A Radar Time-Doppler Dual-Domain Joint Encoding Waveform Design and Processing Method

Abstract

This invention belongs to the field of radar signal processing technology, and relates to a method for designing and processing radar time-Doppler dual-domain joint coding waveforms. The invention designs two sets of coprime Doppler frequency modulation schemes, the essence of which is that each set of Doppler modulation frequencies is non-uniformly spaced, the adjacent Doppler frequency spacing of each transmitting antenna is unique, and there is no common factor with the other transmitting antennas. At this time, the two sets of coprime Doppler frequency division multiplexing waveforms have mismatch characteristics, which can provide richer information during transmission matching and improve the matching success rate.

Description

Technical Field

[0001] This invention belongs to the field of radar signal processing technology and relates to a method for designing and processing radar time-Doppler dual-domain joint coding waveforms. Background Technology

[0002] In recent years, MIMO (Multiple Input Multiple Output) technology has been widely used in automotive radar, with array sizes continuously expanding to provide higher sensing dimensions and better angular resolution. However, MIMO radar applications require the design of orthogonal waveforms to achieve transmit waveform separation. Orthogonal waveform selection typically employs various modes such as time-division multiplexing, frequency-division multiplexing, and code-division multiplexing, achieving waveform orthogonality in the time, frequency, and coding domains, respectively. Time-division multiplexing waveforms transmit signals in a time-division manner, exhibiting good orthogonality, but this leads to reduced pulse repetition frequency and energy loss due to signal accumulation. Frequency-division multiplexing waveforms transmit multiple frequency signals simultaneously, also exhibiting good orthogonality, but this waveform places higher hardware requirements. Code-division multiplexing waveforms transmit signals of the same frequency but different phases simultaneously; orthogonality is limited by coding degrees of freedom and the transmitting antenna, but implementation is simple with low hardware complexity.

[0003] To reduce hardware complexity and achieve better orthogonal performance, some scholars have proposed that the Doppler frequency division multiplexing waveform is a way to achieve orthogonality of MIMO waveforms. This waveform uses phase coding technology to achieve frequency division multiplexing of transmitted signals in the Doppler frequency domain. It has good orthogonality, is simple to implement, and has low hardware requirements, making it highly valuable for applications in the field of automotive radar.

[0004] Traditional Doppler frequency division multiplexing (FDDM) waveforms mainly suffer from transmit matching and target peak collision problems. Transmit matching arises because the echo signals from N transmitting antennas are focused on the Doppler spectrum, but the relationship between the focal point and the transmitting antennas is unknown. Matching is required to determine the target Doppler frequency; this phenomenon is also known as transmit ambiguity. Target peak collision can be described as the indistinguishable collision and coupling of echoes from two different transmitting antennas in a two-dimensional spectrum. Since multiple transmitting antenna echo signals are focused on the Doppler spectrum, the probability of target peak collision is higher. Furthermore, Doppler FDDM waveforms require mapping the focal points of multiple echo signals on the Doppler spectrum to transmitting array elements to form a virtual array angle measurement. Therefore, target peak collision not only causes ambiguity in velocity measurement but also significantly impacts virtual array synthesis and angle measurement.

[0005] For the transmit matching problem in traditional Doppler frequency division multiplexing, two main processing schemes are currently used. The first method uses array signal processing, i.e., angular dimension processing, to resolve transmit ambiguity. However, the resolution effect is greatly affected by the measured angle. The second method uses a non-uniform Doppler frequency offset on the transmitting antenna and matches the transmitting antenna with N transmit echo positions to achieve deambiguity. This method is more robust and widely used. However, there is still no effective solution for the above-mentioned transmit matching methods when generating data frames with target peak collision problems.

[0006] Furthermore, the current Doppler frequency division multiplexing waveform only provides a solution for emission matching. To meet the requirements of high-speed scenarios, it is usually necessary to combine it with other speed deblurring algorithms. Summary of the Invention

[0007] The technical problem solved by this invention is to overcome the shortcomings of the prior art and propose a radar time-Doppler dual-domain joint coding waveform design and processing method. This method can effectively solve the transmission matching problem and the velocity deambiguation problem, meet the angle and velocity measurement requirements of MIMO arrays, and has a certain robustness in multi-target situations. This method jointly designs two sets of waveforms with different time intervals and adopts a coprime Doppler frequency modulation scheme. The two sets of waveforms are alternately transmitted by time division multiplexing. For the transmission matching problem and the velocity deambiguation problem, the non-equal interval characteristics of the coprime Doppler frequency modulation scheme are used to complete the transmission matching by position matching and determine whether there is a collision and the type of collision. The collision-free matching point of the dual waveforms is selected and the velocity deambiguation is performed by delaying the phase of time division multiplexing. The collision-free waveform is selected for MIMO virtual array angle measurement.

[0008] The technical solution of this invention is:

[0009] A method for designing and processing radar time-Doppler dual-domain joint coding waveforms, the steps of which include:

[0010] Step 1: Based on the number of transmitting antennas and the radar velocity measurement range, design two sets of orthogonal waveforms with different coprime Doppler modulation frequency phase coding values ​​and different sequence time intervals;

[0011] Step 2: The radar alternately transmits the two sets of orthogonal waveforms designed in Step 1 using time-division multiplexing. At the same time, the radar receiver collects the two sets of echo signals and performs two-dimensional fast Fourier transform on the two sets of echo signals to obtain two sets of multi-channel range Doppler spectra.

[0012] Step 3: Perform non-coherent multi-channel accumulation of the two sets of multi-channel range Doppler spectra obtained in Step 2 to obtain the range Doppler detection spectrum. Then, perform Doppler dimension constant false alarm rate target detection on the obtained range Doppler detection spectrum at each distance point to obtain target range index information and target Doppler index information. Perform target matching based on the obtained target Doppler index information. Determine whether the target has collided based on the target matching result. If a collision has occurred, determine the collision type to obtain the collision type result.

[0013] Step 4: Based on the target distance index information and target Doppler index information obtained in Step 3, as well as the target matching results, target collision results, and collision type results, the target angle is measured and the velocity is de-ambigued to obtain the target's velocity and angle, thus completing the radar time-Doppler dual-domain joint coding waveform design and processing.

[0014] In step one, the method for designing two sets of orthogonal waveforms with different coprime Doppler modulation frequencies and different sequence time intervals, based on the number of transmitting antennas and the radar velocity measurement range, is as follows:

[0015] 101. Given N transmitting antennas and I bits for the phase shifter, the minimum Doppler frequency shift unit of the Doppler frequency division multiplexing waveform is 1 / 2 of the pulse repetition frequency. I The Doppler frequency offset coefficients of the first group of waveforms are designed as vectors of length N. The Doppler frequency deviation coefficients of the second group of waveforms are vectors of length N. Set D1(1) = D2(1) to 0 to 2 I The random value of -1 is used to obtain D1(n) and D2(n) when n>1 using the following formula;

[0016]

[0017]

[0018] 102. Let the radar carrier frequency be f0 and the speed of light be c = 3·10 8 m / s, wavelength λ=c / f0, if the maximum velocity that the radar needs to measure is V max Set the time T1 of the first waveform sequence and the time T2 of the second waveform sequence to satisfy the following equation.

[0019]

[0020] 103. Since the two sets of waveforms are transmitted alternately, the pulse repetition period for each signal is T = T1 + T2. Continuing from 101, the Doppler modulation frequencies of the two sets of signals are... and Constructing a slow-time phase encoding matrix and Specifically, the phase coding values ​​of the nth transmitting antenna in the lth sequence of the first group of waveforms and the second group of waveforms are shown in Equation (6) and Equation (7), respectively;

[0021]

[0022]

[0023]

[0024]

[0025] In step two, the method for obtaining two sets of multi-channel distance-Doppler spectra is as follows:

[0026] 201. Two sets of waveforms are transmitted alternately using time-division multiplexing. The phase code value of the first set of waveforms is... The time encoding value is T1; the phase encoding value of the second group of waveforms is... The time encoding value is T2. At each moment, the linear frequency modulated continuous wave signal is phase modulated along the slow time and transmitted simultaneously by N transmitting antennas.

[0027] 202. With M receiving antennas and K fast-time sampling points, the echo signal, after de-chirp processing and ADC sampling at the receiver, yields two sets of echo signals, s1∈C. L×K×M With s2∈C L×K×M .

[0028] 203. The echo signals s1 and s2 of the two waveforms are respectively subjected to two-dimensional Fourier transforms to obtain two sets of multi-channel range-Doppler spectra S1∈C. L×K×M With S2∈C L×K×M ;

[0029] In step three, the method for obtaining the distance Doppler detection spectrum is as follows:

[0030] The multi-channel range Doppler spectra S1 and S2 are summed modulo-squared along the receiving channel dimension to obtain the noncoherently accumulated range Doppler detection spectrum C1∈R. L×K With C2∈R L×K ;

[0031] The method for obtaining target distance index information and target Doppler index information is as follows:

[0032] The obtained range-Doppler detection spectrum is used for constant false alarm rate (CFAR) target detection at each range point. This process iterates through all range points, extracting the Doppler spectrum at the k-th range point between C1 and C2 for CFAR detection, resulting in J1 and J2 over-detected Doppler points respectively. The target range index information is recorded as k, and the target Doppler index information is recorded as ind1, ind2, ... …、

[0033] The hypothesis testing method is used for target matching. The specific method is as follows:

[0034] 301. Assuming the current Doppler index point is the focusing position of transmitting antenna 1, initialize the matching identification vector. It is a zero vector of length J1+J2;

[0035] 302. Traverse the first J1 over-detected target Doppler index information, and construct the target Doppler index vector on the j-th target Doppler index. Used to extract the peak vector from the corresponding target points.

[0036]

[0037]

[0038] 303. If the peak vector When the difference between the maximum and minimum values ​​is greater than 3dB, the match fails, and the process returns to step 302 to continue with the next target Doppler index target matching; when the peak vector When the difference between the maximum and minimum values ​​is no greater than 3dB, the peak vectors are considered to originate from the same target point, the match is successful, and the value is recorded. This is the Doppler index vector of the target;

[0039] 304. Index the over-detected Doppler data as ind1, ind2, ... …、 and In comparison, if in the corresponding sequence index Similar to the over-detected Doppler index, the matching identifier vector... Increment the value at the corresponding position by 1, and return to step 302 to continue the target matching of the next target Doppler index;

[0040] 305. After completing target matching of J1 over-detected target Doppler indices, determine the matching identifier vector. If all values ​​are less than or equal to 1, no collision occurs; if there is a value greater than 1 in the vector, the position greater than 1 is located to obtain the collision Doppler index, and the obtained collision Doppler index is compared with the Doppler index vectors of all targets. By comparison, the collision type is obtained.

[0041] In step four, the angle measurement method is as follows:

[0042] 401. Complex sequences are obtained by extracting multi-channel echo peak amplitude and phase information from the multi-channel range-Doppler spectrum S1 and S2 using the target range index and Doppler index. and Based on the collision detection result, if a collision occurs, proceed to step 402; if no collision occurs, proceed to step 403.

[0043] 402. Determine whether the collision occurred in the first or second waveform group based on the collision type. If the collision occurred in the first waveform group, select P2 to form a MIMO virtual array, obtain the spatial spectrum through angle-dimensional Fourier transform, and locate the target angle in the spatial spectrum. If the collision occurred in the second waveform group, select P1 to form a MIMO virtual array, obtain the spatial spectrum through angle-dimensional Fourier transform, and locate the target angle in the spatial spectrum.

[0044] 403, by and Two sets of MIMO virtual arrays are formed respectively. The spatial spectrum is obtained by angle-dimensional Fourier transform. The target angle is located in the spatial spectrum and the peak amplitude and phase anglePeak1 and anglePeak2 at that angle are extracted from the spatial spectrum.

[0045] The speed-based defuzzification method is as follows:

[0046] 411. First, calculate the target blur velocity v using the following formula. amb , where ind amb The Doppler index for the current target point.

[0047]

[0048] 412. Based on the collision detection results, if a collision occurs, let the number of transmitting antennas involved in the collision be... exist and The peak point of the selected non-collision transmitting antenna is denoted as and The phase difference ω is calculated using formula (16). If it is determined that no collision has occurred, the phase difference ω is calculated using formula (21).

[0049]

[0050] and After comparing the phases and obtaining the multi-channel phase differences, phase smoothing is performed to obtain the phase difference ω.

[0051]

[0052] 413. To compensate for the time delay phase of the fuzzy velocity, after rounding the phase, the phase difference δ caused by the velocity fuzziness factor is solved as shown in the following formula;

[0053]

[0054] 414. Determine the coefficient b based on the range of integer phase δ;

[0055]

[0056] 415. And solve for the Doppler blur factor q;

[0057]

[0058] 416. Substitute the velocity fuzzy factor to obtain the true velocity v of the current target.

[0059] v = v amb +q·2V max .

[0060] Compared with existing technologies, its advantages are:

[0061] (1) Effectively solves the transmission matching problem of Doppler frequency division multiplexing waveforms under multi-target conditions, with a high success rate.

[0062] This invention designs two sets of coprime Doppler frequency modulation schemes. The essence of this scheme is that each set of Doppler modulation frequencies is non-uniformly spaced, and the distance between adjacent Doppler frequencies of each transmitting antenna is unique and has no common factor with the other transmitting antennas. At this time, the two sets of coprime Doppler frequency division multiplexing waveforms have mismatch characteristics, which can provide richer information during transmission matching and improve the success rate of matching.

[0063] (2) Effectively solves the Doppler ambiguity problem and expands the velocity measurement range.

[0064] This invention encodes the time of two sets of Doppler frequency division multiplexing waveforms, that is, it designs two sets of waveform sequences with different time intervals for time-division alternating transmission, and uses the dual waveform time delay phase for velocity deambiguation, thereby achieving a larger velocity measurement range, and the multi-channel MIMO array improves the robustness of phase deambiguation.

[0065] (3) Reduce the impact of multi-target peak collision and improve the robustness of multi-target angle measurement and defuzzification.

[0066] This invention is based on two sets of coprime Doppler frequency modulation schemes, which can significantly reduce the impact of collisions. When a collision occurs, the target velocity and angle measurement can continue. First, due to the unmatched characteristics of the two sets of coprime Doppler frequency modulation schemes, the collision only occurs in a single waveform. Therefore, after the dual waveforms are matched, the non-collision waveform can be selected to continue to complete the angle measurement. Furthermore, the coprime modulation scheme of the single waveform can reduce the number of collision points, and the non-collision points can continue to be used for velocity deambiguation.

[0067] (4) It is simple to implement and has low computational cost.

[0068] This invention has low hardware requirements, simple waveform implementation, and the transmission matching method and velocity deambiguation method are numerical calculations, resulting in low overall processing complexity. Attached Figure Description

[0069] Figure 1 Radar time-Doppler dual-domain jointly coded waveform time-frequency phase diagram;

[0070] Figure 2 Distance Doppler detection spectrum;

[0071] Figure 3 Target velocity measurement results;

[0072] Figure 4 Target location measurement results. Detailed Implementation

[0073] This invention proposes a method for designing and processing radar time-Doppler dual-domain joint coding waveforms. The overall processing flow of this method is shown in the figure, and the specific implementation is as follows:

[0074] Step 1: Based on the requirements of the number of transmitting antennas and the velocity measurement range, two sets of coprime Doppler frequency division multiplexing orthogonal waveforms with different time intervals are designed. The two sets of orthogonal waveforms are alternately transmitted through time division multiplexing, while the radar receiver collects the target echo in real time.

[0075] 101. Given N transmitting antennas and I bits for the phase shifter, the minimum Doppler frequency shift unit of the Doppler frequency division multiplexing waveform is 1 / 2 of the pulse repetition frequency. I The Doppler frequency offset coefficients of the first group of waveforms are designed as vectors of length N. The Doppler frequency deviation coefficients of the second group of waveforms are vectors of length N. Set D1(1) = D2(1) to 0 to 2 I The random value of -1 is used to obtain D1(n) and D2(n) when n>1 using the following formula;

[0076]

[0077]

[0078] 102. Let the radar carrier frequency be f0 and the speed of light be c = 3·10 8 m / s, wavelength λ=c / f0, if the maximum velocity that the radar needs to measure is V max Set the time T1 of the first waveform sequence and the time T2 of the second waveform sequence to satisfy the following equation.

[0079]

[0080] 103. Since the pulse repetition period of the two sets of alternating waveforms is T = T1 + T2, continue to obtain the Doppler modulation frequencies of the two sets of signals from 101. and Constructing a slow-time phase encoding matrix and Specifically, the phase coding values ​​of the nth transmitting antenna in the lth sequence of the first group of waveforms and the second group of waveforms are shown in Equation (6) and Equation (7), respectively;

[0081]

[0082]

[0083]

[0084]

[0085] Step 2: Two sets of orthogonal waveforms are alternately transmitted using time-division multiplexing. At the same time, the radar receiver collects the signal echoes and performs two-dimensional fast Fourier transforms on the two sets of echo signals to obtain two sets of multi-channel range Doppler spectra.

[0086] 201. The two sets of waveforms are transmitted alternately using time-division multiplexing, such as... Figure 1 As shown, the phase encoding value of the first group of waveforms is The time encoding value is T1; the phase encoding value of the second group of waveforms is... The time encoding value is T2. At each moment, the linear frequency modulated continuous wave signal is phase modulated along the slow time and transmitted simultaneously by N transmitting antennas.

[0087] 202. With M receiving antennas and K fast-time sampling points, the echo signal, after de-chirp processing and ADC sampling at the receiver, yields two sets of echo signals, s1∈C. L×K×M With s2∈C L×K×M .

[0088] 203. The echo signals s1 and s2 of the two waveforms are respectively subjected to two-dimensional Fourier transforms to obtain two sets of multi-channel range-Doppler spectra S1∈C. L×K×M With S2∈CL×K×M ;

[0089] Step 3: The two sets of multi-channel range-Doppler spectra are non-coherently accumulated to obtain the range-Doppler spectrum. Then, Doppler-dimensional constant false alarm rate target detection is performed point by point. Target matching is performed using Doppler index information to determine whether a collision has occurred and the type of collision.

[0090] 301. The range-Doppler spectra S1 and S2 of the multi-channel range-Doppler spectrum are summed modulo-squared along the receiving channel dimension to obtain the noncoherently accumulated range-Doppler detection spectrum as C1∈R. L×K With C2∈R L×K ;

[0091] 302. Perform constant false alarm rate (CFAR) target detection at each distance point using the obtained range-Doppler detection spectrum. Traverse all distance points, extract the Doppler spectrum at the k-th distance point between C1 and C2 for CFAR detection, obtaining J1 and J2 over-detected Doppler points respectively. Record the target distance index information as k, and the target Doppler index information as ind1, ind2, ... …、

[0092] 303. Traverse the first J1 over-detected target Doppler index information, and construct the target Doppler index vector on the j-th target Doppler index. Used to extract the peak vector from the corresponding target points.

[0093]

[0094]

[0095] 304. If the peak vector When the difference between the maximum and minimum values ​​is greater than 3dB, the match fails, and the process returns to step 303 to continue with the next target Doppler index target matching; when the peak vector When the difference between the maximum and minimum values ​​is no greater than 3dB, the peak vectors are considered to originate from the same target point, the match is successful, and the value is recorded. This is the Doppler index vector of the target;

[0096] 305. Index the over-detected Doppler data as ind1, ind2, ... …、 and In comparison, if in the corresponding sequence index Similar to the over-detected Doppler index, the matching identifier vector... Increment the value at the corresponding position by 1, and return to step 302 to continue the target matching of the next target Doppler index;

[0097] 306. After completing target matching of J1 over-detected target Doppler indices, determine the matching identifier vector. If all values ​​are less than or equal to 1, no collision occurs; if there is a value greater than 1 in the vector, the position greater than 1 is located to obtain the collision Doppler index, and the obtained collision Doppler index is compared with the Doppler index vectors of all targets. By comparison, the collision type is obtained.

[0098] Step 4: Sequentially complete the velocity deblurring and angle measurement for all detected target points. Using the distance index and Doppler index information of the detected target points, extract the target peak value from the two sets of multi-channel range-Doppler spectra. Obtain the velocity delay phase difference by comparing the phases of the corresponding non-collision antennas, and use this velocity delay phase difference to deblur the target velocity.

[0099] 401. Complex sequences are obtained by extracting multi-channel echo peak amplitude and phase information from the multi-channel range-Doppler spectrum S1 and S2 using the target range index and Doppler index. and Based on the collision detection result, if a collision occurs, proceed to step 402; if no collision occurs, proceed to step 403.

[0100] 402. Determine whether the collision occurred in the first or second waveform group based on the collision type. If the collision occurred in the first waveform group, select P2 to form a MIMO virtual array, obtain the spatial spectrum through angle-dimensional Fourier transform, and locate the target angle in the spatial spectrum. If the collision occurred in the second waveform group, select P1 to form a MIMO virtual array, obtain the spatial spectrum through angle-dimensional Fourier transform, and locate the target angle in the spatial spectrum.

[0101] 403, by and Two sets of MIMO virtual arrays are formed respectively. The spatial spectrum is obtained by angle-dimensional Fourier transform. The target angle is located in the spatial spectrum and the peak amplitude and phase anglePeak1 and anglePeak2 at that angle are extracted from the spatial spectrum.

[0102] 404. First, calculate the target blur velocity v using the following formula. amb , where ind amb The Doppler index for the current target point.

[0103]

[0104] 405. Based on the collision detection results, if a collision occurs, let the number of transmitting antennas involved in the collision be... exist and The peak point of the selected non-collision transmitting antenna is denoted as and The phase difference ω is calculated using formula (16). If it is determined that no collision has occurred, the phase difference ω is calculated using formula (21).

[0105]

[0106]

[0107] 406. To compensate for the time delay phase of the fuzzy velocity, after rounding the phase, the phase difference δ caused by the velocity fuzziness factor is solved as shown in the following formula;

[0108]

[0109] 407. Determine the coefficient b based on the range of integer phase δ;

[0110]

[0111] 408. And solve for the Doppler blur factor q;

[0112]

[0113] 409. Substitute the velocity fuzzy factor to obtain the true velocity v of the current target.

[0114] v = v amb +q·2V max .

[0115] 408. Output the target angle and actual speed to complete the processing.

[0116] Example

[0117] The effectiveness of this invention can be demonstrated through the following simulation data experiments.

[0118] In this implementation case, the initial position parameters of the seven targets are set as shown in the table below:

[0119] Table 1 Simulation target parameters

[0120]

[0121]

[0122] Parameter settings:

[0123] Number of transmitting antennas N = 3

[0124] Number of receiving lines M = 4

[0125] Number of frequency modulation sequences L = 512

[0126] Fast time sampling number K = 512

[0127] The first group of Doppler frequency division multiplexing waveform sequences has an interval T1 = 35µs.

[0128] The second group of Doppler frequency division multiplexing waveform sequences has an interval of T2 = 25µs.

[0129] The Doppler modulation coefficients for the first group of waveforms are D1 = [0, 22, 45].

[0130] The Doppler modulation coefficients for the second group of waveforms are D2 = [0, 12, 35].

[0131] In this implementation example, the echo signal, after matched filtering and target detection and transmission matching, yields the over-detection point and the echo point information of transmitting antenna 1, as shown below. Figure 2 As shown, at this point, seven echo points from transmitting antenna 1 can be obtained on each spectrum. It is evident that this waveform can complete the transmission matching process for all targets. In particular, when targets 2 and 3 at the same distance but different speeds collide at the 230th distance point, the spectra of both sequences can correctly match the echo points from transmitting antenna 1 when the collision occurs in sequence 1. Subsequently, velocity de-ambiguity and angle measurement are completed using the over-check points and matching results. In the de-ambiguity process, the blurred velocity is determined by the echo points from transmitting antenna 1, but there are multiple possible velocities under different velocity ambiguity multiples, such as... Figure 3 As shown, after deblurring, the target velocity measurement value is unblurred, indicating that the measured value corresponds to the true value, and the velocity measurement range is better than [-65m / s, 65m / s]. In the target angle measurement processing, angle estimation can be completed for all 7 targets, as shown by... Figure 4 As can be seen, the target position estimation results basically correspond to the true values, indicating that the target angle estimation error is small. Specifically, the estimation results of the velocity and angle parameters of all targets are shown in the table. In this experiment, the ranging deviation of all targets is less than 0.2m, the angle deviation is less than 0.5°, and the velocity deviation is less than 0.2m / s. Furthermore, the velocity de-ambiguation processing can be correctly completed under different velocity ambiguity multiples, and the final velocity measurement range is better than [-65m / s, 65m / s], meeting the requirements of the vehicle environment. In summary, this invention can successfully solve the transmission matching problem of Doppler frequency division multiplexing waveforms, obtain the correct ambiguous velocity of the target, and form a virtual array for angle measurement. Moreover, the experimental results show that after phase de-ambiguation processing using time division multiplexing variable interval sequences, the radar velocity measurement range can be expanded, and this method has a certain robustness to collision scenarios.

[0132] Table 2. Target parameter estimation results

[0133]

[0134] This invention proposes a radar time-Doppler dual-domain joint coding waveform design and processing method. It utilizes two sets of coprime Doppler modulation frequencies to perform transmission matching of the echo signal to complete angle measurement, and uses variable time-phase difference for velocity deambiguation. This invention effectively solves the transmission matching problem and the velocity deambiguation problem, meeting the angle and velocity measurement requirements of MIMO arrays, and exhibits certain robustness in multi-target scenarios. In summary, the above are merely preferred embodiments of this invention and are not intended to limit the scope of protection of this invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this invention should be included within the scope of protection of this invention.

Claims

1. A method for designing and processing radar time-Doppler dual-domain joint coding waveforms, the steps of which include: Step 1: Based on the number of transmitting antennas and the radar velocity measurement range, design two sets of orthogonal waveforms with different coprime Doppler modulation frequency phase coding values ​​and different sequence time intervals; Step 2: The radar alternately transmits the two sets of orthogonal waveforms designed in Step 1 using time-division multiplexing. At the same time, the radar receiver collects the two sets of echo signals and performs two-dimensional fast Fourier transform on the two sets of echo signals to obtain two sets of multi-channel range Doppler spectra. Step 3: Perform non-coherent multi-channel accumulation of the two sets of multi-channel range Doppler spectra obtained in Step 2 to obtain the range Doppler detection spectrum. Then, perform Doppler dimension constant false alarm rate target detection on the obtained range Doppler detection spectrum at each distance point to obtain target range index information and target Doppler index information. Perform target matching based on the obtained target Doppler index information. Determine whether the target has collided based on the target matching result. If a collision has occurred, determine the collision type to obtain the collision type result. Step 4: Based on the target distance index information and target Doppler index information obtained in Step 3, as well as the target matching results, target collision results, and collision type results, the target angle is measured and the velocity is de-ambigued to obtain the target's velocity and angle, thus completing the radar time-Doppler dual-domain joint coding waveform design and processing.

2. The radar time-Doppler dual-domain joint coding waveform design and processing method according to claim 1, characterized in that: In step one, the method for designing two sets of orthogonal waveforms with different coprime Doppler modulation frequencies and different sequence time intervals, based on the number of transmitting antennas and the radar velocity measurement range, is as follows:

101. Given N transmitting antennas and I-bit phase shifters, design the Doppler frequency offset coefficient of the first waveform as a vector of length N. The Doppler frequency deviation coefficients of the second group of waveforms are vectors of length N. Set D1(1) = D2(1) to 0 to 2 I The random value of -1 is used to obtain D1(n) and D2(n) when n>1 using the following formula; 102. Let the radar carrier frequency be f0 and the speed of light be c = 3·10 8 m / s, wavelength λ=c / f0, if the maximum speed that the radar needs to measure is V max Set the time T1 of the first waveform sequence and the time T2 of the second waveform sequence to satisfy the following equation.

103. Since the two sets of waveforms are transmitted alternately, the pulse repetition period for each signal is T = T1 + T2. Continuing from 101, the Doppler frequency offset coefficient is obtained... and Construct the slow-time phase encoding matrices for the two sets of waveforms respectively. and As shown in equations (4) and (5), specifically, the phase coding values ​​of the nth transmitting antenna in the lth sequence of the first group of waveforms and the second group of waveforms are shown in equations (6) and (7), respectively.

3. The radar time-Doppler dual-domain joint coding waveform design and processing method according to claim 2, characterized in that: In step two, the method for obtaining two sets of multi-channel distance-Doppler spectra is as follows:

201. Two sets of waveforms are transmitted alternately using time-division multiplexing. The phase code value of the first set of waveforms is... The time encoding value is T1; the phase encoding value of the second group of waveforms is... The time encoding value is T2. At each moment, the linear frequency modulated continuous wave signal is phase modulated along the slow time and transmitted simultaneously by N transmitting antennas.

202. With M receiving antennas and K fast-time sampling points, the echo signal, after de-chirp processing and ADC sampling at the receiver, yields two sets of echo signals, s1∈C. L×K×M With s2∈C L×K×M ; 203. The echo signals s1 and s2 of the two waveforms are respectively subjected to two-dimensional Fourier transforms to obtain two sets of multi-channel range-Doppler spectra S1∈C. L×K×M With S2∈C L×K×M .

4. The radar time-Doppler dual-domain joint coding waveform design and processing method according to claim 3, characterized in that: In step three, the method for obtaining the distance Doppler detection spectrum is as follows: The multi-channel range Doppler spectra S1 and S2 are respectively summed modulo-squared along the receiving channel dimension to obtain the noncoherently accumulated range Doppler detection spectrum C1∈R. L×K With C2∈R L×K ; The method for obtaining target distance index information and target Doppler index information is as follows: The obtained range-Doppler detection spectrum is used for constant false alarm rate (CFAR) target detection at each range point. All range points are traversed, and the Doppler spectrum is extracted at the k-th range point between C1 and C2 for CFAR detection, resulting in J1 and J2 over-detected Doppler points respectively. The target range index information is recorded as k, and the target Doppler index information is as follows:

5. The radar time-Doppler dual-domain joint coding waveform design and processing method according to claim 4, characterized in that: In step three, the hypothesis testing method is used for target matching. The specific method is as follows:

301. Assuming the current Doppler index point is the focusing position of transmitting antenna 1, initialize the matching identification vector. It is a zero vector of length J1+J2; 302. Traverse the first J1 detected target Doppler index information, starting from the j-th target Doppler index ind. j Construct the Doppler index vector of the target launch channel Used to extract the corresponding target peak vector 303. If the peak vector When the difference between the maximum and minimum values ​​is greater than 3dB, the match fails, and the process returns to step 302 to continue with the next target Doppler index target matching; when the peak vector When the difference between the maximum and minimum values ​​is no greater than 3dB, the peak vectors are considered to originate from the same target point, the match is successful, and the value is recorded. This is the Doppler index vector of the target's transmission channel; 304. Over-detected Doppler index and In comparison, if in the corresponding sequence index Similar to the over-detected Doppler index, the matching identifier vector... Increment the value at the corresponding position by 1, and return to step 302 to continue the target matching of the next target Doppler index; 305. After completing target matching of J1 over-detected target Doppler indices, determine the matching identifier vector. If all values ​​are less than or equal to 1, no collision occurs; if there is a value greater than 1 in the vector, the position greater than 1 is located to obtain the collision Doppler index, and the obtained collision Doppler index is compared with the Doppler index vectors of all targets. By comparison, the collision type is obtained.

6. The radar time-Doppler dual-domain joint coding waveform design and processing method according to claim 5, characterized in that: In step four, the angle measurement method is as follows:

401. Extract multi-channel echo peak amplitude and phase information from the multi-channel range-Doppler spectrum S1 and S2 using the target range index and the target transmission channel Doppler index vector, respectively. and Based on the collision detection result, if a collision occurs, proceed to step 402; if it is determined that no collision occurs, proceed to step 403.

402. Determine whether the collision occurred in the first or second waveform group based on the collision type. If the collision occurred in the first waveform group, then select... A MIMO virtual array is formed, and the spatial spectrum is obtained through angle-dimensional Fourier transform. The target angle is then located within the spatial spectrum. If the collision occurs in the second waveform group, then... A MIMO virtual array is formed, and the spatial spectrum is obtained through angle-dimensional Fourier transform. The target angle is then located in the spatial spectrum. 403, by and Two sets of MIMO virtual arrays are formed respectively. The spatial spectrum is obtained by angle-dimensional Fourier transform. The target angle is located in the spatial spectrum and the peak amplitude and phase anglePeak1 and anglePeak2 at that angle are extracted from the spatial spectrum.

7. The radar time-Doppler dual-domain joint coding waveform design and processing method according to claim 6, characterized in that: In step four, the speed defuzzification method is as follows:

411. First, calculate the target ambiguity velocity v using the following formula. amb , where ind amb Extracted from The first value; 412. Based on the collision detection results, if a collision occurs, let the number of transmitting antennas involved in the collision be... exist and The peak point of the selected non-collision transmitting antenna is denoted as and The phase difference ω is calculated using formula (16). If it is determined that no collision has occurred, the phase difference ω is calculated using formula (21). and After comparing the phases and obtaining the multi-channel phase differences, phase smoothing is performed to obtain the phase difference ω.

413. To compensate for the time delay phase of the fuzzy velocity, after rounding the phase, the phase difference δ caused by the velocity fuzziness factor is solved as shown in the following formula; 414. Determine the coefficient b based on the range of integer phase δ; 415. And solve for the Doppler blur factor q; 416. Substitute the velocity fuzzy factor to obtain the true velocity v of the current target; v=v amb +q·2V max 。

Images