4D millimeter wave radar antenna and nested array and anti-interference design method thereof

Abstract

The application discloses a 4D millimeter wave radar antenna and a nested array and anti-interference design method thereof. The method adopts a multi-transmitting and multi-receiving orthogonal array to construct a larger equivalent virtual aperture under a limited size, a double-layer waveguide is adopted for hardware, and a multi-resonance broadband matching is matched to cover 76-81 GHz; a multifunctional nested subarray is introduced to realize the division and cooperation of long-distance narrow beam, short-distance high resolution and pitch enhancement, and to suppress the sidelobe and reduce the array size through shared aperture and channel multiplexing; the spectral orthogonality and anti-interference ability with the same domain radar are improved through segmented calibration, online interference detection, subband selection and switching; under the premise of meeting the resolution, the module size is reduced through sparse topology optimization. The application can improve the stable detection distance and weak target detection ability, reduce the fusion and misassociation probability, reduce unnecessary braking, reduce the missed detection / false alarm caused by mutual interference, and reduce the constraint on the vehicle modeling and installation window.

Description

Technical Field

[0001] This invention relates to the field of 4D millimeter-wave radar technology, specifically to a 4D millimeter-wave radar antenna and its nested array and anti-interference design method. Background Technology

[0002] As intelligent driving technology evolves from assisted driving to advanced autonomous driving, vehicle-side perception systems typically rely on cameras as the primary sensor, supplemented by millimeter-wave radar and lidar. Millimeter-wave radar is characterized by its low illuminance sensitivity, strong penetration and stability in low-visibility environments such as rain, fog, and dust, and its ability to naturally provide radial velocity information. In typical urban and highway scenarios, millimeter-wave radar can also utilize electromagnetic wave diffraction to perceive partially obscured targets earlier, helping to detect potential hazards in advance. For example:

[0003] 1. "Ghost peek" type: Pedestrians / non-motorized vehicles suddenly cross the road from behind an obstruction;

[0004] 2. A scenario where the car in front is stationary and blocked by the car in front, and the scene only cuts out when the car in front gets very close.

[0005] As an important perception sensor for forward-facing scenarios, forward-facing millimeter-wave radar plays a key role in many advanced driver assistance systems (ADAS) functions, including but not limited to: adaptive cruise control (ACC), automatic emergency braking (AEB), forward collision warning (FCW), traffic jam assist (TJA) and highway pilot / integrated cruise assist (HWA / ICA), evasive steering assist (ESA), and target layer support and cut-in / cut-out detection for lane centering / lane keeping assist (LCC / LKA).

[0006] However, existing forward-facing millimeter-wave radars still face the following limitations and challenges in practical applications:

[0007] 1) Limited detection range: The stable detection range of typical mass-produced forward-facing radars is mostly within about 300m. Under conditions where braking distance increases significantly, such as on slippery roads, long downhill slopes, and high-speed driving, the warning redundancy is insufficient.

[0008] 2) Insufficient angular resolution: Due to the constraints of antenna aperture and array size, it is difficult to effectively distinguish close-range dual targets (such as traffic cones, disabled vehicles, etc.) in adjacent lanes at medium and long distances, which can easily lead to target fusion, false association, or missed detection.

[0009] 3) Insufficient accessibility assessment: Limited pitch resolution and height estimation capabilities make it easy to confuse accessible targets at high altitudes (such as road sign beams and the bottom of height restriction poles) with inaccessible obstacles on the ground, which may lead to misjudgment or unnecessary braking.

[0010] 4) Insufficient robustness in interference scenarios: With the increase in the installation rate of vehicle radar, interference between radars of the same frequency / asynchronous frequency occurs more frequently in the environment.

[0011] 5) Size and integration limitations: To achieve pitch resolution and virtual aperture expansion, existing 4D forward millimeter-wave radars generally increase the number of antenna elements and channels, resulting in array size, module thickness and heat dissipation volume that are significantly higher than ordinary forward radars, making it difficult to achieve friendly integration within the limited installation window at the front of the vehicle. Summary of the Invention

[0012] The purpose of this invention is to address the shortcomings of existing technologies by providing a 4D millimeter-wave radar antenna and its nested array and anti-interference design method.

[0013] To achieve the above objectives, this invention provides a 4D millimeter-wave radar antenna and its nested array and anti-interference design method, comprising:

[0014] Employing a multi-transmitter, multi-receiver orthogonal array to construct a larger equivalent virtual aperture within a limited size, thereby improving azimuth / elevation resolution, the ability to separate multiple targets at the same distance and speed, and the stability of the detection range;

[0015] The hardware used to achieve multiple transmit and receive orthogonal arrays adopts a double-layer waveguide structure and is equipped with multi-resonant broadband matching to cover 76–81 GHz, reduce insertion loss, and improve link gain and long-distance robustness.

[0016] A multi-functional nested subarray is introduced to achieve a balance between angular measurement accuracy and strong / weak target resolution.

[0017] By segmented calibration, online interference detection, sub-band selection and switching, the spectrum orthogonality and anti-interference capability with the radar in the same domain are improved;

[0018] While maintaining resolution, the module size is reduced through sparse topology optimization.

[0019] Furthermore, the hardware used to achieve multiple transmit / receive orthogonal arrays includes:

[0020] Multiple transmitting antennas are used to transmit electromagnetic waves into space;

[0021] Multiple receiving antennas are used to receive the echoes emitted by the transmitting antenna after they hit an object;

[0022] Multiple 0° PCB waveguide ports, one part of which is a transmitting channel used to receive electromagnetic radiation transmitted by the 4D millimeter-wave radar chip and further convert it into the waveguide transmission feed line; the other part is a receiving channel used to receive electromagnetic radiation transmitted by the waveguide transmission feed line and further convert it into the 4D millimeter-wave radar chip.

[0023] Multiple 90° PCB waveguide ports, one part of which is a transmitting channel, used to receive electromagnetic radiation transmitted by the 4D millimeter-wave radar chip, and complete a 90° turn before switching to the waveguide transmission feed line; the other part is a receiving channel, used to receive electromagnetic radiation transmitted by the waveguide transmission feed line, and complete a 90° turn before switching to the 4D millimeter-wave radar chip.

[0024] Multiple parasitic antennas are used to adjust the channel consistency between the real antennas.

[0025] Furthermore, both the transmitting and receiving antennas employ a double-layer cavity waveguide millimeter-wave antenna, specifically including:

[0026] The H-to-E waveguide transition structure uses rounded corners of the upper waveguide and extended length of the lower waveguide to convert the H-waveguide, which is convenient for transmission, into an E-waveguide.

[0027] A multi-waveguide power divider structure includes a first-stage power divider structure, several H-to-E conversion connection structures, and several second-stage power divider structures, used to distribute electromagnetic energy according to the required ratio. The first-stage power divider structure is a symmetrical E-waveguide to two H-waveguide structures, with a chamfered rectangular groove above and below it for adjusting impedance matching. The H-to-E conversion connection structures are located between the first and second-stage power divider structures, with a chamfered rectangular groove above and a chamfered triangular groove below for adjusting impedance matching. The second-stage power divider structure is an asymmetrical E-waveguide to two H-waveguide structures, adjusting the power distribution ratio by the degree of asymmetry. The H-waveguides of the second-stage power divider structure have two rectangular openings above them, from which two straight waveguides radiate energy into space.

[0028] Choke rings are placed on both sides of the straight waveguide for beam shaping.

[0029] Furthermore, the 0° PCB waveguide port includes:

[0030] The first gap waveguide metal pillar is set on the side where the first transition structure waveguide connects to the waveguide on the PCB, and is used to suppress electromagnetic energy leakage.

[0031] The first ridge waveguide is located on the side where the first transition structure waveguide connects to the waveguide on the PCB, in order to expand the operating bandwidth without changing the external dimensions.

[0032] The first-stage impedance adjustment structure is used to achieve first-stage impedance matching adjustment by setting a chamfered rectangular groove in the input / output transition section of the first transition structure waveguide.

[0033] The first and second stage impedance adjustment structures are set after the first stage impedance adjustment structure. The second stage impedance matching adjustment is achieved by setting a rectangular protrusion with chamfers.

[0034] A first-size gradient structure is used to reduce the port size. The first-size gradient structure is connected to one end of a first-transition waveguide, and the other end of the first-transition waveguide is connected to a first-secondary impedance adjustment structure.

[0035] Furthermore, the 90° PCB waveguide port includes:

[0036] The second gap waveguide metal pillar is set on the side where the second transition structure waveguide connects to the waveguide on the PCB, and is used to suppress electromagnetic energy leakage.

[0037] The second ridge waveguide is located on the side where the second transition structure waveguide connects to the waveguide on the PCB, in order to extend the operating bandwidth without changing the external dimensions.

[0038] The 90° bend structure with chamfered corners is used to achieve a 90° turn in the waveguide transmission direction, and the impedance matching at the bend is optimized by setting a chamfered triangular notch at the outer bend.

[0039] The second-size gradient structure is used to reduce the port size. The second-size gradient structure is connected to one end of the second transition structure waveguide, and the other end of the second transition structure waveguide is connected to a 90° bend structure with a chamfer.

[0040] Furthermore, the multifunctional nested subarray includes a physical array and a virtual array;

[0041] The minimum horizontal spacing between the transmitting and receiving antennas of the physical array is ≥1.5 times the wavelength to ensure sufficient physical safety distance between the antennas. When the transmitting and receiving antennas are arranged adjacent to each other in the horizontal direction, their pitch direction is staggered by a distance ≥1.5 times the wavelength to ensure that the transmit / receive isolation is better than -35dB.

[0042] The virtual array includes:

[0043] Subarray 1: Located in the longest row of the virtual array, it uses all the virtual array elements in this row to form a whole column, which is used for the angular dimension accumulation of the distributed average constant false alarm rate, and outputs the strongest target in the corresponding distance and velocity two-dimensional units, and assists Subarray 2 in deblurring;

[0044] Subarray 2: Located in the longest row of the virtual array, a uniform array with a spacing of 7 times the wavelength, used to distinguish strong and weak targets at the same distance and speed but different angles;

[0045] Subarray 3: Located in the longest row of the virtual array, a uniform array with a spacing of 1 wavelength, used to distinguish strong and weak targets at the same distance and speed but with an angle difference of more than a preset threshold, and to assist subarray 2 in deblurring;

[0046] Subarray 4: This is an elevation azimuth array. Its angle estimation includes both azimuth and elevation information. It is decoupled from the azimuth results obtained from subarray 1, subarray 2, and subarray 3 to extract elevation angle information.

[0047] Furthermore, the segmentation calibration includes:

[0048] a) Divide the subbands into: Subband 1: 76–77 GHz; Subband 2: 77–78 GHz; Subband 3: 78–79 GHz; Subband 4: 79–80 GHz; Subband 5: 80–81 GHz;

[0049] b) Perform individual transmit-receive calibration for each 1GHz subband on the production line, as detailed below:

[0050] 1) Transmit 5 frames of calibration waveforms sequentially according to sub-band 1-sub-band 5;

[0051] 2) Measure and record the phase compensation factor at each sub-band, each transmit-receive channel, and each azimuth / elevation angle to form a phase calibration table, with each entry representing the phase using 1 byte;

[0052] c) Runtime use: Based on the calibration table of the corresponding sub-band, perform amplitude and phase compensation for each transmit-receive channel of the sub-band according to the angle.

[0053] Furthermore, the online interference detection includes:

[0054] a) Between different subbands, only the chirp initiation frequency is changed, while the remaining parameters remain consistent across all subbands;

[0055] b) If the energy of certain range-velocity two-dimensional units or the full-band energy is significantly higher than that of the thermal noise and environmental echo statistical model in multiple frames, or if a target appears that is inconsistent with the local waveform parameters, then the frame or time window is recorded as a frequency domain interference event.

[0056] c) Detect narrow pulse / spiking interference in the timing of the analog-to-digital converter, count and integrate it. If the count exceeds the threshold within a given time window, it is determined to be a time-domain interference event.

[0057] d) If a large number of tracked targets are lost or change significantly, and this is clearly inconsistent with the vehicle's operating conditions, it is recorded as a tracking layer interference event.

[0058] Furthermore, the subband selection and switching includes:

[0059] a) If the overall interference score of the current working sub-band is lower than the first score threshold, then the current sub-band is considered to be in a strong interference environment:

[0060] 1) After the subsequent normal FM continuous wave function waveform, insert a listening sequence: transmit the chirps of subband 1 to subband 5 in sequence for several frames;

[0061] 2) For each operating frequency band, online interference assessment is performed based solely on range Fourier transform for domain energy anomaly detection and time-domain impulse interference counting;

[0062] 3) Obtain the interference scores of the 5 subbands, sort them, and select the subband with the least interference as the candidate target subband;

[0063] b) Switching logic with hysteresis:

[0064] 1) If the overall score of the current sub-band deteriorates but is still higher than the second score threshold, then continue to operate the current sub-band without switching.

[0065] 2) At the same time, a minimum dwell time is introduced, prohibiting further switching within a specified time after the most recent subband switch;

[0066] 3) If the current sub-band comprehensive score is lower than the second score threshold and the minimum dwell time requirement is met, then the full sub-band scoring and sorting process will be retried.

[0067] Furthermore, the horizontal aperture of the physical array is ≤33 times half a wavelength, and its elevation aperture is ≤15 times half a wavelength.

[0068] Beneficial effects: 1. For long-range detection: By increasing the link gain through channel size and low-loss radiation / feeding, the stable detection range and the ability to detect weak targets are improved.

[0069] 2. For angle resolution and multi-target separation: A larger equivalent virtual aperture and low sidelobe layout significantly improve azimuth / elevation accuracy and the ability to separate multiple targets at the same distance and speed, while reducing the probability of fusion and false association.

[0070] 3. Regarding accessibility assessment: Improved pitch array aperture and imaging capabilities help distinguish between accessible targets at high altitudes and inaccessible targets on the ground, reducing unnecessary braking.

[0071] 4. Regarding anti-interference: Frequency agility and coding resource scheduling improve orthogonality with external radar, reducing missed detections / false alarms caused by mutual interference.

[0072] 5. Regarding miniaturization and integration: While meeting the requirements of full-band coverage of 76–81GHz and 4D imaging capabilities, the module size is effectively controlled, reducing constraints on the overall vehicle styling and installation window. Attached Figure Description

[0073] Figure 1 This is a schematic diagram of the double-layer waveguide structure according to an embodiment of the present invention;

[0074] Figure 2 This is a schematic diagram of the layout of the 0° PCB waveguide port and the 90° PCB waveguide port according to an embodiment of the present invention.

[0075] Figure 3 This is a schematic diagram of the layout of the transmitting and receiving antennas according to an embodiment of the present invention.

[0076] Figure 4 This is a schematic diagram of a virtual array according to an embodiment of the present invention;

[0077] Figure 5 This is a schematic diagram of the structure of the double-layer cavity waveguide millimeter-wave antenna according to an embodiment of the present invention;

[0078] Figure 6 This is a schematic diagram of the 0° PCB waveguide port structure according to an embodiment of the present invention;

[0079] Figure 7 This is a schematic diagram of the structure of a 90° PCB waveguide port according to an embodiment of the present invention;

[0080] Figure 8 This is the azimuth radiation pattern of a 4D millimeter-wave radar antenna in the 75-81GHz range;

[0081] Figure 9 This is the elevation pattern of the 4D millimeter-wave radar antenna in the 75-81GHz range;

[0082] Figure 10 This is a simulation diagram of the S11 parameters of a 4D millimeter-wave radar antenna at 75-81GHz;

[0083] Figure 11 This is a diagram showing the overall return loss of transmitting antenna 101;

[0084] Figure 12 This is a diagram showing the overall return loss of receiving antenna 102;

[0085] Figure 13 This is the S-parameter diagram of a 0° PCB waveguide port;

[0086] Figure 14 This is the S-parameter diagram of a 90° PCB waveguide port;

[0087] Figure 15 This is a schematic diagram illustrating the process of segment calibration, online interference detection, and subband selection and switching in an embodiment of the present invention. Detailed Implementation

[0088] The present invention will be further illustrated below with reference to the accompanying drawings and specific embodiments. These embodiments are implemented based on the technical solutions of the present invention, and it should be understood that these embodiments are only used to illustrate the present invention and are not intended to limit the scope of the present invention.

[0089] This invention provides a 4D millimeter-wave radar antenna and its nested array and anti-jamming design method, including:

[0090] Employing a multiple-input multiple-output (MIMO) orthogonal array allows for the construction of a larger equivalent virtual aperture within a limited size, thereby improving azimuth / elevation resolution, the ability to separate multiple targets at the same distance and speed, and the stability of the detection range.

[0091] For details, see Figure 1 The hardware used to realize the orthogonal array of multiple transmitters and receivers includes multiple transmitting antennas 101, multiple receiving antennas 102, multiple 0° PCB waveguide ports 2 (metal waveguide port transition structure), multiple 90° PCB waveguide ports 3 (metal waveguide port transition structure), multiple waveguide transmission feed lines 4, and multiple parasitic antennas 5.

[0092] Multiple transmitting antennas 101 are used to transmit electromagnetic waves into space. Preferably, there are eight transmitting antennas 101, but they are not limited to eight.

[0093] Multiple receiving antennas 102 are used to receive the echoes emitted by the transmitting antenna 101 after hitting an object. Preferably, there are eight receiving antennas 102, but they are not limited to eight.

[0094] A portion of the multiple 0° PCB waveguide ports 2 belong to the transmitting channel, used to receive electromagnetic radiation transmitted by the 4D millimeter-wave radar chip and further convert it into the waveguide transmission feed line 4. Another portion of the multiple 0° PCB waveguide ports 2 belong to the receiving channel, used to receive electromagnetic radiation transmitted by the waveguide transmission feed line 4 and further convert it into the 4D millimeter-wave radar chip. Preferably, there are eight 0° PCB waveguide ports 2, but it is not limited to eight.

[0095] A portion of the multiple 90° PCB waveguide ports 3 belongs to the transmitting channel, used to receive electromagnetic radiation transmitted from the 4D millimeter-wave radar chip, complete a 90° turn, and then transfer it into the waveguide transmission feed line 4; another portion of the multiple 90° PCB waveguide ports 3 belongs to the receiving channel, used to receive electromagnetic radiation transmitted from the waveguide transmission feed line 4, complete a 90° turn, and then transfer it into the 4D millimeter-wave radar chip. Preferably, there are eight 90° PCB waveguide ports 3, but it is not limited to eight. See also Figure 2 , Figure 2 The diagram illustrates the layout of multiple 0° PCB waveguide ports 2 and multiple 90° PCB waveguide ports 3.

[0096] In an embodiment where there are eight 0° PCB waveguide ports 2 and eight 90° PCB waveguide ports 3, there are 16 waveguide transmission feed lines 4, each corresponding to one of the 0° PCB waveguide ports 2 and the 90° PCB waveguide ports 3. A portion of the waveguide transmission feed lines 4 belongs to the transmitting channel, used to transfer the electromagnetic radiation output from the 0° PCB waveguide ports 2 and 90° PCB waveguide ports 3 to the transmitting antenna 101; another portion of the waveguide transmission feed lines 4 belongs to the receiving channel, used to transfer the electromagnetic radiation in the receiving antenna 102 to the 0° PCB waveguide ports 2 and 90° PCB waveguide ports 3.

[0097] Multiple parasitic antennas 5 are used to adjust the channel consistency between the real antennas (transmitting antenna 101 and receiving antenna 102). Preferably, there are seven parasitic antennas, but they are not limited to seven. Figure 1 and Figure 3 As can be seen from the layout of the transmitting antenna 101, the receiving antenna 102, and the parasitic antenna 5, Figure 3 In the diagram, T1-T8 represent the positions of the eight transmitting antennas 101, and R1-R8 represent the positions of the eight receiving antennas. The units in the x and y directions are half wavelengths.

[0098] The hardware used to achieve multiple transmit and receive orthogonal arrays employs a double-layer waveguide structure and is equipped with multi-resonant broadband matching to cover 76–81 GHz, while reducing insertion loss and improving link gain and long-distance robustness.

[0099] See Figure 5 The aforementioned transmitting antenna 101 and receiving antenna 102 both adopt a double-layer cavity waveguide millimeter-wave antenna, specifically including an H-to-E waveguide transition structure 11, a one-to-many waveguide power divider structure 12, and a choke ring 13.

[0100] The H-to-E waveguide transition structure 11 uses a rounded corner cut in the upper waveguide and an extended length in the lower waveguide to convert the H-waveguide, which is convenient for transmission, into an E-waveguide. Because double-layer splicing requires segmenting the waveguide, and the H-waveguide, when segmented in the middle, experiences minimal cutting of the electric field and the lowest energy leakage, it is suitable for transmission. Since the two paths split by the E-waveguide as a power divider are 180° out of phase, when the transmission directions are opposite, the 180° phase difference can actually be superimposed in phase. Specifically, this H-waveguide has a width of 2.2mm and a thickness of 1mm. The thickness of this E-waveguide is a portion of the width of the H-waveguide, i.e., 1.1mm, and its width is 2.7mm.

[0101] The 1-to-multi-waveguide power divider structure 12 includes a first-stage power divider structure 121, several H-to-E conversion connection structures 122, and several second-stage power divider structures 123, used to distribute electromagnetic energy according to the required proportion. As a preferred embodiment, the 1-to-multi-waveguide power divider structure 12 is a 1-to-4 waveguide power divider structure; specifically, it consists of two 1-to-2 waveguide power divider cascades (specifically, the first-stage power divider structure). 1+H to E connection structure 2+ Second-stage power divider structure 2) Distribute the electromagnetic energy into 4 channels according to the required ratio. Although the given case is 4 channels, it is not limited to 4 channels. As the pitch FOV (field of view) requirement decreases and the ranging requirement increases, the number of channels needs to be increased, such as 6 channels and 8 channels.

[0102] The first-stage power divider structure 121 is a symmetrical E-waveguide to two H-waveguide structures. Above and below the first-stage power divider structure are chamfered rectangular grooves 124 for impedance matching adjustment. An H-to-E connection structure 122 is positioned between the first-stage power divider structure 121 and the second-stage power divider structure 123. Above the H-to-E connection structure 122 is a chamfered rectangular groove, and below it is a chamfered triangular groove for impedance matching adjustment. The second-stage power divider structure 123 is an asymmetrical E-waveguide to two H-waveguide structures. The power distribution ratio is adjusted by the degree of asymmetry. Above the H-waveguide of the second-stage power divider structure 123 are two rectangular openings, each 0.8 mm wide and 2.2 mm long (half a wavelength). Two straight waveguides radiate from these openings for energy radiation into space. Choke rings 13 are positioned on both sides of the straight waveguides for beamforming. Specifically, the choke ring 13 is positioned 0.85mm apart on both sides of the straight waveguide, with a width of 0.8mm and a length equal to that of the antenna body. The overall width of the antenna is 6.25mm, supporting placement at 1.5 times the wavelength interval. h) Traditional full-vehicle-band waveguide antenna solutions have an S-parameter depth of only -10dB or -15dB in the 76-81GHz range. Due to the inevitable left / right and upward shifts in S-parameters during manufacturing, the robustness of these designs is clearly insufficient. However, using the solution of this application, the S-parameter depth in the 76-81GHz range is below -30dB (the reflected signal energy of -30dB is only 1% of that of -10dB), which can improve the robustness of the manufactured parts with sufficient design redundancy.

[0103] See Figure 6 The aforementioned 0° PCB waveguide port includes a first gap waveguide metal pillar 21, a first ridge waveguide 22, a first primary impedance adjustment structure 23, a first secondary impedance adjustment structure 24, a first transition structure 25, and a first size gradient structure 26.

[0104] The first gap waveguide metal pillar 21 is disposed on the side where the first transition structure 25 connects to the waveguide on the PCB, and is used to suppress electromagnetic energy leakage. Specifically, there are preferably eight first gap waveguide metal pillars 21, which are arranged around the first ridge waveguide 22. Even if the transition structure (PCB waveguide port to rectangular waveguide port) is not completely attached to the PCB, it can still ensure effective energy transmission and almost no leakage.

[0105] The first ridge waveguide 22 is located on the side where the first transition structure waveguide 25 connects to the waveguide on the PCB, in order to expand the operating bandwidth without changing the external dimensions.

[0106] The first-stage impedance adjustment structure 23 is used to achieve first-stage impedance matching adjustment by setting a chamfered rectangular groove in the input / output transition section of the first transition structure waveguide 25.

[0107] The first and second stage impedance adjustment structure 24 is set after the first stage impedance adjustment structure 23. The second stage impedance matching adjustment is achieved by setting a rectangular protrusion with chamfers.

[0108] The first size gradient structure 26 is used to reduce the port size, preferably to 2.2mm × 1mm. The first size gradient structure 26 is connected to one end of the first transition structure waveguide 25, and the other end of the first transition structure waveguide 25 is connected to the first secondary impedance adjustment structure 24. Under the combined effect of the above structures, the S-parameters of this transition structure in the 71.6–81.4GHz frequency band are better than -25dB, providing sufficient design margin and significantly improving the robustness of the actual manufactured parts.

[0109] See Figure 7 The aforementioned 90° PCB waveguide port includes a second gap waveguide metal pillar 31, a second ridge waveguide 32, a 90° bend structure with chamfered corners 33, a second transition structure waveguide 34, and a second size gradient structure 35.

[0110] The second gap waveguide metal pillar 31 is disposed on the side where the second transition structure waveguide 34 connects to the waveguide on the PCB, and is used to suppress electromagnetic energy leakage. Similarly, there are preferably eight second gap waveguide metal pillars 31, which are arranged around the second ridge waveguide 32, so that even if the transition structure is not completely attached to the PCB, energy can be effectively transmitted with almost no leakage.

[0111] The second ridge waveguide 32 is located on the side where the second transition structure waveguide 34 connects to the waveguide on the PCB, in order to expand the operating bandwidth without changing the external dimensions.

[0112] The 90° bend structure 33 with chamfered corners is used to achieve a 90° turn in the waveguide transmission direction, and the impedance matching at the bend is optimized by setting a chamfered triangular notch at the outer bend.

[0113] The second-size gradient structure 35 is used to reduce the port size, which is 2.2mm × 1mm. The second-size gradient structure 35 is connected to one end of the second transition structure waveguide 34, and the other end of the second transition structure waveguide 34 is connected to the 90° bend structure 33 with a chamfer. Under the combined effect of the above structures, the S-parameter of this transition structure is better than -29dB in the 74.9–81GHz frequency band, with sufficient design margin, which can significantly improve the robustness of the actual manufactured parts.

[0114] A multifunctional nested subarray is introduced to achieve a balance between angular measurement accuracy and strong / weak target resolution. Specifically, the multifunctional nested subarray includes both physical and virtual arrays.

[0115] Among them, the minimum horizontal spacing between the transmitting and receiving antennas of the physical array is ≥1.5 times the wavelength to ensure sufficient physical safety distance between the antennas. When the transmitting and receiving antennas are arranged adjacent to each other in the horizontal direction, their pitch direction is staggered by a distance of ≥1.5 times the wavelength to ensure that the transmission and reception isolation is better than -35dB.

[0116] See Figure 4 The aforementioned virtual array includes:

[0117] Subarray 1: Located in the longest row of the virtual array, using all virtual array elements in that row to form a complete column. 3dB beamwidth 1.8264° (after DML angular resolution <1°), 6×6=36 array elements, aperture 63 times half wavelength, sidelobes -12.5470dB. Primarily used for angular dimension accumulation of distributed average constant false alarm rate (DA-CFAR), outputting the strongest target at the corresponding range and velocity bin, while also assisting subarray 2 in deblurring; due to its low sidelobes, it can output non-extremely weak targets to a certain extent.

[0118] Subarray 2: Located in the longest row of the virtual array, this is a uniform array with a spacing of 7 times the wavelength. It has a 3dB beamwidth of 1.6205° (angular resolution <1° after DML), 10 array elements, a blur factor of 7, and an aperture of 7 × (10⁻¹) = 63 times half a wavelength. It is used to distinguish targets at the same distance and speed but with different angles (angles that are very close or significantly different are acceptable; when the angle difference is large, it needs to be deblurred in conjunction with subarrays 1 and 3). Because it is a uniform array, it has a strong ability to distinguish between strong and weak targets.

[0119] Subarray 3: Located in the longest row of the virtual array, this is a uniform array with an interval of one wavelength. It has 11 array elements, an aperture of 10 times half a wavelength, and no angular ambiguity. It is used to distinguish between strong and weak targets at the same distance and speed but with an angular difference greater than a preset threshold (5.08°), and assists subarray 2 in de-ambiguation; also, due to the characteristics of a uniform array, it has a strong ability to distinguish between strong and weak targets.

[0120] Subarray 4: Elevation azimuth array. Its angle estimation includes both azimuth and elevation information. It needs to be decoupled from the azimuth angle results obtained from subarrays 1-3 to extract the elevation angle information. The decoupled elevation 3dB beamwidth is 3.1815° (after DML, angular resolution <1.7°), and the sidelobes are -8.9625dB.

[0121] By segmented calibration (executed offline on the production line), online interference detection (continuously executed while the vehicle is running), and sub-band selection and switching (triggered in strong interference scenarios), the spectrum orthogonality and anti-interference capability with the radar in the same domain are improved.

[0122] See Figure 15 Specifically, the segmented calibration includes:

[0123] a) Divide the subbands into: Subband 1: 76–77 GHz; Subband 2: 77–78 GHz; Subband 3: 78–79 GHz; Subband 4: 79–80 GHz; Subband 5: 80–81 GHz.

[0124] b) Perform individual transmit-receive calibration for each 1GHz subband on the production line, as detailed below:

[0125] 1) Transmit 5 frames of calibration waveforms sequentially from subband 1 to subband 5.

[0126] 2) Measure and record the phase compensation factor at each sub-band, each transmit-receive channel, and each azimuth / elevation angle. (tx, rx, angle), forming a phase calibration table, where each entry uses 1 byte to represent the phase. The phase calibration table is preferably 5 bytes. 8 8 (141+34), where 5 is the number of sub-bands, 8 is the number of transmit antennas, 8 is the number of receive antennas, 141 is the number of azimuth calibration angles from -70° to 70°, and 34 is the number of elevation calibration angles from -17° to +17° excluding 0°. The total size of the phase calibration table is approximately 54.7KB, and the maximum quantization error is approximately ±0.70°.

[0127] c) Runtime use: Regardless of which 1GHz subband the radar switches to, amplitude and phase compensation is performed on each TX-RX channel of the corresponding subband according to the calibration table, so as to ensure that the equivalent array response under different subbands is as consistent as possible.

[0128] The above-mentioned online interference detection includes:

[0129] a) Between different subbands, only the chirp start frequency is changed, while the other parameters remain consistent across all subbands (sweep bandwidth, sweep slope, chirp duration, sampling rate, number of sampling points, distance dimension / Doppler dimension fast Fourier transform size, etc.).

[0130] b) If the energy of certain range-velocity two-dimensional units or the full-band energy is significantly higher than that of the thermal noise and environmental echo statistical model within multiple frames, or if a target appears that is inconsistent with the local waveform parameters, then the frame or time window is recorded as a frequency domain interference event.

[0131] c) Detect narrow pulse / spiking interference in the timing of the analog-to-digital converter, count and integrate it. If the count exceeds the threshold within a given time window, it is determined to be a time-domain interference event.

[0132] d) If a large number of tracked targets are lost or change significantly, and this is clearly inconsistent with the vehicle's operating conditions, it is recorded as a tracking layer interference event.

[0133] By combining the above three types of events, the interference score of the current subband within the window is obtained, which is used to drive the subsequent subband selection logic.

[0134] The above sub-band selection and switching includes:

[0135] a) If the overall interference score of the current working sub-band is lower than the first score threshold, then the current sub-band is considered to be in a strong interference environment:

[0136] 1) After the normal FM continuous wave function waveform, insert a listening sequence: transmit the chirps of subband 1 to subband 5 in sequence for several frames (designed according to delay / overhead).

[0137] 2) For each working frequency band, online interference assessment is performed based solely on the distance Fourier transform for domain energy anomaly detection and time-domain impulse interference counting (at this time, the Tracking layer statistics are not relied upon).

[0138] 3) Obtain the interference scores of the 5 subbands, sort them, and select the subband with the least interference (cleanest) as the candidate target subband.

[0139] b) Switching logic with hysteresis:

[0140] 1) If the overall score of the current sub-band deteriorates but is still higher than the second score threshold, the current sub-band will continue to operate without switching, in order to avoid frequent sub-band jumps.

[0141] 2) At the same time, a minimum dwell time is introduced, prohibiting switching again within a specified time after the most recent subband switch, in order to avoid frequent subband jumps.

[0142] 3) If the current sub-band comprehensive score is lower than the second score threshold and the minimum dwell time requirement is met, then the full sub-band scoring and sorting process will be retried.

[0143] While maintaining resolution, the module size is reduced through sparse topology optimization. Specifically:

[0144] a) The horizontal aperture of the physical array is ≤33 times half the wavelength to ensure that the width of the waveguide antenna is 73.5mm and the width of the whole unit is 80mm (excluding the latch), which is less than the width of the whole unit of a common 4D radar with the same resolution of about 100mm.

[0145] b) The elevation aperture of the physical array is ≤15 times half wavelength to ensure that the waveguide antenna height is 54mm and the overall height is 60.5mm (excluding the latch), which is less than the overall height of about 80mm for common 4D radars with the same resolution.

[0146] The performance of the 4D millimeter-wave radar antenna designed using the above methods is as follows: Figures 8 to 12 As shown, where, Figure 8 The diagram illustrates the azimuth radiation pattern of the 4D millimeter-wave radar antenna in the 75-81 GHz range, demonstrating that the radiation pattern bandwidth of the 4D millimeter-wave radar antenna can cover the 75 to 81 GHz range and has good consistency. Figure 9 The diagram illustrates the elevation pattern of the 4D millimeter-wave radar antenna in the 75-81 GHz range, demonstrating the stability and consistency of the antenna's pattern characteristics within this frequency band. Figure 10 The S11 parameters of the 4D millimeter-wave radar antenna in the 75-81 GHz range are illustrated. This design scheme achieves an S11 below -30 dB, exhibiting significant design redundancy and contributing to improved robustness of the fabricated parts. Figure 11 The overall return loss diagram of the transmitting antenna 101 is shown, and the result is -20 dB. In the survey with peers, the threshold is usually -10 dB and -15 dB. Therefore, the performance and robustness of the 4D millimeter-wave radar antenna designed in this application are superior to the industry standard. Figure 12 The diagram illustrates the overall return loss of the receiving antenna 102, which is similar to that of the transmitting antenna 101, reflecting its good performance. Figure 13 The S-parameters of 0° PCB waveguide port 2 are shown, and the result reaches -25 dB across the entire frequency band, a level that meets the industry's common threshold. Figure 14 The S-parameters of the 90° PCB waveguide port are shown, achieving -29 dB across the entire frequency band, which is better than the industry-standard threshold of -25 dB, demonstrating superior performance.

[0147] The above description is merely a preferred embodiment of the present invention. It should be noted that for those skilled in the art, other parts not specifically described are existing technology or common knowledge. Several improvements and modifications can be made without departing from the principle of the present invention, and these improvements and modifications should also be considered within the scope of protection of the present invention.

Claims

1. A 4D millimeter wave radar antenna and its nested array and anti-jamming design method, characterized in that, include: Employing a multi-transmitter, multi-receiver orthogonal array to construct a larger equivalent virtual aperture within a limited size, thereby improving azimuth / elevation resolution, the ability to separate multiple targets at the same distance and speed, and the stability of the detection range; The hardware used to achieve multiple transmit and receive orthogonal arrays adopts a double-layer waveguide structure and is equipped with multi-resonant broadband matching to cover 76–81 GHz, reduce insertion loss, and improve link gain and long-distance robustness. A multi-functional nested subarray is introduced to achieve a balance between angular measurement accuracy and strong / weak target resolution. By segmented calibration, online interference detection, sub-band selection and switching, the spectrum orthogonality and anti-interference capability with the radar in the same domain are improved; While maintaining resolution, the module size is reduced through sparse topology optimization. The hardware used to implement a multi-transmit, multi-receive orthogonal array includes: Multiple transmitting antennas are used to transmit electromagnetic waves into space; Multiple receiving antennas are used to receive the echoes emitted by the transmitting antenna after they hit an object; Multiple 0° PCB waveguide ports, one part of which is a transmitting channel used to receive electromagnetic radiation transmitted by the 4D millimeter-wave radar chip and further convert it into the waveguide transmission feed line; the other part is a receiving channel used to receive electromagnetic radiation transmitted by the waveguide transmission feed line and further convert it into the 4D millimeter-wave radar chip. Multiple 90° PCB waveguide ports, one part of which is a transmitting channel, used to receive electromagnetic radiation transmitted by the 4D millimeter-wave radar chip, and complete a 90° turn before switching to the waveguide transmission feed line; the other part is a receiving channel, used to receive electromagnetic radiation transmitted by the waveguide transmission feed line, and complete a 90° turn before switching to the 4D millimeter-wave radar chip. Multiple parasitic antennas are used to adjust the channel consistency between real antennas; The multifunctional nested subarray includes a physical array and a virtual array; The minimum horizontal spacing between the transmitting and receiving antennas of the physical array is ≥1.5 times the wavelength to ensure sufficient physical safety distance between the antennas. When the transmitting and receiving antennas are arranged adjacent to each other in the horizontal direction, their pitch direction is staggered by a distance ≥1.5 times the wavelength to ensure that the transmit / receive isolation is better than -35dB. The virtual array includes: Subarray 1: Located in the longest row of the virtual array, it uses all the virtual array elements in this row to form a whole column, which is used for the angular dimension accumulation of the distributed average constant false alarm rate, and outputs the strongest target in the corresponding distance and velocity two-dimensional units, and assists Subarray 2 in deblurring; Subarray 2: Located in the longest row of the virtual array, a uniform array with a spacing of 7 times the wavelength, used to distinguish strong and weak targets at the same distance and speed but different angles; Subarray 3: Located in the longest row of the virtual array, a uniform array with a spacing of 1 wavelength, used to distinguish strong and weak targets at the same distance and speed but with an angle difference of more than a preset threshold, and to assist subarray 2 in deblurring; Subarray 4: This is an elevation azimuth array. Its angle estimation includes both azimuth and elevation information. It is decoupled from the azimuth results obtained from subarray 1, subarray 2, and subarray 3 to extract elevation angle information.

2. The 4D mmWave radar antenna and its nested array and anti-jamming design method according to claim 1, characterized in that, Both the transmitting and receiving antennas are double-layer cavity waveguide millimeter-wave antennas, specifically including: The H-to-E waveguide transition structure uses rounded corners of the upper waveguide and extended length of the lower waveguide to convert the H-waveguide, which is convenient for transmission, into an E-waveguide. A multi-waveguide power divider structure includes a first-stage power divider structure, several H-to-E conversion connection structures, and several second-stage power divider structures, used to distribute electromagnetic energy according to the required ratio. The first-stage power divider structure is a symmetrical E-waveguide to two H-waveguide structures, with a chamfered rectangular groove above and below it for adjusting impedance matching. The H-to-E conversion connection structures are located between the first and second-stage power divider structures, with a chamfered rectangular groove above and a chamfered triangular groove below for adjusting impedance matching. The second-stage power divider structure is an asymmetrical E-waveguide to two H-waveguide structures, adjusting the power distribution ratio by the degree of asymmetry. The H-waveguides of the second-stage power divider structure have two rectangular openings above them, from which two straight waveguides radiate energy into space. Choke rings are placed on both sides of the straight waveguide for beam shaping.

3. The 4D mmWave radar antenna and its nested array and anti-jamming design method according to claim 1, characterized in that, The 0° PCB waveguide port includes: The first gap waveguide metal pillar is set on the side where the first transition structure waveguide connects to the waveguide on the PCB, and is used to suppress electromagnetic energy leakage. The first ridge waveguide is located on the side where the first transition structure waveguide connects to the waveguide on the PCB, in order to expand the operating bandwidth without changing the external dimensions. The first-stage impedance adjustment structure is used to achieve first-stage impedance matching adjustment by setting a chamfered rectangular groove in the input / output transition section of the first transition structure waveguide. The first and second stage impedance adjustment structures are set after the first stage impedance adjustment structure. The second stage impedance matching adjustment is achieved by setting a rectangular protrusion with chamfers. A first-size gradient structure is used to reduce the port size. The first-size gradient structure is connected to one end of a first-transition waveguide, and the other end of the first-transition waveguide is connected to a first-secondary impedance adjustment structure.

4. The 4D mmWave radar antenna and its nested array and anti-jamming design method according to claim 1, characterized in that, The 90° PCB waveguide port includes: The second gap waveguide metal pillar is set on the side where the second transition structure waveguide connects to the waveguide on the PCB, and is used to suppress electromagnetic energy leakage. The second ridge waveguide is located on the side where the second transition structure waveguide connects to the waveguide on the PCB, in order to extend the operating bandwidth without changing the external dimensions. The 90° bend structure with chamfered corners is used to achieve a 90° turn in the waveguide transmission direction, and the impedance matching at the bend is optimized by setting a chamfered triangular notch at the outer bend. The second-size gradient structure is used to reduce the port size. The second-size gradient structure is connected to one end of the second transition structure waveguide, and the other end of the second transition structure waveguide is connected to a 90° bend structure with a chamfer.

5. The 4D mmWave radar antenna and its nested array and anti-jamming design method according to claim 1, characterized in that, The segmentation calibration includes: a) Divide the subbands into: Subband 1: 76–77 GHz; Subband 2: 77–78 GHz; Subband 3: 78–79 GHz; Subband 4: 79–80 GHz; Subband 5: 80–81 GHz; b) Perform individual transmit-receive calibration for each 1GHz subband on the production line, as detailed below: 1) Transmit 5 frames of calibration waveforms sequentially according to sub-band 1-sub-band 5; 2) Measure and record the phase compensation factor at each sub-band, each transmit-receive channel, and each azimuth / elevation angle to form a phase calibration table, with each entry representing the phase using 1 byte; c) Runtime use: Based on the calibration table of the corresponding sub-band, perform amplitude and phase compensation for each transmit-receive channel of the sub-band according to the angle.

6. The 4D mmWave radar antenna and its nested array and anti-jamming design method according to claim 5, characterized in that, The online interference detection includes: a) Between different subbands, only the chirp initiation frequency is changed, while the remaining parameters remain consistent across all subbands; b) If the energy of certain range-velocity two-dimensional units or the full-band energy is higher than the thermal noise and environmental echo statistical model within multiple frames, or if a target appears that is inconsistent with the local waveform parameters, then the frame or time window is recorded as a frequency domain interference event. c) Detect narrow pulse / spiking interference in the timing of the analog-to-digital converter, count and integrate it. If the count exceeds the threshold within a given time window, it is determined to be a time-domain interference event. d) If a large number of tracked targets are lost or change significantly, and this is clearly inconsistent with the vehicle's operating conditions, it is recorded as a tracking layer interference event.

7. The 4D mmWave radar antenna of claim 6, wherein, The subband selection and switching includes: a) If the overall interference score of the current working sub-band is lower than the first score threshold, then the current sub-band is considered to be in a strong interference environment: 1) After the subsequent normal FM continuous wave function waveform, insert a listening sequence: transmit the chirps of subband 1 to subband 5 in sequence for several frames; 2) For each operating frequency band, online interference assessment is performed based solely on range Fourier transform for domain energy anomaly detection and time-domain impulse interference counting; 3) Obtain the interference scores of the 5 subbands, sort them, and select the subband with the least interference as the candidate target subband; b) Switching logic with hysteresis: 1) If the overall score of the current sub-band deteriorates but is still higher than the second score threshold, then continue to operate the current sub-band without switching. 2) At the same time, a minimum dwell time is introduced, prohibiting further switching within a specified time after the most recent subband switch; 3) If the current sub-band comprehensive score is lower than the second score threshold and the minimum dwell time requirement is met, then the full sub-band scoring and sorting process will be retried.

8. The 4D mmWave radar antenna of claim 1 and its nested array and anti-jamming design method, characterized in that, The horizontal aperture of the physical array is ≤33 times half a wavelength, and its elevation aperture is ≤15 times half a wavelength.

Images