A complex scene multi-beam millimeter wave radar point cloud imaging method
By combining phase coding with MIMO and phased array radar modes, the problem of target indistinguishability in complex scenarios is solved, multi-beam imaging is achieved, hardware costs are reduced, and real-time detection is improved.
Patent Information
- Application Number
- CN202311662587.8
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2023-12-06
- Publication Date
- 2026-11-06
- Estimated Expiration
- 2043-12-06
AI Technical Summary
In complex scenarios, MIMO radar has difficulty distinguishing targets with similar distances and speeds, the angle resolution algorithm is complex, and phased array radar has poor real-time detection performance and high hardware costs.
By employing phase coding and combining MIMO and phased array radar modes, the system achieves simultaneous detection of wide and narrow beams by transmitting phase-coded signals, and uses two-dimensional FFT to process the data matrix to form multi-beam imaging.
It enables effective detection of weak targets in complex scenarios, reduces hardware computing requirements and costs, and improves the real-time performance and angular resolution of the detection.
Smart Images

Figure CN117687024B_ABST
Abstract
Description
Technical Field
[0001] This invention belongs to the field of point cloud imaging technology, specifically relating to a point cloud imaging method using multi-beam millimeter-wave radar for complex scenes. Background Technology
[0002] Currently, most millimeter-wave radars employ TDMA (Time Division Multiple Access) MIMO technology. In MIMO radar, the transmitting antenna emits orthogonal signals, which increases the number of virtual array elements and the equivalent array aperture, enabling the radar to obtain a narrower beam, lower sidelobes, and higher angle estimation accuracy and resolution.
[0003] Common millimeter-wave radar waveforms use the FMCW (Frequency Modulated Continuous Wave) waveform. For one transmitting antenna to transmit Q repeated FMCW waveforms, such as... Figure 1 A schematic diagram.
[0004] Figure 1 In this context, B represents the bandwidth of the transmitted signal, and t... p The time T is the time to transmit an FMCW signal. p The period for transmitting Q repetitive FMCW signals.
[0005] A schematic diagram of a TDMA MIMO millimeter-wave radar transmission mode is shown below. Figure 2 As shown.
[0006] Figure 2 It can be seen that different transmitting antennas (Tx1, Tx2, Tx3, Tx4) transmit alternately in a time-sharing manner.
[0007] A schematic diagram of a phased array radar transmitting FMCW signals is shown below. Figure 3 As shown.
[0008] Figure 3 It can be seen that all phased array radar transmitters transmit simultaneously, and all transmitting antennas are configured with the same phase control word. In this way, the phased array radar needs a total of L narrow beams to scan the detection area to form the i-th narrow beam.
[0009] In complex scenarios, because MIMO radar emits a wide beam, when two targets are close in distance and speed, making them indistinguishable, the energy of the object reflecting the light more strongly will obscure the target. This is illustrated by the beam emitted by a vehicle-mounted millimeter-wave MIMO radar in a tunnel. Figure 4 As shown.
[0010] Figure 4The walls of the tunnel are strong reflectors, which can cause the echoes of weak targets such as pedestrians and cones at the same distance to be covered by the strong targets. They cannot be distinguished by distance and speed, and can only be distinguished by angle resolution algorithms. However, these algorithms are highly complex and difficult to implement for point cloud radar, which needs to output a large number of detection points.
[0011] In the same scenario, a schematic diagram of phased array radar detection is shown below. Figure 5 As shown.
[0012] from Figure 5 As can be seen, phased array radar can detect weak targets by scanning narrow beams. However, in order to ensure detection capability, each beam transmits Q FMCW waveforms, which requires a certain transmission time. Therefore, assuming a total of L beams are transmitted, it takes L times the transmission time compared to MIMO radar transmitting Q FMCW waveforms, resulting in poor real-time performance. Moreover, phased array radar is more complex to design and more expensive than MIMO radar. Summary of the Invention
[0013] To address the challenges of MIMO radar's wide-beam detection in complex scenarios, where it struggles to distinguish targets with similar distances and velocities, and the computational complexity and high hardware requirements of angle-resolved algorithms, as well as the poor real-time performance and high hardware cost associated with phased array radar's narrow-beam scanning detection, this invention provides a multi-beam millimeter-wave radar point cloud imaging method for complex scenarios. This method utilizes phase coding to achieve simultaneous detection of both wide and narrow beams, transmitting a single set of phase-coded signals to simultaneously enable MIMO radar to detect wide beams and phased array radar to detect narrow beams.
[0014] To achieve the above technical objectives, the present invention adopts the following technical solution:
[0015] A method for imaging point clouds in complex scenes using multi-beam millimeter-wave radar includes the following steps:
[0016] (1) Perform phase modulation transmission on the phase encoding method of the transmitting antenna and determine the phase modulation frequency of different transmitting antennas. A total of Q FMCW waveforms were transmitted;
[0017] (2) Sample N points for each of the Q FMCW waveforms received to obtain an N-row Q-column two-dimensional sampled data matrix, which is the two-dimensional matrix of data sampling in MIMO mode. Then perform a two-dimensional FFT on this two-dimensional matrix.
[0018] (3) Extract data from the N rows and Q columns of the two-dimensional sampling data column by column, with an extraction interval of M, to obtain a total of M groups, each group consisting of N rows. The two-dimensional sampling data matrix of the column is used to obtain the two-dimensional data sampling matrix of M beams in phased array mode;
[0019] (4) Select L sets of two-dimensional matrices with better beamforming effect from the data sampling two-dimensional matrices of M beams, and perform two-dimensional FFT on them.
[0020] Furthermore, in step (1), the phase modulation of different waveforms of the same antenna changes periodically.
[0021] Furthermore, in step (1)
[0022] Furthermore, Q = 512 as described in step (1).
[0023] Furthermore, N = 256 as stated in step (2).
[0024] Furthermore, M = 32 as described in step (3).
[0025] Furthermore, L = 12 as described in step (4).
[0026] Compared with the prior art, the present invention has the following advantages:
[0027] This invention combines MIMO transmission mode and phased array transmission mode using MIMO radar through phase encoding of the transmitted waveform. This enables both wide-beam detection and multi-beam detection, thus overcoming the problems of poor real-time performance of phased array radar and poor detection capability of MIMO radar in complex scenarios. Furthermore, it does not require high computing power from the hardware, thus controlling hardware costs. Attached Figure Description
[0028] Figure 1 This is a schematic diagram of the FMCW waveform.
[0029] Figure 2 This is a schematic diagram of a TDMA MIMO radar.
[0030] Figure 3 This is a schematic diagram of a phased array radar.
[0031] Figure 4 This is a wide beam diagram of a vehicle-mounted millimeter-wave MIMO radar.
[0032] Figure 5 This is a narrow-beam scan diagram of a phased array radar.
[0033] Figure 6 This is a schematic diagram illustrating the complex scene detection implemented by the present invention.
[0034] Figure 7 This is the transmit phase encoding diagram.
[0035] Figure 8 This is a graph of the received signal sampling data.
[0036] Figure 9 This is a schematic diagram of the extraction process.
[0037] Figure 10 This is the radiation pattern of the transmitting antenna.
[0038] Figure 11 This is the radiation pattern of the transmitting antenna. Detailed Implementation
[0039] like Figure 6 A method for imaging point clouds using multi-beam millimeter-wave radar in complex scenes includes the following steps:
[0040] (1) Perform phase modulation transmission on the phase encoding method of the transmitting antenna and determine the phase modulation frequency of different transmitting antennas. A total of Q FMCW waveforms were transmitted;
[0041] (2) Sample N points for each of the Q FMCW waveforms received to obtain an N-row Q-column two-dimensional sampled data matrix, which is the two-dimensional matrix of data sampling in MIMO mode. Then perform a two-dimensional FFT on this two-dimensional matrix.
[0042] (3) Extract data from the N rows and Q columns of the two-dimensional sampling data column by column, with an extraction interval of M, to obtain a total of M groups, each group consisting of N rows. The two-dimensional sampling data matrix of the column is used to obtain the two-dimensional data sampling matrix of M beams in phased array mode;
[0043] (4) Select L sets of two-dimensional matrices with better beamforming effect from the data sampling two-dimensional matrices of M beams, and perform two-dimensional FFT on them.
[0044] To make the present invention more fully disclosed, more specific embodiments are described below.
[0045] This invention combines MIMO mode and phased array mode using MIMO radar and transmit signal phase coding. The basic principle is that each antenna performs different phase modulation coding on each transmitted waveform, and the phase modulation of different waveforms from the same antenna changes periodically, as illustrated in the diagram. Figure 7 As shown.
[0046] Figure 7 In the middle, φ ij This represents the initial phase of the j-th waveform modulation of the i-th antenna. This represents the phase modulation frequency of the k-th transmitting antenna. Using the above transmission mode, Q = 512 FMCW signals are transmitted. Each receiving antenna receives Q = 512 FMCW signals, and each FMCW signal is sampled at N = 256 points, such as... Figure 8 As shown.
[0047] Figure 8If we directly sample the 512 FMCW signals in the above figure to obtain a 256*512 two-dimensional matrix, and then perform two-dimensional FFT processing, since the phase change of each transmitting antenna is periodic and the phase modulation frequency is... Since they are different, different transmitting antennas occupy different positions on the Doppler frequency, realizing the multiplexing of the Doppler domain, which realizes DDMA (Doppler Division Multiple Access) MIMO.
[0048] Furthermore, if let Because the signal phase range is 0-2π, the phase of all FMCW waveforms on the same transmitting antenna changes periodically, with 32 waveforms per cycle. If this invention decimates the 512 transmitted waveforms at intervals of 32, then all 512 / 32 = 16 decimated waveforms will have the same transmit phase modulation. For example, if this invention decimates waveforms [1, 33, 65, 97, ..., 481], then these 16 waveforms will have the same transmit modulation phase. Figure 9 As shown.
[0049] pass Figure 9 It can be seen that the phase modulation of the transmitted signal of these 16 extracted FMCW waveforms is as follows:
[0050]
[0051] Under this phase modulation, the transmitting antenna pattern is as follows: Figure 10 As shown.
[0052] This achieves digital beamforming (DBF) for phased array radar, enabling the detection of targets within ±15° of the radar beam. Similarly, if the [1st, 33rd, 65th, 97th, ..., 481st]th waveform is extracted, the transmit modulation phase of these 16 waveforms is the same, as shown in the following equation:
[0053]
[0054] Under this phase modulation, the transmitting antenna pattern is as follows: Figure 11 As shown.
[0055] Targets detected by this beam are within the radar range of [0°-30°].
[0056] Similarly, this invention can extract 32 sets of waveform sampling matrices with the same modulation. Thus, this invention can obtain 32 beams. Since phase coding must accommodate both MIMO and phased array transmission modes, not all 32 sets will meet the beamforming requirements. Finally, this invention selects 12 sets with good beamforming performance as the final data, and these 12 beams can cover the detection range in MIMO mode.
Claims
1. A method for point cloud imaging in complex scenes using multi-beam millimeter-wave radar, characterized in that, Includes the following steps: (1) Perform phase modulation transmission on the phase encoding method of the transmitting antenna and determine the phase modulation frequency of different transmitting antennas. A total of Q FMCW waveforms were transmitted; (2) Sample N points for each of the Q FMCW waveforms received to obtain an N-row Q-column two-dimensional sampled data matrix, which is the two-dimensional matrix of data sampling in MIMO mode. Then perform a two-dimensional FFT on this two-dimensional matrix. (3) Extract data from the N rows and Q columns of the two-dimensional sampling data column by column, with an extraction interval of M, to obtain a total of M groups, each group consisting of N rows. The two-dimensional sampling data matrix of the column is used to obtain the two-dimensional data sampling matrix of M beams in phased array mode; (4) Select L sets of two-dimensional matrices with better beamforming effect from the data sampling two-dimensional matrices of M beams, and perform two-dimensional FFT on them.
2. The method for point cloud imaging in complex scenes using multi-beam millimeter-wave radar according to claim 1, characterized in that, The phase modulation periodic changes of different waveforms of the same antenna in step (1).
3. The method for point cloud imaging in complex scenes using multi-beam millimeter-wave radar according to claim 1, characterized in that, In step (1) 4. The method for point cloud imaging in complex scenes using multi-beam millimeter-wave radar according to claim 1, characterized in that, Q = 512 as mentioned in step (1).
5. The method for point cloud imaging in complex scenes using multi-beam millimeter-wave radar according to claim 1, characterized in that, The N = 256 mentioned in step (2).
6. The method for point cloud imaging in complex scenes using multi-beam millimeter-wave radar according to claim 1, characterized in that, The M = 32 mentioned in step (3).
7. The method for point cloud imaging in complex scenes using multi-beam millimeter-wave radar according to claim 1, characterized in that, L = 12 as mentioned in step (4).
Citation Information
Patent Citations
Imaging radar device and imaging method
CN109407092A
Non-coherent fusion target detection method of distributed drone MIMO radar
CN110412559A
