A radar waveform and signal processing method combining TDM and DDM

Abstract

The application relates to the technical field of MIMO radars, and discloses a radar waveform combining TDM and DDM, a signal processing method and device and a medium, which comprise a modulated signal transmitting module, a return signal receiving module and a return signal processing module, the application combines the advantages of TDM and DDM, solves the problem of speed ambiguity caused by TDM through slow-time random sampling, the waveform can be well applied to large-scale arrays, avoids speed ambiguity, improves the utilization rate of a Doppler band, a relatively wide Doppler sub-band improves the accuracy of target detection, meanwhile, the application provides a corresponding signal processing procedure for the waveform, and provides a method for separating channels through cyclic correlation using a template, the corresponding target transmitting channels can be quickly and accurately separated, the demand for angle measurement can be met, and the method provided by the application can be compatible with a traditional FMCW radar, and has strong application flexibility.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of MIMO radar, in particular to a radar waveform and signal processing method combining TDM and DDM. BACKGROUND

[0002] Multiple-input Multiple-output (MIMO) radar is a kind of radar that uses multiple antennas for signal transmission and reception at the transmitting end and the receiving end. MIMO radar uses multi-dimensional processing of signals to synthesize a virtual array using the positional relationship of the transmitting and receiving elements, expand the radar aperture, and achieve high-precision detection and tracking of targets. Different transmitting antennas are allocated to different time slots to virtually form MxN channel data, and the correspondence between each channel data and the transmitting antenna is determined. After rearranging the channel data according to the positional relationship of the virtual elements, the spatial spectrum of the target is estimated.

[0003] TDM (Time Domain Multiplexing) waveform is a design method that divides the radar transmission waveform in time. Different transmitting antennas are used in different time slots to separate the transmission channels. Traditional TDM waveforms use a fixed sequence for transmission, and each transmission has a large gap in time, which causes a decrease in unambiguous velocity. At the same time, there is a coupling between velocity and space. On a large-scale array, as the number of transmissions increases, the unambiguous velocity further decreases, making it difficult for TDM waveforms to meet the speed range requirements.

[0004] DDM (Doppler Domain Multiplexing) waveform is a design method based on Doppler frequency. It divides the radar transmission signal in the Doppler domain to achieve simultaneous detection and tracking of multiple targets. DDM uses different transmission initial phases in the slow time to make different transmitting antennas occupy different Doppler frequency bands, thereby achieving channel separation. DDM waveforms allow multiple transmitting antennas to transmit simultaneously, reducing time waste and improving unambiguous velocity. However, there is a problem of determining the correspondence between Doppler frequency bands and transmitting antennas. For a large-scale array, the more transmitting antennas there are, the more Doppler sub-bands need to be divided, which can cause frequency band congestion and affect target detection.

[0005] Therefore, for a large-scale array, using either of the above two waveforms alone cannot effectively solve the problem of channel separation. SUMMARY

[0006] To address the deficiencies of the prior art, the present application provides a radar waveform and signal processing method combining TDM and DDM, which solves the problems raised in the background art.

[0007] To solve the above technical problems, according to one aspect of the present application, more specifically, a radar waveform and signal processing method combining TDM and DDM, comprising TDM phase and DDM phase combination, specifically comprising the following steps:

[0008] S1, divide all M to be allocated transmitting antennas into M Group groups, wherein M can be divided by M Group ;

[0009] S2, determine the multiplexing times of each group of transmission and determine the random transmission sequence of M Group groups, at the same time, determine the initial phase of each transmitting antenna in each time slot, and transmit signals according to the determined transmission initial phase, and receive echo signals;

[0010] S3, mix the echo signal with the reference signal and sample, and perform distance dimension FFT on the sampled signal;

[0011] S4, according to the transmission sequence of each group determined in S2, perform 2D FFT in slow time, and divide the obtained results according to Doppler sub-band;

[0012] S5, non-coherent accumulation is performed on the Doppler sub-band dimension, transmission group dimension and receiving channel dimension;

[0013] S6, CFAR detection is performed on the non-coherent accumulation result, and 2D FFT data in all sub-bands corresponding to each target is taken out;

[0014] S7, cyclic correlation is performed on the target data and the template data, wherein the channel separation cyclic correlation method is adopted, the displacement number corresponding to the maximum value is recorded, which is the position corresponding to the first transmitting antenna, and the remaining transmitting channel data is taken out in turn;

[0015] S8, rearrange all channel data according to virtual channel position, perform angle dimension FFT on the rearranged data, and obtain target spatial spectrum;

[0016] S9, CFAR detection is performed on the spatial spectrum to obtain the angle information of the target;

[0017] S10, output the distance, velocity, angle and signal-to-noise ratio information of the target.

[0018] Further, it comprises a modulated signal transmitting module, an echo signal receiving module and an echo signal processing module, characterized in that the modulated signal transmitting module: for modulated signal generation and signal transmission;

[0019] The echo signal receiving module: for receiving the signal transmitted by the modulated signal transmitting module;

[0020] Echo signal processing module: used for echo signal distance dimension, speed dimension and angle dimension FFT.

[0021] Further, the working process of the modulation signal transmitting module is:

[0022] (1) modulate the signal and generate a random Tx transmission sequence;

[0023] (2) determine the transmitting subarray according to the current pulse index, and calculate the modulation phase in this transmission according to the subarray information;

[0024] (3) generate a chirp signal and change the initial phase through a phase shifter;

[0025] (4) select the corresponding Tx group and transmit the corresponding modulation signal.

[0026] Further, the working process of the echo signal processing module is:

[0027] (1) rearrange the data according to the transmission sequence, zero-pad the rearranged data, and perform speed dimension FFT;

[0028] (2) group the speed dimension according to the number of sub-bands, and perform non-coherent accumulation on the channel dimension and sub-band dimension;

[0029] (3) separate the transmitting channels according to the target, rearrange the channel data according to the virtual channel position relationship, and obtain the target spatial spectrum;

[0030] (4) perform CFAR detection on the spatial spectrum, obtain the target information, and deliver the target information.

[0031] Further, the transmission grouping method is:

[0032] (1) determine the number of groups M Group to be divided, and M Group can be evenly divided by the total number of Tx M, that is, each group contains Tx;

[0033] (2) determine the transmitting Tx number in each group according to non-replacement extraction.

[0034] Further, the transmission sequencing method is:

[0035] (1) determine the multiplexing number N of each Tx group, then the total number of chirp is M Group ×N, and one transmission of all Tx groups is called a Burst, that is, N Bursts are transmitted;

[0036] (2) determine the transmission order in each Burst, that is, generate a 1~M GroupA random sequence, the transmission sequence of all bursts constitutes an array with a size of 1xNM Group , recording the transmission Tx group number of each chirp;

[0037] (3) According to the sequence of each burst, determine the transmission time of each Tx group.

[0038] Further, the method of slow-time 2DFFT is:

[0039] (1) According to the transmission sequence, select the corresponding N echo data of one Tx group from all M Group ×N echo data;

[0040] (2) Fill N echo data into the corresponding transmission sequence according to the transmission sequence, and fill 0 in the position where the Tx group does not transmit, to construct a data matrix with a size of 1xNM Group ;

[0041] (3) Perform NM Group point FFT on the data matrix.

[0042] Further, the channel separation cyclic correlation method is:

[0043] (1) According to the detected target index, take its value at the corresponding position of all Doppler sub-bands, construct an array with a size of k band according to the sub-band position, and construct a cyclic correlation template;

[0044] (2) Calculate the cyclic correlation result of the template sequence and the data array;

[0045] (3) Find the peak value in the correlation result, which is the Tx peak position with the transmission number 1, and calculate the peak position of the remaining m-1 Tx according to the Doppler sub-band width.

[0046] The beneficial effects of the TDM and DDM combined radar waveform and signal processing method are:

[0047] The application combines the advantages of TDM and DDM, and solves the problem of speed ambiguity caused by TDM by slow-time random sampling, the waveform can be well applied to large-scale array, avoids speed ambiguity, improves the utilization rate of Doppler band, a wider Doppler sub-band improves the accuracy of target detection, at the same time, for the waveform, the application provides a corresponding signal processing process, especially provides a method of using a template to perform cyclic correlation to separate channels, which can quickly and accurately separate the transmission channels of the corresponding target, can meet the demand of angle measurement, in addition, the method provided by the application can be compatible with the traditional FMCW radar, when the transmission is sequenced, the transmission in sequence is the traditional FMCW waveform, and has strong application flexibility. BRIEF DESCRIPTION OF DRAWINGS

[0048] The application will be further described in detail below in combination with the drawings and specific implementation methods.

[0049] Figure 1 It is a radar system flowchart provided by the application based on TDM and TDM and DDM combined waveform.

[0050] Figure 2 It is a modulation signal generation method flowchart provided by the application based on TDM and TDM and DDM combined waveform.

[0051] Figure 3 It is a schematic diagram of a modulation signal transmitting module provided by the application.

[0052] Figure 4 It is a flowchart of a return signal processing method provided by the application.

[0053] Figure 5 It is a schematic diagram of a return signal processing module provided by the application.

[0054] Figure 6 It is a radar device schematic diagram provided by the application based on TDM and TDM and DDM combined waveform.

[0055] Figure 7 It is a modulation signal time domain schematic diagram provided by an embodiment of the application.

[0056] Figure 8 It is a physical array schematic diagram provided by another embodiment of the application.

[0057] Figure 9 It is a virtual array schematic diagram obtained from the physical array shown in Figure 8

[0058] Figure 10 It is a target speed-amplitude spectrum in an embodiment.

[0059] ​Figure 11 is Figure 10 Target velocity index-amplitude spectrum in an embodiment.

[0060] Figure 12 is Figure 10 FFT result figure of S2 of a target in an embodiment.

[0061] Figure 13 is Figure 10 Spatial FFT result figure of a target in an embodiment.

[0062] Figure 14 is a velocity FFT result figure of a target in another embodiment. DETAILED DESCRIPTION

[0063] Hereinafter, the present application will be described in detail with reference to the accompanying drawings and embodiments. It should be noted that the embodiments in the present application and the features in the embodiments can be combined with each other without conflict.

[0064] As Figure 1 shown, according to one aspect of the present application, a radar waveform and signal processing method combining TDM and DDM is provided, including TDM phase and DDM phase combination, specifically including the following steps:

[0065] S1, divide all M to-be-allocated transmitting antennas into M Group groups, wherein M can be divided by M Group groups evenly;

[0066] S2, determine the multiplexing times of each group of transmissions and determine the random transmission sequence of M Group groups, at the same time, determine the initial phase of each transmitting antenna in each time slot, and transmit a signal according to the determined transmission initial phase, and receive a return signal;

[0067] S3, mix the return signal with a reference signal and sample, and perform distance dimension FFT on the sampled signal;

[0068] S4, according to the transmission sequence of each group determined in S2, perform 2D FFT in slow time, and divide the obtained result according to Doppler sub-band;

[0069] S5, perform non-coherent accumulation on the Doppler sub-band dimension, the transmitting group dimension and the receiving channel dimension;

[0070] S6, perform CFAR detection on the non-coherent accumulation result, and take out the 2D FFT data in all sub-bands corresponding to each target;

[0071] S7, cyclically correlating the target data and the template data, wherein a channel separation cyclic correlation method is adopted, a displacement number corresponding to a maximum value is recorded as a position corresponding to a first transmitting antenna, and remaining transmitting channel data is sequentially taken out;

[0072] S8, rearranging all channel data according to virtual channel positions, performing angle dimension FFT on the rearranged data, and obtaining target spatial spectrum;

[0073] S9, performing CFAR detection on the spatial spectrum, and obtaining angle information of the target;

[0074] S10, outputting distance, speed, angle and signal-to-noise ratio information of the target.

[0075] In the embodiment, the system comprises a modulated signal transmitting module, an echo signal receiving module and an echo signal processing module, and the modulated signal transmitting module is configured to generate and transmit modulated signals.

[0076] The echo signal receiving module is configured to receive signals transmitted by the modulated signal transmitting module.

[0077] The echo signal processing module is configured to perform distance dimension, speed dimension and angle dimension FFT on echo signals.

[0078] In the embodiment, the working process of the modulated signal transmitting module is as follows:

[0079] (1) modulate signals and generate a random Tx transmitting sequence;

[0080] (2) determine a transmitting subarray according to a current pulse index, and calculate a modulated phase in the transmission according to subarray information;

[0081] (3) generate a chirp signal and change an initial phase through a phase shifter;

[0082] (4) select a corresponding Tx group, and transmit a corresponding modulated signal.

[0083] In the embodiment, the working process of the echo signal processing module is as follows:

[0084] (1) rearrange data according to a transmitting sequence, zero-pad the rearranged data, and perform speed dimension FFT;

[0085] (2) group speed dimensions according to a number of sub-bands, and perform non-coherent accumulation on channel dimensions and sub-band dimensions;

[0086] (3) separate transmitting channels according to targets, rearrange channel data according to virtual channel position relationships, and obtain target spatial spectrum;

[0087] (4) CFAR detection is performed on the spatial spectrum to obtain target information and deliver the target information.

[0088] In the embodiment, the method for transmitting the packets is as follows:

[0089] (1) Determine the number M of groups to be divided Group , and M Group is an integer that can divide the total number M of Tx, that is, each group contains Tx;

[0090] (2) Determine the Tx number in each group according to non-replacement extraction.

[0091] In the embodiment, the method for transmitting the packets is as follows:

[0092] (1) Determine the multiplexing number N of each Tx group, and the total number of chirps is M Group ×N, and one Burst is completed when all Tx groups complete one transmission, that is, N Bursts are transmitted;

[0093] (2) Determine the transmission sequence in each Burst, that is, generate a random sequence of 1~M Group , and the transmission sequence of all Bursts forms an array of 1×NM Group , which records the Tx group number of each chirp;

[0094] (3) Determine the transmission time of each Tx group according to the sequence of each Burst.

[0095] Further, the method for slow-time 2D FFT is as follows:

[0096] (1) According to the transmission sequence, select N echo data corresponding to one Tx group from the total M Group ×N echo data;

[0097] (2) Fill the N echo data into the corresponding transmission sequence according to the transmission sequence, and fill 0 in the positions where the Tx group does not transmit, to construct a data matrix of 1×NM Group ;

[0098] (3) Perform NM Group point FFT on the data matrix.

[0099] Further, the channel separation cyclic correlation method is as follows:

[0100] (1) According to the detected target index, take the values at the corresponding positions of all Doppler sub-bands, construct an array of k band according to the sub-band positions, and construct a cyclic correlation template;

[0101] (2) Calculate the cyclic correlation result of the template sequence and the data array;

[0102] (3) Find the peak value in the correlation result, which is the Tx peak position of the group with the number 1. According to the Doppler sub-band width, the peak positions of the remaining m-1 Txs are calculated in turn.

[0103] Embodiment 1:

[0104] The transmit array is 16 elements, the number of Tx groups M Group = 4, each group contains four transmit elements, and each time one Tx group transmits simultaneously. The grouping situation is shown in the following table:

[0105] Table 1 Tx grouping situation

[0106]

[0107] This embodiment illustrates the phase modulation process of DDMA. According to the attached Figure 2 The steps of generating the transmit waveform according to the method are as follows:

[0108] First, a Tx group transmit sequence is randomly generated, which contains 4 Chirps in each Burst. The sequence in each Burst is a permutation of [1, 2, 3, 4]. The Tx group sequence of two Bursts in this embodiment is shown in the following table:

[0109] Table 2 Tx group sequence

[0110]

[0111] Then, according to Table 2, the corresponding transmit group is selected, and the modulation phase in the group is calculated. In this embodiment, the number of empty bands k empty = 2 for unwrapping, and the total number of Doppler sub-bands k band = 6. The modulation phase required for each transmission is calculated according to the following formula:

[0112]

[0113] Where p represents the pth Tx group, n p,r represents the chirp number when the rth transmission of the Tx in the group, and q represents the qth Tx in the group. The modulation phases in the above two Bursts are shown in the following table, and the empty positions in the table represent no transmission:

[0114]

[0115] Table 3 Transmit modulation phase

[0116] Finally, the calculated phase is sent to the transmitting channel in the current Tx group, and the phase modulation is completed through the phase shifter. The four Txs in the group simultaneously transmit the phase-modulated linear frequency modulation signal, and the transmission waveform is as shown in Figure 7 .

[0117] Embodiment 2

[0118] The antenna array used is as shown in Figure 8 , which contains 48 Tx elements and 48 Rx elements, and the grouping condition is that the four consecutive Txs in the vertical direction are one group, and there are M Group =12 groups, the number of fast time sampling points is 256, the number of slow time Tx group multiplexing is N=192, and the total number of Chirps is 2304. Therefore, the size of the echo data cube is: 256x48x2304, and the target information setting is: 60m, -10m / s, 10° in azimuth, and 5° in elevation.

[0119] The signal processing flow is as shown below:

[0120] 1. First, 256-point distance dimension FFT is performed on the echo data, and the latter half data is discarded. At this time, the data size is 128x48x2304;

[0121] 2. The data of different Tx groups is extracted according to the Tx transmission sequence. At this time, the data size becomes 128x192x48x12;

[0122] 3. The positions not transmitted by the slow time dimension Tx group are supplemented with 0, and the data size is 128x2304x48x12;

[0123] 4. FFT processing is performed on the slow time dimension, and the result is as shown in Figure 12 . Due to the non-uniform sampling of the slow time dimension, the sidelobe of the velocity dimension is raised, and the peak-to-average sidelobe ratio is approximately , N is the number of Tx group multiplexing, and m is the number of Txs in each group. In this embodiment, N=192 and m=4, and the calculated peak-to-average sidelobe ratio is 16.8dB, Figure 10 , the simulation result peak-to-average sidelobe ratio is about 17.02dB, in Figure 14 , N=768 and m=12, and the calculated peak-to-average sidelobe ratio is 18.1dB, Figure 14 , the simulation result peak-to-average sidelobe ratio is about 19.53dB;

[0124] 5. The velocity spectrum is as shown in Figure 13 . According to the sub-band division data, in this embodiment, the total number of frequency bands is 8, the number of velocity dimension FFT points is 2304, and therefore each sub-band contains 288 velocity units. The data size after division is 128x288x8x48x12;

[0125] 6. The divided data is post-3D non-coherent accumulation. Since the information of the target is not contained in the null band, the peak of the target will not be improved during the non-coherent accumulation, that is, too many null bands should not be set. The 2D spectrum after the non-coherent accumulation is subjected to 2dCFAR to obtain the r, v indexes of the target as [104, 67];

[0126] 7. The target is subjected to channel separation. According to the target index, the data at the corresponding position in the sub-band is sequentially taken out, and the speed index number of the target in each sub-band is sequentially calculated as [67, 355, 643, 931, 1219, 1507, 1795, 2083] according to the sub-band width. The template [1 1 1 1 0 0 0 0] is used to cyclically correlate with the above sequence, and the correlation peak is obtained at k=4. Therefore, the index number 931 corresponds to the Tx numbered 1 in the group, and the peak positions corresponding to the Txs numbered 2, 3 and 4 are sequentially obtained as 1219, 1507 and 1795.

[0127] 8. The channel data is rearranged according to the position of the virtual true element as shown in Figure 9 , and the spatial dimension 2DFFT is performed after windowing. The spatial spectrum result is shown in Figure 13 .

[0128] 9. The spatial spectrum shown in Figure 13 is subjected to 2dCFAR detection to obtain the azimuth angle and the elevation angle information of the target. The steps 7-9 are repeated to complete the angle detection of all targets on the range-speed spectrum.

[0129] Of course, the above description is not a limitation of the present application, and the present application is not limited to the above examples. Changes, modifications, additions or replacements made by ordinary skilled in the art within the essential scope of the present application also belong to the protection scope of the present application.

Claims

1. A radar waveform and signal processing method combining TDM and DDM, comprising the following steps of combining TDM phase and DDM phase: S1, divide all M transmit antennas to be allocated into M Group groups, where M can be divided by M Group integers. S2, determine the multiplexing number of each group of transmissions and determine M Group determining the initial phase of each transmitting antenna in each time slot, and transmitting signals according to the determined initial phase of the transmission, receiving the echo signals; S3, mixing the echo signal with the reference signal and sampling, and performing distance dimension FFT on the sampled signal; S4, performing 2D FFT on the slow time according to each group of transmission sequences determined in S2, and dividing the obtained results according to Doppler sub-bands; S5, performing non-coherent accumulation on the Doppler sub-band dimension, the transmission group dimension and the receiving channel dimension; S6, performing CFAR detection on the non-coherent accumulation result, and taking out the 2D FFT data in all sub-bands corresponding to each target; S7, performing cyclic correlation on the target data and the template data, wherein the channel separation cyclic correlation method is adopted, the displacement number corresponding to the maximum value is recorded, the position corresponding to the first transmission antenna is recorded, and the remaining transmission channel data is taken out in turn; S8, rearranging all channel data according to the virtual channel position, performing angle dimension FFT on the rearranged data, and obtaining target spatial spectrum; S9, performing CFAR detection on the spatial spectrum, and obtaining the angle information of the target; S10, outputting the distance, speed, angle and signal-to-noise ratio information of the target.

2. The radar waveform and signal processing method of claim 1, comprising a modulated signal transmitting module, a return signal receiving module and a return signal processing module, wherein, The modulated signal transmission module is used for modulated signal generation and signal delivery. The echo signal receiving module is used for receiving the signal delivered by the modulated signal transmission module. The echo signal processing module is used for performing distance dimension, speed dimension and angle dimension FFT on the echo signal.

3. The method of claim 2, wherein: the TDM and DDM combined radar waveform and signal processing method further comprises: transmitting a radar signal using a first radar waveform; receiving a radar signal using a second radar waveform; and transmitting a radar signal using a third radar waveform. The working process of the modulated signal transmission module is as follows: (1) modulating the signal and generating a random Tx transmission sequence; (2) determining the transmission sub-array according to the current pulse index, and calculating the modulation phase in this transmission according to the sub-array information; (3) generating a chirp signal and changing the initial phase through a phase shifter; (4) selecting the corresponding Tx group and transmitting the corresponding modulated signal.

4. The radar waveform and signal processing method combining TDM and DDM according to claim 2, characterized in that: The echo signal processing module includes a transmission grouping step and a transmission sequencing step, and the working process is as follows: (1) rearranging the data according to the transmission sequence, zero padding the rearranged data and performing speed dimension FFT; (2) grouping the speed dimension according to the number of sub-bands, and performing non-coherent accumulation on the channel dimension and the sub-band dimension; (3) separating the transmission channels according to the target, rearranging the channel data according to the virtual channel position relationship, and obtaining the target spatial spectrum; (4) performing CFAR detection on the spatial spectrum, obtaining the target information, and delivering the target information.

5. The method of claim 4, wherein: The transmission grouping method is as follows: ​ (1) Determine the number of groups M to be split Group , and M Group can be evenly divided by the total number of Tx M, that is, each group contains Tx; (2) determining the transmission Tx number in each group according to non-replacement extraction in turn.

6. The method of claim 4, wherein: The transmission sequencing method is as follows: ​ (1) Determine the multiplexing number N of each group of Tx, and the total number of chirp is M Group ×N, and all Tx groups complete one transmission, which is called a Burst, that is, N Bursts are transmitted in total; (2) Determine the transmission order within each Burst, that is, generate a 1 ~ M Group random sequence, and the transmission order of all Bursts constitutes a 1 × NM Group array, recording the transmission Tx group number of each chirp; (3) determining the transmission time of each Tx group according to the sequencing of each Burst.

7. The radar waveform and signal processing method combining TDM and DDM according to claim 1, characterized in that: The slow time 2D FFT method is as follows: (1) According to the transmitting sequence, one Tx group corresponding N echo data is selected from all M Group N echo data. (2) Fill N echo data into corresponding transmit sequence according to the transmit sequence, and fill 0 in the position where the Tx group does not transmit, to construct a 1xNM Group data matrix; (3) Perform an NM Group point FFT on the data matrix.

8. The method of claim 1, wherein: The channel separation cyclic correlation method is as follows: ​ (1) According to the detected target index, the values are taken out at the corresponding positions of all Doppler sub-bands, an array with a size of k band is constructed according to the sub-band positions, and a circular correlation template is constructed; (2) calculating the cyclic correlation result of the template sequence and the data array; (3) finding the peak value in the correlation result, which is the Tx peak value position of the transmission number 1, and calculating the peak value positions of the remaining m-1 Txs according to the Doppler sub-band width.

Images