MIMO radar target detection method and device based on DDMA waveform modulation

Abstract

The application discloses a MIMO radar target detection method and device based on a DDMA waveform modulation, and the method steps comprise the following steps: receiving echo signals of a DDMA waveform transmitted by a MIMO radar system, sequentially performing distance dimension FFT calculation and velocity dimension FFT calculation to obtain a distance Doppler data matrix; dividing a to-be-searched space range into a plurality of sub-space ranges, sequentially performing cyclic hypothesis testing of the distance Doppler data matrix on each sub-space range, wherein the cyclic hypothesis testing of the distance Doppler data matrix comprises cyclic hypothesis transmission antenna sequence and constructing a virtual MIMO array to perform DBF, and whether a target signal exists in a corresponding direction of each sub-space range is judged according to a virtual MIMO array DBF result obtained in each cyclic hypothesis. The application has the advantages of simple implementation method, low cost, good weak target detection performance, long detection distance, low false negative rate and the like.

Description

Technical Field

[0001] This invention relates to the field of radar target detection technology, and in particular to a MIMO radar target detection method and apparatus based on DDMA waveform modulation. Background Technology

[0002] MIMO (Multiple-Input Multiple-Output) radar utilizes multiple transmitting antennas to simultaneously transmit diversity waveforms and multiple receiving antennas to receive the echo signals. MIMO technology increases the number of virtual channels in the radar, thereby improving its detection performance. For a radar containing N... Tx Root transmitting antenna, N Rx A MIMO radar system with a root receiving antenna can form an N-band radar by employing appropriate antenna layout and waveform design. Tx ×N Rx A virtual antenna array, by performing receive DBF (Digital Beam Forming) on ​​the received signal in the desired direction, can improve the system's detection gain in that direction. However, to form a virtual antenna array, the transmitter must be able to project N signals in a certain dimension. Tx The waveforms from the transmitting antennas are multiplexed, and the receiving end, after receiving this waveform, must be able to reinterpret N in the same dimension. Tx The waveform of the transmitting antenna is separated.

[0003] DDMA (Doppler Multiple Access) separates the signals from all transmitting antennas simultaneously, with each antenna's signal offset by a specific frequency, thus enabling signal separation in the Doppler domain. In MIMO radar, DDMA allows for the simultaneous transmission of signals from all transmitting antennas (N...) Tx Simultaneous transmission of both antennas (roots) reduces the transmission time by one pulse compared to traditional TDMA (Time Division Multiple Access). Under the same transmission signal duration, it can guarantee the detection gain of the virtual antenna DBF (Dead-Factor Filter). In contrast, DDMA requires first detecting the target signal at the receiver and then separating the different transmit antennas at the receiver. That is, the order of the transmit antennas needs to be determined at the receiver. Without determining the transmit antenna order, virtual antenna DBF cannot be performed; only the receive channel DBF can be performed. Therefore, the prerequisite for using DDMA is that the target signal can be analyzed during the target signal detection phase; otherwise, the transmit antenna DBF gain cannot be realized.

[0004] In existing technologies, when using the DDMA method in MIMO radar, pulse accumulation of the echo signal is typically performed first. Since the transmitting antenna is not separated, only DBF (Digital Filtering) of the receiving channel can be performed at this stage. Then, a detection algorithm is used to detect the two-dimensional data matrix after the receiving DBF. Based on the detected values, the transmission is decoupled, and the order of the transmitting antennas is determined for subsequent signal processing algorithms such as angle measurement. For targets with strong echo signals, target signal detection can be achieved through pulse accumulation gain and receiving channel DBF gain, thereby determining the order of different transmitting antennas, and then using a virtual array for angle measurement and other algorithms. However, for targets with weak echo signals (perhaps due to a small target RCS or a large distance from the radar system), pulse accumulation and receiving channel DBF signal gain are insufficient to detect the target signal. That is, the target signal cannot be detected during the target detection stage, resulting in the inability to separate different transmitting antennas and hindering subsequent algorithm processing. In summary, MIMO radar systems using DDMA have poor performance in detecting weak targets, and the system's detection range is relatively short, leading to a risk of missed detections. Summary of the Invention

[0005] The technical problem to be solved by the present invention is: in view of the technical problems existing in the prior art, the present invention provides a MIMO radar target detection method and device based on DDMA waveform modulation that is simple to implement, low in cost, has good weak target detection performance, long detection range and low false alarm rate.

[0006] To solve the above-mentioned technical problems, the technical solution proposed by this invention is as follows:

[0007] A MIMO radar target detection method based on DDMA waveform modulation includes the following steps:

[0008] The echo signal of the DDMA waveform transmitted by the MIMO radar system is received, and range-dimensional FFT (Fast Fourier Transform) calculation and velocity-dimensional FFT calculation are performed sequentially to obtain the range-Doppler data matrix.

[0009] The search area is divided into multiple sub-space domains. The cyclic hypothesis test of the range-Doppler data matrix is ​​performed on each sub-space domain in turn. The cyclic hypothesis test of the range-Doppler data matrix includes cyclically assuming the transmitting antenna order and constructing a virtual MIMO array for DBF. Based on the virtual MIMO array DBF result obtained in each cyclic hypothesis, it is determined whether there is a target signal in the corresponding direction of each sub-space domain.

[0010] Furthermore, the cyclic assumption of the transmit antenna order includes cyclically assuming that some Doppler units in all Doppler units contain the target signal, extracting each Doppler unit that is assumed to contain the target signal, and obtaining the assumed transmit antenna order.

[0011] Furthermore, the construction of the virtual MIMO array for DBF includes:

[0012] Find the corresponding receive channel data based on the transmit antenna data obtained from the transmit antenna sequence obtained in each assumption;

[0013] Based on the received channel arrangement order determined by the retrieved received channel data, a corresponding virtual MIMO array is constructed;

[0014] Based on the pointing angle of the current sub-space, the target signal vector of the virtual MIMO array is subjected to DBF to obtain a synthesized value output.

[0015] Furthermore, the step of determining whether a target signal exists in the direction corresponding to each sub-spatial domain based on the virtual MIMO array DBF result obtained in each cycle includes: judging the composite value obtained in each hypothesis; if there are multiple composite values ​​greater than a preset threshold, it is determined that a target signal exists in the direction of the corresponding sub-spatial domain; and determining the position of the first transmitting antenna in the sub-spatial domain based on the position of the maximum value of the composite value, that is, determining the order of the transmitting antennas.

[0016] Furthermore, the number of hypothesis tests in the cyclic hypothesis testing is N. Chirp N Chirp The length of the Doppler element is assumed to correspond to a certain transmit antenna channel sequence each time.

[0017] Furthermore, in the process of dividing the searched spatial domain into multiple sub-spatial domains, the number of sub-spatial domains is N. th The subspace pointing angle is θ i = -θ + [(i-1)Δθ, iΔθ], the subspace interval is [-θ,+θ] represents the spatial range to be searched, where θ i Let be the subspace pointing angle of the i-th subspace, i = 1, 2, ..., N. th .

[0018] Furthermore, the sequential performance of range-dimensional FFT calculation and velocity-dimensional FFT calculation to obtain the range-Doppler data matrix includes:

[0019] Perform a one-dimensional range-dimensional FFT calculation on each receiving channel of the echo signal to obtain a one-dimensional range matrix;

[0020] The range-Doppler data matrix is ​​obtained by performing pulse accumulation and two-dimensional velocity-dimensional FFT calculation on each distance cell in the distance matrix.

[0021] Angle measurements are performed based on the constructed virtual MIMO array and the detected target signal;

[0022] Furthermore, after the cyclic detection, the method also includes performing angle measurement based on the constructed virtual MIMO array and the detected target signal, and confirming whether the detected target signal is a real target based on the correspondence between the angle measurement result and the pointing angle range of the corresponding sub-space domain.

[0023] Furthermore, the step of confirming whether the detected target signal is a real target based on the correspondence between the angle measurement result and the pointing angle range of the corresponding sub-space domain includes: if the angle measurement result of the target signal is within the pointing angle range of the corresponding sub-space domain, then the target signal is determined to be a real target; if the angle measurement result of the target signal is not within the pointing angle range of the corresponding sub-space domain, then the target signal is determined to be a false target.

[0024] A MIMO radar target detection device based on DDMA waveform modulation includes a processor and a memory, wherein the memory is used to store a computer program, and the processor is used to execute the computer program to perform the method described above.

[0025] Compared with the prior art, the advantages of the present invention are as follows:

[0026] 1. This invention divides the airspace range to be searched into multiple sub-airspace ranges. For each range cell in each sub-airspace range, a cyclic hypothesis test of the range-Doppler data matrix is ​​performed sequentially. This pre-assigns different hypothetical transmit antenna sequences to all target signals. Then, a target signal vector for a virtual MIMO array is constructed accordingly. When the cyclic hypothesis condition is met, the echo signal is superimposed with the virtual MIMO array DBF pointing gain. When the target signal is weak, the superimposed virtual array DBF gain improves the signal-to-noise ratio (SNR) of the weak echo signal target. This effectively allows the location and transmit antenna sequence of the weak echo target signal to be found in the Doppler domain. This solves the problem in traditional MIMO radar systems where weak echo signals cannot be detected in the range-Doppler data matrix under DDMA mode, greatly improving the weak target detection performance of MIMO radar in DDMA mode and reducing the probability of missed detection.

[0027] 2. This invention can improve the detection capability of MIMO radar systems for targets with weak echo signals while maintaining consistency with existing DDMA waveform modulation methods in terms of hardware. Under the same target detection range, it can reduce the minimum RCS detection limit and increase the maximum target detection range under the same target RCS, thereby ensuring the detection performance of the entire radar system. Attached Figure Description

[0028] Figure 1 This is a schematic diagram of the structure and data processing principle of a MIMO radar in a specific application embodiment.

[0029] Figure 2 This is a schematic diagram illustrating the principle of equivalent transceiver digital beamforming for a MIMO virtual array in a specific application embodiment.

[0030] Figure 3 This is a schematic diagram of the structural principle of DDMA in a specific application embodiment.

[0031] Figure 4 This is a schematic diagram of the DDMA waveform modulation effect used in this embodiment.

[0032] Figure 5 This is a schematic diagram of the results of DDMA on four transmit antennas in a specific application embodiment.

[0033] Figure 6 This is a schematic diagram of the results of DDMA on four transmit antennas in a specific application embodiment.

[0034] Figure 7 This is a schematic diagram illustrating the implementation process of MIMO radar target detection based on DDMA waveform modulation in this embodiment.

[0035] Figure 8 This is a schematic diagram of the process for performing cyclic hypothesis testing on sub-space domains in this embodiment.

[0036] Figure 9 This is a schematic diagram of the target range-Doppler two-dimensional spectrum of a strong echo signal obtained in a specific application embodiment.

[0037] Figure 10 This is a schematic diagram of the target range-Doppler two-dimensional spectrum of the weak echo signal obtained in a specific application embodiment.

[0038] Figure 11 This is a schematic diagram of the Doppler frequency domain spectrum corresponding to the target's spatial domain and the distance unit, obtained in a specific application embodiment.

[0039] Figure 12 The DBF pattern corresponding to 11° pointing of the virtual MIMO array is obtained in a specific application embodiment.

[0040] Figure 13 This is a schematic diagram of the detection results of the virtual array DBF synthesized data after the Doppler unit cyclic assumption was obtained in a specific application embodiment.

[0041] Figure 14 This is a schematic diagram of the MIMO angle measurement results obtained in a specific application embodiment. Detailed Implementation

[0042] The present invention will be further described below with reference to the accompanying drawings and specific preferred embodiments, but this does not limit the scope of protection of the present invention.

[0043] like Figure 1 As shown, in MIMO radar, transmit beamforming is equivalently implemented at the receiver. Assuming the transmit and receive arrays are co-located, let the transmit array contain N... T There are 1 array element, and the spacing between the array elements is d. T The receiver array contains N R There are 1 array element, and the spacing between the array elements is d. R The signals transmitted by a MIMO radar are orthogonal to each other. These orthogonal signals are superimposed in space. When the signals are reflected back and received by the receiving array, the orthogonal signals can be separated by matched filtering. That is, a receiving array element can obtain multiple outputs after passing through a set of matched filters. The outputs of each matched filter maintain the relative phase relationship between the transmitting array elements, so that the MIMO radar can form an equivalent transmit beam at the receiving end.

[0044] To form a virtual beam pointing in a specific direction, a set of weight vectors w is used. s The output of the matched filter bank will be weighted, assuming w s It can be represented as:

[0045]

[0046] in, It is the nth r The receiving array element pairs with the nth... t The weight vector coefficients corresponding to the output of each transmitted signal matched filter, where n r =0,1,...,N R -1, n t =0,1,...,N T -1.

[0047] Based on spatial beamforming theory and matched filtering theory, the weight vector w s For a specific direction, to maximize the output SNR in that direction, w s The value should be:

[0048]

[0049] in For the Kronecker product, w T and w R Let s represent the weight vectors for virtual transmit beamforming and receive beamforming, respectively. T (θ0) is the launch steering vector, s R (θ0) is the receiving steering vector, and the beam points towards θ0, wT (θ0) can be expressed as:

[0050]

[0051] w R (θ0) can be expressed as:

[0052]

[0053] like Figure 2 As shown, the equivalent transmit / receive DBF of the MIMO virtual array first performs receive beamforming, with the receive beam pointing in the θ0 direction. All receive array elements match and filter the same transmit signal through w. R (θ0) weighted summation, and then equivalent transmit beamforming is performed on the received beamforming output, that is, through w T (θ0) is a weighted sum of the received beamforming output. The MIMO radar transmit-receive pattern is equivalent to the product of the received pattern and the transmitted pattern, and the equivalent MIMO transmit-receive pattern can form a narrower main beam, which is equivalent to a beam of length N. T N R The radiation pattern formed by a uniform linear array.

[0054] DDMA's MIMO radar uses different transmitting antennas (T) to transmit the echo signal from the same target. x By occupying different positions φ at the target Doppler frequency, the Doppler domain can be multiplexed, such as... Figure 3 As shown, in a DDMA waveform, the velocity values ​​detected in the echoes formed by a target under the signal illumination of different transmitting antennas are different. For a single target, the signals from different transmitting antennas can be separated at the receiving end by utilizing this difference in velocity.

[0055] Assuming the MIMO radar uses DDMA waveform modulation, the radar system contains N Tx Root transmitting antenna, N Rx Root receiving antenna, N chirp The pulse and the single pulse period T c For the transmitting antenna k, the phase shift ω added to the adjacent chirp... k as follows:

[0056]

[0057] Where N = N Tx Specific waveform modulation such as Figure 4 As shown, N Tx One transmitting antenna, N chirp Indicates the number of pulses.

[0058] In DDMA mode, between different chirs of the same transmit antenna, in addition to the Doppler frequency shift caused by target motion, there is also an added DDMA frequency offset, that is, the phase shift difference between pulses of two adjacent transmit antennas is:

[0059]

[0060] For the same target, the Doppler frequency difference in the echoes generated by two adjacent transmitting antennas is:

[0061]

[0062] The corresponding speed difference is:

[0063]

[0064] Among them, V max This represents the maximum velocity value of the echo signal.

[0065] As can be deduced above, DDMA waveform modulation causes the radar's maximum unambiguous velocity range to become 1 / N of the original value. Tx In the echoes generated by two adjacent transmitting antennas, the corresponding Doppler elements differ in phase. (usually an integer).

[0066] Taking four transmitting antennas as an example, without using DDMA, the radar's unambiguous velocity range is [-V max V max After adopting DDMA, it is equivalent to dividing the interval into four sub-intervals of equal length: A, B, C, and D. At this point, the radar's unambiguous velocity range is [-V]. max / 4,V max / 4]. For the same target, the echo signals from the four transmitting antennas will fall into the four intervals respectively, and the velocity difference V between the echo signals corresponding to two adjacent transmitting antennas will be... max / 2. For example... Figure 5 As shown, (a) corresponds to case 1: when V≥0 and V≤V max / 2,Tx1 falls in region C, (b) corresponds to case 2: when V>V max / 2 and V≤V max Tx1 falls in region D, and (c) corresponds to case 3: when V≥-V max And V≤-V max / 2,Tx1 falls in region A, (d) corresponds to case 4: when V>-V max / 2 and V<0, Tx1 falls in region B. From the graph, we can see that when 0≤V≤V maxAt step / 2, the echo signals of Tx1 / Tx2 / Tx3 / Tx4 will fall into sub-intervals C / D / A / B respectively. This does not cause velocity ambiguity. After detecting the target signal, the data of Tx1 / Tx2 / Tx3 / Tx4 can be separated in the order of C / D / A / B, which is DDMA demodulation. However, if the velocity V of the echo signal exceeds this range (such as cases 2 / 3 / 4 corresponding to (b), (c), and (d)), the correspondence between Tx1 / Tx2 / Tx3 / Tx4 and sub-intervals C / D / A / B will change. In this case, different transmit antenna positions cannot be separated by the position of the sub-interval, resulting in DDMA velocity ambiguity. Other methods are needed to complete DDMA demodulation.

[0067] If the number N of the unambiguous velocity sub-intervals of the DDMA waveform is configured in equation (5) above such that N≠N Tx (The choice of N and N) Tx And the set number of pulses N chirp This is relevant; the number of pulses N must be guaranteed. chirp (This is an integer multiple of the number of subintervals N), by adding empty subintervals (subbands), making N... Tx When the signal from a single transmitting antenna is mapped to N sub-intervals, there will be instances where no target signal is mapped in empty sub-intervals. By adding two empty sub-intervals (empty sub-bands) to the existing four transmitting antennas, the radar's unambiguous velocity range [-V] is improved. max V max The system is divided into six sub-intervals of equal length: A, B, C, D, E, and F. The maximum unambiguous velocity for each sub-interval becomes V. max / 3, such as Figure 6 As shown, (a) corresponds to case 1: when V≥0 and V≤V max / 3,Tx1 falls in region D; (b) Corresponding case 2: when V>V max / 3 and V≤2V max / 3,Tx1 falls in region E; (c) Corresponding to case 3: when V>2V max / 3 and V≤V max Tx1 falls in region F; (d) Corresponding to case 4: when V≥-V max And V≤-2V max / 3,Tx1 falls in region A; (e) Corresponding to case 5: when V>-2Vmax / 3 and V≤-V max / 3, Tx1 falls in region B; (f) Corresponding to case 6: when V>-V max / 3 and V<0, Tx1 falls in region C. For a target signal, the echo signals of Tx1 / Tx2 / Tx3 / Tx4 fall into four consecutive cyclic sub-intervals in sequence, and there are no echo signals corresponding to the transmitting antenna in the remaining two empty sub-intervals.

[0068] Depend on Figure 6 It can be seen that the target's actual velocity is within the entire unambiguous velocity range [-V max V max There are six possible distributions of the position on [Tx1] (Tx1 falls within one of the sub-intervals A / B / C / D / E / F). If it can be determined that there is no signal in two of the six sub-intervals ( Figure 6 If (a) is a subinterval of B and C, (b) is a subinterval of C and D, (c) is a subinterval of D and E, (d) is a subinterval of E and F, (e) is a subinterval of A and F, and (f) is a subinterval of A and B, then we can determine which of the six possibilities it belongs to, that is, we can determine:

[0069] 1) The sub-interval numbers (transmitting antenna order) corresponding to Tx1 / Tx2 / Tx3 / Tx4;

[0070] 2) In which sub-interval of the entire unambiguous velocity range does the target's actual velocity lie (velocity deambiguity)?

[0071] In summary, the DDMA method requires identifying sub-intervals that do not contain target signals among all sub-intervals. If the echo signal is weak (small radar cross-section RCS or the target is far from the radar system), it cannot identify empty sub-intervals and sub-intervals containing target signals in the range-Doppler data matrix using detection algorithms. If empty sub-intervals are added, making N... Tx When the signal from the transmitting antenna is mapped to N sub-intervals, there will be a situation where no target signal is mapped in the empty sub-intervals. If it is possible to determine whether there is a signal in each sub-interval, the order of the transmitting antennas can be determined, thereby achieving speed deambiguity.

[0072] This invention performs range-dimensional FFT and velocity-dimensional FFT on the echo signal of the DDMA waveform transmitted by a MIMO radar system. Within the searchable spatial range, the spatial range is divided into multiple sub-spatial ranges. For each range cell in each sub-spatial range, a cyclic hypothesis test of the range-Doppler data matrix is ​​performed on all Doppler cells. This cyclically assumes that some Doppler cells contain target signals, thus pre-assigning different transmit antenna sequences to all target signals, implementing the transmit antenna sequence hypothesis. A target signal vector for a virtual MIMO array is then constructed. If the hypothesis is valid, the output signal is superimposed with the DDMA transmit antenna DBF gain. Therefore, DBF on the target signal vector of this virtual MIMO array can improve the target signal echo gain. When the target signal is weak, by comparing the DBF results of each hypothesis, the presence of target signals in each sub-spatial direction and range cell, as well as the order of the target signals within the sub-ranges (i.e., the transmit antenna sequence), can be determined. This means that even in weak echo signals, the target signal can still be accurately detected, the transmitting antenna sequence of the weak target signal in the corresponding DDMA echo can be determined, the DBF gain of the transmitting antenna can be guaranteed, and the problem of the traditional MIMO radar system being unable to detect the target signal in the range-Doppler data matrix for weak echo signals in DDMA mode can be solved. This greatly improves the weak target detection performance of MIMO radar in DDMA mode and reduces the probability of missed detection.

[0073] like Figure 7 As shown, the steps of the MIMO radar target detection method based on DDMA waveform modulation in this embodiment include:

[0074] Step S01. Receive the radar echo signal of the DDMA waveform transmitted by the MIMO radar system, and perform range-dimensional FFT calculation and velocity-dimensional FFT calculation in sequence to obtain the range-Doppler data matrix.

[0075] Step S101. Perform a one-dimensional range FFT calculation on each receiving channel of the echo signal to obtain a one-dimensional range matrix.

[0076] Step S102. Perform a two-dimensional velocity-dimensional FFT calculation on each distance cell in the distance matrix to obtain the range-Doppler data matrix.

[0077] Pulse accumulation involves merging multiple pulses from the same range gate to improve the target signal-to-noise ratio (SNR). Specifically, this is achieved by performing FFT calculations on all pulses within the same range cell. For the radar echo signal of the DDMA waveform transmitted by the MIMO radar system, performing range-dimensional FFT calculations and velocity-dimensional FFT calculations sequentially yields a range-Doppler data matrix containing range and velocity information.

[0078] Step S02. Cyclic Hypothesis Testing: Divide the search area into multiple sub-space domains and perform cyclic hypothesis testing of the range-Doppler data matrix for each sub-space domain in turn. Cyclic hypothesis testing of the range-Doppler data matrix includes cyclically assuming the transmitting antenna order and constructing a virtual MIMO array for DBF. Based on the virtual MIMO array DBF results obtained from each cyclic hypothesis, determine whether there is a target signal in the direction corresponding to each sub-space domain.

[0079] In this embodiment, the spatial domain to be searched [-θ, +θ] is divided into N... th Subspace, the subspace pointing angle is Subspace Spacing Within the searchable spatial range [-θ, +θ], each sub-spatial region θ is then searched sequentially. i Cyclic hypothesis testing of the range-Doppler data matrix is ​​performed to make cyclic assumptions about the target coordinate information. That is, the cyclic assumption of the transmitting antenna sequence includes cyclically making assumptions about the target signal spatial coordinate information, cyclically making assumptions about the range cell position, and cyclically making assumptions about the sequence of transmitting antennas. Each assumption includes a transmitting antenna channel sequence.

[0080] In this embodiment, the cyclical assumption of the transmit antenna order specifically includes: cyclically assuming that some Doppler cells in all Doppler units contain the target signal, extracting the Doppler cells that are assumed to contain the target signal, and obtaining the assumed transmit antenna order so as to assign different transmit antenna orders to all target signals. Hypothesis testing mainly involves performing a cyclical assumption on all Doppler cells corresponding to each range cell, assuming that some Doppler cells in all Doppler units contain the target signal, and then extracting the Doppler cells that are assumed to contain the target signal. The number of extracted Doppler cells is the number of transmit antenna channels N. Tx In cyclic hypothesis testing, the specific number of hypothesis tests is N. Chirp N Chirp The length of the Doppler element is given.

[0081] In this embodiment, constructing a virtual MIMO array for DBF includes:

[0082] Find the corresponding receive channel data based on the transmit antenna data obtained from the transmit antenna sequence obtained in each assumption;

[0083] Based on the received channel arrangement order determined by the retrieved received channel data, a corresponding virtual MIMO array is constructed, which is the target signal vector array;

[0084] Based on the pointing angle of the current sub-space, the target signal vector of the virtual MIMO array is subjected to DBF to obtain a synthesized output value.

[0085] The target signal vector size of the virtual MIMO array constructed above is specifically N. Tx *N Rx , where N Tx N represents the number of transmit antenna channels. Rx Let N be the number of receiving channels. For a virtual array DBF at a certain spatial angle, the synthesized signal has 10*log10(N) of the virtual array. Tx ×N Rx )dB detection gain.

[0086] In this embodiment, during the cyclic hypothesis testing process, after each hypothesis determines a transmit antenna channel order, the receive channel (the number of channels is N) is then used as the basis for the cyclic hypothesis testing. Rx A virtual MIMO array echo target signal vector is constructed by arranging the signals in the correct order, and then the sub-spatial pointing angle θ is used to determine the target signal vector. i A Virtual MIMO Array DBF is performed on the target signal vector of the virtual MIMO array. The data after the loop is then merged and detected. If all assumptions are true simultaneously, the corresponding data will have a maximum value, thereby determining whether a target signal exists and the order of the target signal's transmitting antennas. In this embodiment, determining whether a target signal exists in the direction corresponding to each sub-spatial domain based on the Virtual MIMO Array DBF results obtained from each loop assumption includes: judging the composite values ​​obtained from each assumption; if multiple composite values ​​are greater than a preset threshold, it is determined that a target signal exists in the direction of the corresponding sub-spatial domain; and determining the position of the first transmitting antenna in the sub-spatial domain based on the location of the maximum value of the composite value, thus determining the order of the transmitting antennas.

[0087] In specific application embodiments, such as Figure 8 As shown, the detailed steps of cyclic hypothesis testing include:

[0088] Step S201. Loop through the search space range [-θ, +θ], where the loop step is... Number of loops N th The range of the i-th subspace is: θ i =-θ+[(i-1)Δθ,iΔθ], the center line is 0°, θ i Let be the subspace pointing angle of the i-th subspace, i = 1, 2, ..., N. th ;

[0089] Step S202. Loop through all distance cell ranges [1, M], with a loop step of 1 and a loop count of M, where the j-th distance cell is R. j ;

[0090] Step S203. For all Doppler cell ranges [1, N] Chirp The transmitting antenna sequence is assumed to be looped, with a loop step of 1 and a loop count of N. ChirpThe k-th type of Doppler unit sequence is Where, N Eb This indicates the number of empty subspaces. When the sequence value of a selected Doppler cell exceeds the pulse number N... Chirp Then subtract N from the sequential value. Chirp express;

[0091] Step S204. Find the corresponding receive channel data for the transmit antenna data obtained in step S203, and construct a virtual MIMO array vector (vector size: [1, N). Tx *N Rx ]), and perform virtual MIMO array DBF (DBF steering vector size is: [1, N Tx *N Rx ]), to obtain a composite value;

[0092] Step S205. Repeat steps S203 and S204 to obtain a set of length N. Chirp The combined values ​​are used to select the highest target signal using a detection algorithm. If a target signal is detected, the corresponding Doppler unit order and transmitting antenna order are determined. If no target signal is detected, the loop is exited and step S202 is entered for the next loop.

[0093] This embodiment combines cyclic hypothesis testing and MIMO virtual array DBF to achieve target signal detection with superimposed transmit antenna gain, enabling the detection of target signals across the entire airspace and determining the Doppler cell positions and transmit antenna sequences in each sub-airspace. This effectively improves the radar system's detection performance for weak signal targets (those with small radar cross-section σ or targets far from the radar system).

[0094] This embodiment includes, but is not limited to, performing multiple cyclic hypothesis tests on the range Doppler echo signal spectrum data of the MIMO radar system to improve the overall detection performance of the radar system.

[0095] Step S03. Perform angle measurement based on the constructed virtual MIMO array and the detected target signal. Based on the correspondence between the angle measurement results and the pointing angle range of the corresponding sub-space domain, confirm whether the detected target signal is a real target.

[0096] Besides distance and velocity, the angle of the target relative to the radar is a crucial piece of information characterizing the target. However, the target signal detected in step S02, containing only distance and velocity information, may include false targets. Since the distance from the target to each receiving antenna in the MIMO radar varies, the phase of the received signal will differ. In this embodiment, after detecting the target signal in step S02, angle measurement is further performed based on the virtual MIMO array and the detected target signal. The angle measurement search interval is the sub-spatial angle coverage range. The angle measurement results are used to ultimately confirm whether the detected target signal is a real target. If the target signal angle measurement result is within the pointing angle range of the corresponding sub-spatial domain, the target signal is determined to be a real target; if the target signal angle measurement result is not within the pointing angle range of the corresponding sub-spatial domain, the target signal is determined to be a false target. This can further improve the accuracy of target detection, suppress false alarms caused by misjudgment, and reduce the false alarm rate. The angle measurement algorithm can employ algorithms such as subspace-based DOA angle estimation algorithms, and the specific configuration can be selected according to actual needs.

[0097] This embodiment divides the spatial domain range [-θ, +θ] to be searched into multiple sub-spatial domain ranges (the number of sub-spatial domains is N). th The subspace pointing angle is θ i Subspace interval For each sub-spatial region, range-Doppler data matrix hypothesis testing is performed sequentially. For each range cell, a cyclic hypothesis is made for all Doppler cells, assuming that some Doppler cells contain the target signal. Each hypothesis includes a transmit antenna channel order. The Doppler cells that are assumed to contain the target signal are then extracted (the number of extracted Doppler cells is equal to the number of transmit antenna channels N). Tx Based on the order of the receiving channels, construct an N... Tx *N Rx The virtual MIMO array echo target signal vector, and then based on the sub-spatial pointing angle θ i The virtual MIMO array target signal vector is subjected to a virtual MIMO array DBF (the DBF pointing angle is the set sub-spatial pointing angle θ). i This yields a composite value that can increase the echo gain of the target signal to 10*log10(N). Tx ), by considering a certain subspace direction θ i The Rth i N distance units are used for N Chirp Second hypothesis, by comparing N ChirpThe synthesized value of this hypothesis can determine whether a target signal exists in the direction and range cell of the sub-space domain, and the order of the target signal in the sub-interval (transmit antenna order). By searching the entire spatial range [-θ, +θ] according to the above method, the target signal detection with superimposed transmit antenna gain can be completed using the MIMO virtual array DBF, improving the radar system's detection performance for weak signal targets (small radar cross-section σ or target far from the radar system). Finally, angle measurement is performed based on the detected target signal and the constructed virtual MIMO array. If the target signal angle measurement result is within the pointing angle range of the corresponding sub-space domain, the target signal is determined to be real; if the target signal angle measurement result is not within the pointing angle range of the corresponding sub-space domain, the target signal is determined to be false and discarded, which can further suppress false alarms caused by misjudgment.

[0098] To verify the effectiveness of the present invention, a simulation experiment of MIMO radar target detection based on DDMA waveform modulation was conducted using the method of the present invention in a specific application embodiment, wherein the DDMA step phase was set as follows: When the target RCS is 20 square meters (relatively large RCS), the target distance to the radar R = 300 m, the target speed relative to the radar V = 100 km / h, and the target angle relative to the radar A = 11° (strong echo target), the obtained range-Doppler two-dimensional spectrum corresponding to the second receiving channel is as follows: Figure 9 As shown. By Figure 9 It can be seen that the range-Doppler two-dimensional spectrum is divided into 8 sub-intervals (6 sub-intervals occupied by the transmitting antennas + 2 empty sub-intervals). The rectangles represent the positions of the target signal within the 6 transmitting antenna sub-intervals, and the empty sub-intervals do not contain the target signal. For targets with a large RCS, the positions of the target signal in each sub-interval and the empty sub-intervals can be separated from the range-Doppler spectrum directly by pulse accumulation or superimposing the receiver DBF gain. However, when the RCS is 0.1 square meters and the target distance from the radar R = 500m (weak echo signal target), such as Figure 10 As shown, the range-Doppler two-dimensional spectrum of a weak echo signal target is almost entirely noise, making it impossible to find the target signal from the spectrum. The target signal position and transmitting antenna sequence cannot be determined by DDMA waveform modulation and demodulation, which will result in the target being missed.

[0099] This invention divides the spatial scanning range [-15°, 15°] into 10 sub-spatial domains, with a sub-spatial domain cyclic step of 3°, a range element range of [1, 1024], a cyclic step of 1, and a Doppler element range of [1, 256] for assuming a cyclic transmission antenna sequence. When the sub-spatial domain search interval is [9°, 12°] and the range search element is 500, the corresponding Doppler element spectrum of the second receiving channel is as follows: Figure 11 As shown. From Figure 11 It can be seen that the target echo signal cannot be found from the noise signal in the Doppler frequency domain, let alone the order of the transmitting antennas. The DBF pattern of the virtual MIMO array pointing to the 11° position in the spatial domain is shown below. Figure 12 As shown, the virtual MIMO array has a DBF gain of 16.8dB in the 11° direction, which meets the design theoretical specifications. After applying the cyclic assumption and virtual MIMO array DBF to these 256 Doppler elements, a set of synthetic data with a length of 256 is obtained. By setting a threshold to detect this set of data, if multiple values ​​are greater than the set threshold, the position of the maximum value is the position of the first transmit antenna in the sub-spatial domain, thus determining the order of the transmit antennas in the corresponding frequency domain.

[0100] The detection results of the virtual array DBF synthesized data after the Doppler element cyclic assumption are as follows: Figure 13 As shown, since the current assumption is that the antenna order is not perfectly aligned, the detection threshold detected 8 data points. The Doppler element has its maximum value in the 139th assumption, so the 139th Doppler element can be determined as the coordinates of the target signal falling in the sub-spatial domain corresponding to the first transmitting antenna. Transmitting antenna T x1 ,T x2 ,T x3 ,T x4 ,T x5 ,T x6 The corresponding Doppler elements are [139, 172, 203, 235, 11, 43]. Therefore, the method of this invention can effectively detect the target signal and determine the transmitting antenna sequence from weak echo signals. Further, based on the position of the detected target signal and the corresponding transmitting antenna sequence in the Doppler domain, MIMO angle measurement is performed, such as... Figure 14 As shown, the angle measurement result is 10.98°, which is basically consistent with the midline angle of 11° in the set sub-spatial search range [9°, 12°]. Therefore, the currently detected weak target signal is considered to be a real target signal (the angle of the real target signal is 11°). That is, combining the angle measurement result can further filter out false alarms.

[0101] This invention, based on traditional DDMA, performs cyclic hypothesis testing on the airspace to be searched, range domain, and Doppler domain within a sub-airspace. When the cyclic hypothesis is valid, the echo signal is superimposed with the pointing gain of a virtual MIMO array DBF. When the target echo signal is weak, the possible Doppler domain position of the target signal is cyclically assumed, and the hypothesis results are then used for synthetic data detection. Since the superimposed virtual array DBF gain improves the signal-to-noise ratio (SNR) of the weak echo signal target, the position of the weak echo target signal (transmit antenna sub-airspace position) and the sequence of the transmit antennas can be effectively found in the Doppler domain. Then, through secondary confirmation by MIMO angle measurement, the radar system can accurately detect the weak echo signal and suppress false alarms caused by interference signals.

[0102] This invention can improve the detection capability of MIMO radar systems for targets with weak echo signals while maintaining consistency with existing DDMA waveform modulation in hardware. Under the same target detection range, it can reduce the minimum RCS detection limit and increase the maximum target detection range under the same target RCS, thereby ensuring the detection performance of the entire radar system.

[0103] This embodiment also provides a MIMO radar target detection device based on DDMA waveform modulation, including a processor and a memory. The memory is used to store computer programs, and the processor is used to execute the computer programs to perform the methods described above.

[0104] The above description is merely a preferred embodiment of the present invention and is not intended to limit the invention in any way. Although the present invention has been disclosed above with reference to preferred embodiments, it is not intended to limit the invention. Therefore, any simple modifications, equivalent changes, and alterations made to the above embodiments based on the technical essence of the present invention without departing from the scope of the present invention should fall within the protection scope of the present invention.

Claims

1. A MIMO radar target detection method based on DDMA waveform modulation, characterized in that the steps include... include: The echo signal of the DDMA waveform transmitted by the MIMO radar system is received, and range-dimensional FFT calculation and velocity-dimensional FFT calculation are performed sequentially to obtain the range-Doppler data matrix. The search area is divided into multiple sub-space domains. The cyclic hypothesis test of the range-Doppler data matrix is ​​performed on each sub-space domain in turn. The cyclic hypothesis test of the range-Doppler data matrix includes cyclically assuming the transmitting antenna order and constructing a virtual MIMO array for DBF. Based on the virtual MIMO array DBF result obtained in each cyclic hypothesis, it is determined whether there is a target signal in the corresponding direction of each sub-space domain.

2. The MIMO radar target detection method based on DDMA waveform modulation according to claim 1, characterized in that, The cyclically assumed transmit antenna order includes cyclically assuming that some Doppler elements in all Doppler elements contain the target signal, extracting each Doppler element that is assumed to contain the target signal, and obtaining the assumed transmit antenna order.

3. The MIMO radar target detection method based on DDMA waveform modulation according to claim 1, characterized in that, The construction of the virtual MIMO array for DBF includes: The corresponding receive channel data is found based on the transmit antenna data obtained from the assumed transmit antenna sequence each time. Based on the received channel arrangement order determined by the retrieved received channel data, a corresponding virtual MIMO array is constructed; Based on the pointing angle of the current sub-space, the target signal vector of the virtual MIMO array is subjected to DBF to obtain a synthesized value output.

4. The MIMO radar target detection method based on DDMA waveform modulation according to claim 3, characterized in that, The step of determining whether a target signal exists in the direction corresponding to each sub-space domain based on the virtual MIMO array DBF result obtained in each cycle includes: judging the composite value obtained in each hypothesis; if there are multiple composite values ​​greater than a preset threshold, it is determined that a target signal exists in the direction of the corresponding sub-space domain; and determining the position of the first transmitting antenna in the sub-space domain based on the position of the maximum value of the composite value, that is, determining the order of the transmitting antennas.

5. The MIMO radar target detection method based on DDMA waveform modulation according to claim 1, characterized in that, The number of hypothesis tests in the cyclic hypothesis testing is N. Chirp N Chirp The length of the Doppler element is assumed to correspond to a certain transmit antenna channel sequence each time.

6. The MIMO radar target detection method based on DDMA waveform modulation according to any one of claims 1 to 5, characterized in that, In the process of dividing the spatial domain to be searched into multiple sub-spatial domains, the number of sub-spatial domains is N. th The subspace pointing angle is θ i = -θ + [(i-1)Δθ, iΔθ], the subspace interval is [-θ,+θ] represents the spatial range to be searched, where θ i Let be the subspace pointing angle of the i-th subspace, i = 1, 2, ..., N. th .

7. The MIMO radar target detection method based on DDMA waveform modulation according to any one of claims 1 to 5, characterized in that, The sequential performance of range-dimensional FFT and velocity-dimensional FFT calculations yields the range-Doppler data matrix, including: Perform a one-dimensional range-dimensional FFT calculation on each receiving channel of the echo signal to obtain a one-dimensional range matrix; The range-Doppler data matrix is ​​obtained by performing pulse accumulation and two-dimensional velocity-dimensional FFT calculation on each distance cell in the distance matrix.

8. The MIMO radar target detection method based on DDMA waveform modulation according to any one of claims 1 to 5, characterized in that, The cyclic detection process further includes angle measurement based on the constructed virtual MIMO array and the detected target signal. Based on the correspondence between the angle measurement results and the pointing angle range of the corresponding sub-space domain, it is confirmed whether the detected target signal is a real target.

9. The MIMO radar target detection method based on DDMA waveform modulation according to claim 8, characterized in that, The step of confirming whether the detected target signal is a real target based on the correspondence between the angle measurement result and the pointing angle range of the corresponding sub-space domain includes: if the angle measurement result of the target signal is within the pointing angle range of the corresponding sub-space domain, then the target signal is determined to be a real target; if the angle measurement result of the target signal is not within the pointing angle range of the corresponding sub-space domain, then the target signal is determined to be a false target.

10. A MIMO radar target detection device based on DDMA waveform modulation, comprising a processor and a memory, wherein the memory is used to store a computer program, characterized in that, The processor is used to execute the computer program to perform the method as described in any one of claims 1 to 9.

Images