Tx4, Tx5, Tx6, and Tx7, and the receiving antenna array of the second radar chip includes four receiving antennas, namely Rx4, Rx5, Rx6, and Rx7.

[0189] like Figure 15s27As shown, in the first direction, transmitting antennas Tx0, Tx1, and Tx2 are set without gaps, transmitting antennas Tx4, Tx6, and Tx7 are set without gaps, transmitting antennas Tx0 and Tx6 are set with a gap of 16.5λ, and transmitting antennas Tx0 and Tx3, Tx4 and Tx5 are set with a gap of 1.5λ; in the second direction, transmitting antennas Tx2, Tx3, Tx5, and Tx5 are set with no gaps between their phase centers, transmitting antennas Tx0 and Tx6 are set with no gaps between their phase centers, transmitting antennas Tx1 and Tx7 are set with no gaps between their phase centers, the phase centers of transmitting antennas Tx0 and Tx1, and Tx0 and Tx2 are set with gaps of 7.5λ and 6λ respectively, and the phase centers of transmitting antennas Tx6 and Tx7, and Tx6 and Tx4 are set with gaps of 7.5λ and 6λ respectively.

[0190] like Figure 15s27 As shown, in the first direction, the transmitting antenna Tx0 and the receiving antenna Rx0 are spaced 2.25λ apart. The receiving antennas Rx0, Rx1, Rx3, and Rx5 are equally spaced, with a spacing of 3λ between adjacent receiving antennas. The receiving antennas Rx5 and Rx6, and Rx6 and Rx7 are spaced 2λ and λ apart, respectively. In the first direction, the receiving antennas Rx2, Rx3, and Rx4 are not spaced apart. In the second direction, the receiving antennas Rx0, Rx1, Rx3, Rx5, Rx6, and Rx7 are arranged with no gap between their phase centers. The phase centers between the receiving antennas Rx3 and Rx2, and Rx3 and Rx4 are spaced 4.5λ and 10.5λ apart, respectively.

[0191] Figure 15s28 The MIMO virtual array of the 8T8R radar system according to this embodiment is shown, such as Figure 15s28 As shown, it contains 64 virtual channels, and the position of each virtual channel is calculated by convolution operation between discrete antenna elements of the antenna array.

[0192] like Figure 15s28 As shown, the MIMO virtual array includes 9 elevation subarrays, of which the main subarray has 24 elements (red circles in Figure 15s). The virtual channels in the main subarray are obtained by convolution operation between the receiving antennas Rx0, Rx1, Rx2, Rx5, Rx6, Rx7 and the transmitting antennas Tx2, Tx3, Tx5, Tx4.

[0193] Hidden within the main subarray is a 20-element ULA array. The virtual channels of this ULA array are formed by convolution operations between receiving antennas Rx0, Rx1, Rx2, Rx5, and Rx7 and transmitting antennas Tx2, Tx3, Tx5, and Tx4, resulting in a 20-element ULA with a spacing of 1.5λ. Its DBF result is as follows... Figure 15s29As shown, its maximum unambiguous angular range is 41.8°, the -3dB beamwidth is 1.7°, and the sidelobe level is -13dB. With Taylor windowing, the -3dB beamwidth widens to 2.3°, but the sidelobe level decreases to -35dB. Because the main subarray exhibits angular ambiguity, we need to perform angular ambiguity resolution across all main subarray MIMOs. The DBF results are as follows: Figure 15s30 As shown, there is no angular blurring within the entire field of view.

[0194] The radar system of the present invention further includes a control unit and a signal processing unit. The control unit provides parameters such as the signal slope and signal waveform of the transmitted electromagnetic wave signal, and the signal processing unit performs calculations on the received electromagnetic wave signal to obtain the distance, speed and angle of the target to be detected.

[0195] In any embodiment of the present invention, a constant false alarm rate (CFAR) detector is used by a first radar chip and / or a second radar chip to identify targets in clutter and noise. To ensure the accuracy of CFAR detection, the feed line lengths of the transmitting antenna array and receiving antenna array of the first radar chip and / or the second radar chip are minimized as much as possible.

[0196] In any embodiment of the present invention, the direction and / or position of the first radar chip and / or the second radar chip can be adaptively adjusted according to the actual antenna array design.

[0197] In any embodiment of the present invention, the feed lines of the receiving antenna array are designed to be of equal length, and the feed lines of the transmitting antenna array can be designed to be of equal or unequal length. Optionally, when the feed lines of the transmitting antenna array are of unequal length, DFE (Decision Feedback Equalizer) compensation is used.

[0198] This invention uses a microstrip antenna as an example. In any embodiment of this invention, the transmitting and receiving antennas can also be various forms such as external waveguide antennas, on-chip packaged antennas, and PCB antennas. The types of transmitting and receiving antennas in this invention can be the same or different; for example, the transmitting antenna can be a microstrip antenna, and the receiving antenna can be an external waveguide antenna. The antennas in any embodiment of this invention can be fabricated individually or as a whole.

[0199] In any embodiment of the present invention, the radar system includes at least two radar chips, and the number of radar chips may be three, four, five, etc.

[0200] This embodiment takes a radar system containing a first radar chip, a second radar chip, a third radar chip, and a fourth radar chip as an example. Each radar chip includes a transmitting antenna array and a receiving antenna array, and all radar chips are mounted on a PCB board.

[0201] like Figure 15s31 As shown, each radar chip is a 4T4R. The transmitting antenna arrays of the first and third radar chips are located on edge a of the PCB board, the transmitting antenna arrays of the second and fourth radar chips are located on edge c of the PCB board, the receiving antenna arrays of the first and third radar chips are located on edge b of the PCB board, and the receiving antenna arrays of the second and fourth radar chips are located on edge d of the PCB board.

[0202] like Figure 15s32 As shown, the radio frequency antenna transmit and receive interfaces on the radar chip are located in the second and fourth quadrants, respectively. This design of the transmit and receive interfaces can effectively reduce interference between channels.

[0203] In this embodiment, both the first radar chip and the second radar chip adopt... Figure 15s32 The antenna interface layout is shown; the antenna interface layout of the third radar chip is rotated 180° relative to the antenna interface layout of the second radar chip, and the antenna interface layout of the fourth radar chip is rotated 180° relative to the antenna interface layout of the first radar chip.

[0204] like Figure 15s33 As shown, each radar chip is a 4T4R. The transmitting antenna arrays of the first and second radar chips are located on edge b of the PCB board, the transmitting antenna arrays of the third and fourth radar chips are located on edge d of the PCB board, the receiving antenna arrays of the first and third radar chips are located on edge a of the PCB board, and the receiving antenna arrays of the second and fourth radar chips are located on edge c of the PCB board.

[0205] like Figure 15s34 As shown, the radio frequency antenna transmitting and receiving interfaces on the radar chip are located in the second and fourth quadrants, respectively, and the transmitting and receiving interfaces are alternately arranged in sequence.

[0206] The first radar chip uses Figure 15s34 The antenna interface layout is shown below; the antenna interface layout of the second radar chip is rotated 90° relative to the antenna interface layout of the first radar chip, the antenna interface layout of the third radar chip is rotated 180° relative to the antenna interface layout of the second radar chip, and the antenna interface layout of the fourth radar chip is rotated 180° relative to the antenna interface layout of the first radar chip.

[0207] In this embodiment, the radio frequency antenna transmitting and receiving interfaces of the first radar chip, the second radar chip, the third radar chip, and the fourth radar chip are connected to the transmitting antenna array and the receiving antenna array on the PCB via feed lines; in another embodiment, the radio frequency antenna transmitting and receiving interfaces of the first radar chip, the second radar chip, the third radar chip, and the fourth radar chip can also be waveguide interfaces, which are connected to an external waveguide antenna.

[0208] In this embodiment, the radar chips can be the same model or different models. In this embodiment, at least one transmitting antenna and at least one receiving antenna in the radar system are placed at the edge of the PCB board, thereby maximizing the aperture of the antenna array and improving the angular resolution of the radar.

[0209] according to Figure 15s31 The antenna layout shown forms a virtual array, such as Figure 15s35 As shown, the spacing coefficients di, hi (i=1-7) and D and H can be flexibly adjusted according to the algorithm and PCB size. In this embodiment, depending on the different requirements of the algorithm, the antenna array can be a uniform full array or a sparse array.

[0210] An example of the implementation of the antenna interface layout of the radar chip of the present invention. Figure 15s36 As shown. Figure 15s36 As shown in the left-middle figure, in this embodiment, the transmitting antenna interface and the receiving antenna interface are arranged perpendicularly to each other, effectively reducing signal crosstalk between the transmitting interface and the receiving port; as Figure 15s36 As shown in the middle right figure, the port arrangement in this example can ensure that the polarization between adjacent ports is perpendicular to each other, thereby reducing signal crosstalk at the ports.

[0211] Examples of radar chip layout configurations in the radar system of this invention Figure 15s37 As shown. Figure 15s37 As shown, in this embodiment, the first radar chip and the second radar chip can adopt the antenna interface layout of any embodiment of the present invention. The third radar chip is set in the same direction as the first radar chip or is set 180° relative to the first radar chip. The fourth radar chip is set in the same direction as the second radar chip or is set 45° relative to the second radar chip.

[0212] The radar system in any embodiment of the present invention may further include a fifth radar chip, which is a chip located in the middle. The fifth chip also includes a transmitting antenna array and a receiving antenna array. The fifth chip is placed near the center of the PCB to compensate for the problem of excessively high sidelobes caused by the sparse array when the aperture is too large.

[0213] In this embodiment of the invention, the shape of the PCB in the radar system is not limited to quadrilateral. In a polygonal, circular, or elliptical PCB, chips and corresponding antennas can be selectively arranged at each corner to maximize the antenna array aperture.

[0214] This application also provides an electronic device, such as... Figure 16 As shown, it includes: a carrier 1610, an integrated circuit 140, and an antenna 1620; The integrated circuit 1620 is disposed on the carrier 1610; The antenna 1620 is disposed on the carrier 1610 and is integrated with the integrated circuit 140 as a single device or disposed separately; it is connected to the integrated circuit 140 and is used to transmit radio frequency transmission signals and / or receive radio frequency reception signals.

[0215] In some exemplary embodiments, the antenna includes a MIMO antenna array as described in any of the above embodiments.

[0216] In some exemplary embodiments, the electronic device includes: a millimeter-wave radar receiver, a smart vehicle device, etc.

[0217] In some exemplary embodiments, the MIMO antenna array is configured to form a virtual antenna array as described in any embodiment of this disclosure.

[0218] It will be understood by those skilled in the art that all or some of the steps, systems, or apparatuses disclosed above, and their functional modules / units, can be implemented as software, firmware, hardware, or suitable combinations thereof. In hardware implementations, the division between functional modules / units mentioned above does not necessarily correspond to the division of physical components; for example, a physical component may have multiple functions, or a function or step may be performed collaboratively by several physical components. Some or all components may be implemented as software executed by a processor, such as a digital signal processor or microprocessor, or as hardware, or as an integrated circuit, such as an application-specific integrated circuit (ASIC). Such software may be distributed on a computer-readable medium, which may include computer storage media (or non-transitory media) and communication media (or transient media). As is known to those skilled in the art, the term computer storage media includes volatile and non-volatile, removable and non-removable media implemented in any method or technology for storing information (such as computer-readable instructions, data structures, program modules, or other data). Computer storage media include, but are not limited to, RAM, ROM, EEPROM, flash memory or other memory technologies, CD-ROM, digital versatile disc (DVD) or other optical disc storage, magnetic cartridges, magnetic tape, disk storage or other magnetic storage devices, or any other medium that can be used to store desired information and can be accessed by a computer. Furthermore, it is well known to those skilled in the art that communication media typically contain computer-readable instructions, data structures, program modules, or other data in modulated data signals such as carrier waves or other transmission mechanisms, and may include any information delivery medium.

Claims

1. A method of radar signal processing, characterized by, The method comprises: acquiring a range-Doppler spectrum of a radar signal; for a range gate, acquiring K Doppler cells with targets corresponding to the range gate in the range-Doppler spectrum; for a Doppler cell with targets, determining a corresponding spectrum estimation by using an iterative adaptive (IA) spectrum estimation method accelerated by fast Fourier transform (FFT); and selecting one of the spectrum estimations corresponding to the K Doppler cells with targets as an angle spectrum of the range gate according to a set condition; wherein the FFT is performed on the K Doppler cells with targets, and K is an integer greater than 0.

2. The radar signal processing method of claim 1, wherein, The method further comprises: for a range gate, if there is no Doppler cell with targets corresponding to the range gate in the range-Doppler spectrum, setting an angle spectrum of the range gate as a noise value.

3. The radar signal processing method of claim 1, wherein, The method of determining the corresponding spectrum estimation by using the IA spectrum estimation method accelerated by FFT comprises: performing FFT operation on signal data corresponding to the Doppler cell in the radar signal to obtain an initial value of an energy spectrum; performing the IA spectrum estimation accelerated by FFT according to the energy spectrum, and iteratively updating the energy spectrum until an iteration exit condition is met, to determine the spectrum estimation corresponding to the Doppler cell.

4. The radar signal processing method according to claim 3, wherein the method of performing the IA spectrum estimation accelerated by FFT according to the energy spectrum and iteratively updating the energy spectrum comprises: According to the energy spectrum The operation result of the FFT operation is the equation root r. performing a Levenberg-Dubbin (LD) recursive algorithm on the equation root r to optimize the covariance matrix R and obtain an optimized signal Y. According to the optimized signal Y, an FFT operation is performed to update the energy spectrum .

5. The radar signal processing method according to claim 4, wherein The energy spectrum is calculated according to the equation The operation result of the FFT operation is obtained, and an equation root r is obtained. to the energy spectrum performing a P-point FFT operation to obtain an initial equation root ; take the first Q data of the initial equation root as the equation root r; wherein P is the number of angle cells, and Q is the number of radar signal receiving channels.

6. The radar signal processing method according to claim 5, wherein inverse of the optimized covariance matrix R ; optimized signal ; wherein , is a strict lower triangular matrix of size P x P with all elements of the first sub-diagonal equal to 1 and all other elements equal to 0; ； ； ； ； ； X is the signal data corresponding to the Doppler unit in the radar signal, x is the signal data of each radar signal receiving channel in X; is the i-th element of the equation root is the i-th element of the equation root H is the vector conjugate transpose operation, * is the vector conjugate operation, LD() is the LD recursion operation.

7. The radar signal processing method according to claim 5, wherein The FFT operation is performed according to the optimized signal Y, and the energy spectrum is updated , comprising: performing FFT operation according to the optimized signal Y, and obtaining an angle spectrum estimation by using the following method: ； updating the energy spectrum ; wherein ； ； ； ； ； C = [ ]； ； ； ； Let i be the i-th element of vector c. Let be the conjugate of the i-th element of vector c. This is the reversal operation on vector c. For FFT operation; is the inversion operation for vector c, * is the conjugate operation for vectors, and T is the transpose operation for vectors; Let be the i-th element of vector t. It is the conjugate of the i-th element of vector t; Let be the i-th element of vector s. Let be the conjugate of the i-th element of vector s. For the P angle units, the first p indivual.

8. The radar signal processing method according to any one of claims 1-7, wherein the method of selecting one of the spectrum estimations corresponding to the K Doppler cells with targets as the angle spectrum of the range gate according to a set condition comprises: selecting a spectrum estimation corresponding to an energy maximum value of the spectrum estimations of the K Doppler cells with targets as the angle spectrum of the range gate.

9. The radar signal processing method according to any one of claims 1-7, wherein the radar signal comprises an MBC waveform radar signal, and each frame of the radar signal comprises a plurality of chirps; the method further comprises generating the range-Doppler spectrum after bandwidth synthesis according to the radar signal, comprising: acquiring sampling point data of the plurality of chirps, performing zero padding on sampling point data outside a bandwidth of each chirp to obtain zero-padded sampling point data; dividing the zero-padded sampling point data into a plurality of batches according to range gates, and performing range-dimension FFT on each batch to obtain a range gate spectrum corresponding to each batch. The phase compensation is performed on the spectrum of each range gate, and a velocity dimension FFT is performed to obtain a range Doppler spectrum of the plurality of chirps.

10. The radar signal processing method according to any one of claims 1-7, characterized in that, Applied to a frequency-modulated continuous wave FMCW radar with a multiple-input multiple-output MIMO antenna array, The MIMO antenna array includes at least one main subarray, and at least part of the main subarrays includes a uniform linear array ULA subarray. An angle spectrum of the range gate is obtained based on a radar signal corresponding to the ULA subarray.

11. The radar signal processing method of claim 10, wherein, The radar signal corresponding to the main subarray to which the ULA subarray belongs is configured to perform an unblurring operation.

12. The radar signal processing method of any one of claims 1-7, wherein The method further comprises: performing an angle resolving operation according to the angle spectrum of each range gate to obtain an angle resolving result, and performing an unblurring operation on the angle resolving result to determine a direction of arrival DoA.

13. An integrated circuit, comprising: a digital signal processing module configured to perform digital signal processing according to the method of any one of claims 1-12.

14. A virtual antenna array, comprising: comprising: at least one main subarray, each main subarray including at least four array elements distributed along a first direction in sequence; at least part of the main subarrays includes a uniform linear array ULA subarray.

15. The virtual antenna array of claim 14, wherein, For any main subarray including a ULA subarray, the ULA subarray is configured to perform an angle resolving operation in a radar signal processing in the first direction, and the main subarray is configured to perform an unblurring operation on the angle resolving result.

16. An antenna array, characterized by comprising: a plurality of receiving antennas and a plurality of transmitting antennas, the plurality of receiving antennas being arranged in sequence along a first direction to form a row of non-equidistantly arranged receiving antenna array, wherein part of the receiving antennas in the receiving antenna array are equidistantly distributed in sequence; the plurality of transmitting antennas are distributed in a plurality of rows, wherein part of the transmitting antennas are equidistantly distributed along the first direction, and the plurality of transmitting antennas are non-equidistantly arranged in the first direction when projected into a row.

17. An electronic device, comprising: comprising: a carrier, the integrated circuit of claim 13, and an antenna; wherein the integrated circuit is disposed on the carrier; the antenna is disposed on the carrier and integrated with the integrated circuit as an integrated device or disposed separately; and connected with the integrated circuit for transmitting radio frequency transmitting signals and / or receiving radio frequency receiving signals.

18. The electronic device of claim 17, wherein the antenna comprises a multiple-input multiple-output MIMO antenna array; the MIMO antenna array forms a virtual antenna array, and comprises at least one main subarray, and at least part of the main subarrays includes a uniform linear array ULA subarray.

19. The electronic device of claim 18, wherein the MIMO antenna array is the antenna array of claim 16, and / or the MIMO antenna array is configured to form the virtual antenna array of claim 14 or 15.

Images