CN113325384A

CN113325384A - Communication radar joint processing method

Info

Publication number: CN113325384A
Application number: CN202110888586.XA
Authority: CN
Inventors: 吴浩; 鞠宏浩; 方旭明; 龙彦; 何蓉
Original assignee: Southwest Jiaotong University
Current assignee: Southwest Jiaotong University
Priority date: 2021-08-04
Filing date: 2021-08-04
Publication date: 2021-08-31
Anticipated expiration: 2041-08-04
Also published as: CN113325384B

Abstract

The invention discloses a communication radar joint processing method, which comprises the following steps: firstly, the transmitting terminal blanks the data field part in the communication frame, and designs the pulse radar signal by taking the preamble as a phase coding sequence. The receiving end segments the echo signals in the distance dimension, designs pulse pressure filter coefficients corresponding to echo data of all segments according to the characteristics of the transmitted signal coding sequence, and inhibits pulse pressure side lobes through segmented pulse pressure; and then improving the CLEAN algorithm, and combining the two-dimensional FGOS-CA CFAR algorithm to realize the one-by-one extraction and detection of targets from the pulse pressure echoes. The invention effectively solves the problem of pulse pressure sidelobe caused by poor non-periodic autocorrelation performance of the transmitting sequence, improves the robustness of the system, simultaneously has no obvious increase of the calculation complexity of the scheme compared with the traditional radar signal processing algorithm, can be used for various lead code sequences, and has universal applicability.

Description

Communication radar joint processing method

Technical Field

The invention belongs to the technical field of communication radar integration, and particularly relates to a communication radar joint processing method.

Background

A Joint Communication-Radar (Joint Communication-Radar) system realizes integration of Radar and Communication functions through hardware multiplexing and waveform sharing, realizes full digital domain baseband processing of echo signals by utilizing hardware such as a high-speed ADC (analog to digital converter) and the like of the Communication system, and simultaneously carries out target detection and parameter estimation based on the traditional Radar signal processing technology.

At present, the JCR system is mainly applied to a commercial WLAN frequency band, is different from a traditional narrow-band radar signal, and has the advantages of high data transmission rate, high distance resolution and the like because the available frequency bands in a commercial authorized frequency band and a millimeter wave unauthorized frequency band are wider. Documents [ p, Kumari, j, Choi, n, Gonz-lez-precic and r.w. heat, "IEEE 802.11ad-Based Radar: An Approach to Joint vehicle Communication-Radar System," in IEEE Transactions on vehicle Technology, vol 67, No. 4, pp. 3012-3027, April 2018] propose a special time-frequency structure using IEEE802.11ad single carrier physical layer frame data packet, and design a single-frame and multi-frame processing strategy Based on the conventional synchronization and frequency offset estimation algorithm, and after the transceiving end beam is aligned and a directional link is established, extract parameter information such as target distance, speed, etc. from the echo reflected by the target. The scheme assumes that a full-duplex radar is realized by a self-interference cancellation scheme, but the current digital domain self-interference cancellation scheme is difficult to realize ideal self-interference cancellation and has extremely high computational complexity. Meanwhile, the residual self-interference after cancellation can greatly increase the background noise of the radar receiver. The invention considers taking IEEE802.11 protocol as a basic frame, utilizes a typical lead code structure thereof, realizes the receiving and transmitting isolation by configuring a communication frame of a transmitting terminal into a phase coding pulse radar signal form, and designs a multi-target detection algorithm to realize the speed measurement and the distance measurement of a target.

Pulse compression is one of the key technologies of pulse radar, and not only maintains the high distance resolution of narrow pulses, but also can obtain the high signal-to-noise ratio of a wide pulse radar system. Wherein, the non-periodic autocorrelation of the transmitted signal sequence can largely determine the detection accuracy of the system. Because the preamble sequence in the physical layer frame structure of the IEEE802.11 protocol is mainly designed for timing and synchronization of a communication receiving end, the JCR system based on the preamble has the problem that the main side lobe of the ambiguity function distance dimension of the pulse signal is relatively low. When the pulse pressure side lobe value of the echo signal is too high, the CFAR module can cover the main lobe of a nearby weak target by a strong target side lobe to cause a false alarm condition in a multi-target scene, and the system can also cause overload due to too high actual false alarm rate to cause target misjudgment. Therefore, it is necessary to adopt a corresponding sidelobe suppression algorithm to solve the problems of strong and weak, adjacent target shading, and the like. On the other hand, when the target is too close, the received echo is incomplete, and if the full code is continuously adopted for pulse compression, grating lobes appear in an output result, so that the target detection accuracy is reduced.

In order to suppress range side lobes of pulse pressure output, researchers have proposed a side lobe suppression technique based on the CLEAN algorithm. The CLEAN algorithm is a deconvolution technology for removing single-frequency components and simultaneously removing signal side lobes, and the main idea is to remove the distance side lobes of strong scatterers in sequence so as to uncover the covered weak scatterers. The computation time cost of the CLEAN algorithm depends on the target number. Eliminating target pulse pressure output side lobes by using a CLEAN algorithm, searching a strongest target for multiple times, reconstructing ideal matched filtering output of the strongest target according to corresponding frequency point information, cancelling the ideal matched filtering output with echo data, and finally detecting target information.

Although the CLEAN algorithm is computationally efficient, it cannot suppress the side lobes to noise levels, and its performance drops significantly when the targets are close to each other. On the other hand, the clear threshold p0 has no explicit solution, and if the clear processing is directly performed on the pulse pressure output of the echo, the false alarm rate and the false alarm rate of the system may increase due to improper setting of the threshold. Meanwhile, the technology is mainly applied to conventional phase coding radar signals, and has certain requirements on the non-periodic autocorrelation performance of a signal sequence, so that the algorithm has certain limitations when facing a JCR system based on an IEEE802.11 protocol.

Table 1 gives the definitions of the abbreviations and chinese and english in this patent:

TABLE 1 abbreviations and Key term definitions

Disclosure of Invention

In order to solve the technical problems of missing detection and error detection of a CFAR (circulating fluid dynamics) detection machine, poor robustness of a CLEAN algorithm and the like caused by high pulse pressure side lobe value when the existing JCR system based on the full-duplex hypothesis adopts the target detection scheme, the invention provides a communication radar joint processing method.

The invention discloses a communication radar combined processing method, which comprises the following steps:

step 1: and performing phase coding modulation on a preamble sequence in the IEEE802.11 protocol communication frame, and meanwhile, emptying the data field part in the frame, thereby constructing and transmitting a pulse radar signal.

Step 2: the acquired radar echo data is segmented into two parts, including a front occlusion part and an non-occlusion part.

And step 3: corresponding pulse pressure filter coefficients are designed for each section of echo data, and output is spliced after segmented pulse pressure.

And 4, step 4: and sending the pulse pressure output to a two-dimensional FGOS-CA CFAR module for strong target detection after MTD moving target detection to obtain target information.

And 5: and (4) inversing an ideal echo signal corresponding to the strong target by utilizing a CLEAN algorithm, and removing the ideal echo signal from the original pulse pressure output.

Step 6: and the residual pulse pressure echoes are subjected to MTD processing again and sent to a CFAR module for successive detection.

And 7: and repeating the steps 4-5 until all the targets are detected.

Further, step 1 specifically comprises:

the data field portion of the communication frame is first zeroed out, followed by the remaining preamble sequenceC(k) BPSK or MPSK modulation is carried out, a baseband modulation scheme is determined by a communication protocol physical layer, and then single carrier modulation or OFDM modulation is carried out according to a carrier modulation scheme adopted by a JCR system, so that a pulse radio frequency signal is obtained. The transmitted phase encoded pulse string baseband signal is of the form:

wherein K is the length of the pulse code sequence, N is the number of pulses, M is the number of phases which can be used by the modulation scheme,

represents the complex envelope of the nth pulse,

indicating the modulation phase of the k-th sub-pulse within each pulse,T _PRIin order to be the pulse repetition interval,tas a matter of time, the time is,

。

further, step 2 specifically comprises:

the radar receiver receives N pulse echoes according to a set distance window, and a radio frequency RF module outputs baseband sampling signals of the pulse echoesx(n) Will bex(n) Divided into two parts in the distance dimensionx ₁(n) Andx ₂(n) Respectively corresponding to the non-shielding condition and the front shielding condition, wherein the corresponding distance segments of each part are as follows:

no shielding:

front shielding:

wherein R is the target distance,T _dfor the duration of the sub-pulses,cis the speed of light.

Further, step 3 specifically comprises:

step 3.1: initializing system parameters to bePreamble sequencesC(k) The STF in (1) is divided into two parts including a repeated sequence partC _stf1And non-repetitive sequence portionsC _stf2(ii) a Will be provided withC _stf1Obtaining the cross-correlation value between the partial sequence and each sequence in the pseudo-random sequence set S through a cross-correlation selectorR _c(k) If the length of each sequence is L, the form is as follows:

wherein S = { m sequence, Gold sequence, Kasami sequence, Oppermann sequence },l=0,1,2,…,L-1。

step 3.2: comparing the peak side lobe ratio PSLR of the cross-correlation values:

whereinP _sThe peak value is the highest peak value,P _mthe highest sidelobe value.

Then selecting a sequence S corresponding to the PSLR highest value from the pseudo-random sequence set, and matching the coefficient of the pulse pressure filterh _cPart of the repetitive sequence of (1)C _stf1Replacing the pseudo-random sequence with the same length to obtain the mismatched pulse pressure filterh1。

Step 3.3: will be provided withC(k) The LTF coding sequence in the filter is used as a filter coefficient to obtain a mismatched pulse pressure filterh2。

Step 3.4: will be provided withx ₁(n) Andx ₂(n) Respectively passing through mismatched pulse pressure filtersh1 andh2, splicing the results according to the distance dimension to obtain pulse pressure outputy(n)。

Further, step 4 specifically includes:

step 4.1: performing coherent accumulation on N groups of pulse pressure output data to obtain RDM matrix of range-Doppler dimension

In which N is_FFTAnd the number of FFT points adopted in the phase-coherent accumulation process is represented, U is the number of single-pulse sampling points, and the RDM matrix is sent into a two-dimensional FGOS-CACFAR module after square rate detection.

Step 4.2: the module firstly processes from a fast time dimension, namely a distance dimension, of an RDM matrix, and carries out Doppler unit processing on each row of data in a two-dimensional reference windowiCorresponding toU _iA reference distance unit

By a one-dimensional OSCAGOCFAR detector in which Q reference cells employ the CACFAR algorithm, and the remainderU _i-Q reference cells are then processed using the OSCFAR algorithm and only cells whose internal coordinates are odd are sorted; the values calculated by the OSCFAR and CACFAR portions are then compared and the maximum value is selected as the estimate of the noise and clutter power in the Doppler cell

(ii) a After all Doppler units in the two-dimensional reference window are subjected to the same processing, a one-dimensional column vector is obtained

，

Representing a transpose operator; finally, the obtained vector is processed by CACFAR, namely the arithmetic mean value of the vector elements is calculated so as to obtain the average power level of the background noise and the clutter

The calculation formula is as follows:

wherein the content of the first and second substances,q is the number of reference units processed using the CACFAR algorithm,

as a Doppler unitiThe corresponding qth reference range unit,

is left overU _i-k-th ordered statistic of Q reference units processed by OSCFAR algorithm.

The decision threshold is formed by a threshold multiplication factor

Threshold offset

And

jointly determining:

wherein the content of the first and second substances,

related to false alarm probability:

，Y ₀the event that clutter and noise only exist in the unit to be detected is represented, so that the event can be obtained through calculation through a Monte Carlo experiment; is provided withY ₀The method aims to reduce the influence of using a window function and limited reference unit quantity, and improve the accuracy of detecting a single target in each subsequent iteration, wherein the value range is 1.27-1.52.

Step 4.3: carrying out target judgment on the RDM matrix according to the threshold value obtained by calculation, selecting the point with the highest peak value as an output target from the RDM matrix if a plurality of targets are detected, and obtaining the value of the coordinate value of the pointTo target parameter estimation

。

Further, step 5 specifically comprises:

step 5.1: constructing a theoretical echo signal corresponding to the target according to the strong target parameters detected by the CFAR module:

wherein the content of the first and second substances,

are respectively the targetpCorresponding to the amplitude, round trip delay and doppler shift of the theoretical echo signal, P is the target number,s(t) is the baseband waveform of the transmitted signal,Pis the target number.

Step 5.2: to pair

Digital sampling and segmentation are performed, and then the sampled data is passed through a mismatched pulse pressure filterh1 andh2 obtaining the targetpCorresponding ideal echo pulse pressure output

。

Step 5.3: from pulse pressure outputy(n) Minus

The result is used as input data for the next iteration.

Furthermore, when the pulse pressure filter is applied to a communication radar integrated system based on an IEEE802.11ad protocol, the matched pulse pressure filter coefficient is selectedh _cThe STF sequence in (1) is replaced by m sequence with the same lengthq ₂₀₄₈Thereby obtaining the mismatched pulse pressure filter in the non-shielding distance sectionh1; meanwhile, the CEF coding sequence is used as a filter coefficient to obtain a mismatched pulse pressure filterh2; mismatched pulse pressure filterh1 andh2 the construction is as follows:

lead code

；

；

。

The beneficial technical effects of the invention are as follows:

1. the communication radar integrated system based on the preamble of the PHY physical layer of the IEEE802.11 protocol can multiplex the existing radio frequency hardware, and can realize the updating of the multi-target detection function on the basis of the data transmission of the original communication transceiver by carrying out phase coding on the preamble sequence and emptying the data part in the communication frame.

2. The invention provides a method for combining a constant false alarm algorithm and a CLEAN algorithm, which can ensure that the previously detected target is completely dug out (including information such as modulation mode, amplitude, phase and the like) in the original echo on the premise of not losing any useful target information, and the target-by-target searching and elimination are realized through an iterative algorithm, thereby greatly improving the accuracy of multi-target detection.

3. Aiming at the problem that the side lobe of a fuzzy function of a traditional lead code sequence is higher, a part of pulse pressure sequences with poor autocorrelation performance are replaced by pseudo-random sequences to design a mismatch filter, namely, the correlation between the part of sequences and the corresponding part of echo data is eliminated to improve the main-side lobe ratio of pulse pressure output. Meanwhile, the grating lobe caused by incomplete echo signals due to the fact that the targets are too close is reduced by the segmented pulse pressure method.

4. The method provided by the invention is not only suitable for a low data rate JCR system, but also suitable for a high speed JCR system; compared with the traditional radar signal processing algorithm, the scheme obviously improves the multi-target detection rate of the radar communication integrated system and has very high application value.

Drawings

Fig. 1 is a schematic diagram of an integrated system transmission signal based on an IEEE802.11 protocol preamble.

FIG. 2 is a schematic diagram of a reference sliding window of a two-dimensional FGOS-CA CFAR module.

Fig. 3 is a diagram of an IEEE802.11ad SC PHY preamble frame structure.

FIG. 4 is a complete flow chart of the multi-target detection method of the present invention.

Fig. 5-6 are graphs comparing the pulse pressure output of the first pulse echo (where fig. 5 is the matched pulse pressure filter output and fig. 6 is the mismatched pulse pressure filter output).

Fig. 7 is a diagram of a first iteration processing result of the communication radar joint processing method.

Fig. 8 is a diagram of a second iteration processing result of the communication radar joint processing method.

Fig. 9 is a diagram of a third iteration processing result of the communication radar joint processing method.

Fig. 10 is a diagram of a fourth iteration processing result of the communication radar joint processing method.

Fig. 11 is a diagram of a fifth iteration processing result of the communication radar joint processing method.

FIG. 12 is a graph of the two-dimensional FGOS-CA CFAR algorithm detection results.

Detailed Description

The invention is described in further detail below with reference to the figures and the detailed description.

The application object of the invention is a radar communication integrated system based on IEEE802.11 protocol framework, namely, a lead code in a communication frame is taken as a phase coding sequence and the single-base-station radar system works in a time division duplex mode. As shown in fig. 1, the system achieves radar sounding only during communication idle times by partially blanking out the payload data.

The preamble contains a short training sequence STF mainly used for frame synchronization and coarse frequency correction and a long training sequence LTF mainly used for fine frequency correction and channel estimation. The repetition period of the sequence group consisting of STF and LTFSimilarly, for example, the L-LTF sequence in the 802.11a/g physical layer is composed of two sets of identical sequences T1 and T2 and a guard interval GI, and the L-STF is composed of 10 sets of repeated sequences P, so that the whole preamble has poor aperiodic autocorrelation performance, i.e., the peak side lobe is lower than the PSLR. In order to solve the problem of sidelobe masking of the JRC system in the multi-target environment, the coefficient of the matched pulse pressure filter needs to be matchedh _cThe improvement is carried out to reduce pulse pressure sidelobe. Considering that the number of the STF internal repeating sequence groups is large, the autocorrelation function of the STF internal repeating sequence groups has a plurality of groups of repeating peaks with the same height, which greatly affects the accuracy of the CFAR module target detection. Therefore, the invention willh _cThe STF part in the filter is replaced by a pseudo-random sequence to obtain a mismatched pulse pressure filterh2, thereby suppressing fixed position side lobes in the echo pulse pressure result.

The invention improves the CLEAN algorithm aiming at the preamble sequence adopted by the JCR system transmitting signal and aims to improve the target detection accuracy of the constant false alarm CFAR module in the multi-target environment in the traditional radar detection process. As the algorithm needs to extract and eliminate the targets in the echoes one by one, and the CFAR detection machine is still adopted in the process of extracting the targets, the overall calculation complexity of the algorithm is higher. To this end, the invention proposes a two-dimensional FGOS-CACFAR module, i.e. a one-dimensional OSCAGOCFAR detector in the range dimension and a CACFAR in the Doppler dimension. Meanwhile, in order to further reduce the calculation amount of the algorithm, the one-dimensional OSCAGOCFAR only orders the Doppler unit sample points of the odd columns, and the schematic diagram of the algorithm is shown in FIG. 2, wherein "

"denotes a reference cell processed with OSCFAR"

"denotes a reference unit processed by CACFAR"

"represents a protection unit"

"indicates the unit under investigation.

Example 1:

the application object of embodiment 1 of the present invention is a JCR system based on the IEEE802.11ad protocol, and the transmitted two-phase coded radar signal uses a preamble in an IEEE802.11ad SC PHY frame as a binary coding sequence.

The IEEE802.11ad SC PHY preamble consists of two parts (as shown in fig. 3) of STF and CEF, where STF consists of 16 repeated sets of Golay one-sided sequences, Ga128, and suffix-Ga 128, and CEF field consists of two sets of Golay complementary sequence pairs, Gu512, Gv512, and suffix-Gb 128.

Each Golay complementary sequence pair consists of Ga and Gb. Since the autocorrelation of the complementary sequence is completely opposite in amplitude and phase at the non-zero shift, the sidelobe levels are completely cancelled and the peak value is doubled when the entire sequence pair is subjected to autocorrelation processing, so that the autocorrelation function of the complementary sequence has the characteristic of zero sidelobe level.

The STF part in the IEEE802.11ad lead code consists of repeated Ga, does not have the characteristics of the complementary sequence pair, and has rich sidelobe of a non-periodic autocorrelation function and low main sidelobe ratio. Under the influence, the CFAR detection module of the system is very easy to miss detection and error detection under the environment that a plurality of targets are adjacent to each other.

Therefore, the pulse pressure sidelobe is reduced by designing the mismatched filter, the clear algorithm is improved to eliminate the target masking effect, and the two-dimensional FGOS-CA CFAR module is constructed to further reduce the calculation complexity of the clear algorithm.

Fig. 4 shows the complete signal processing flow of the present embodiment. The mismatch filter is designed according to a typical preamble adopted by the transmitting end of the JCR system. After the radar receiver obtains radar echo data, mismatch filtering processing is carried out to obtain pulse pressure output. And then the pulse pressure data is subjected to MTD moving target detection to obtain an RDM matrix, and the RDM matrix is sent to a constructed two-dimensional FGOS-CACFA module for strong target detection to obtain single target information. Then, the ideal echo signal corresponding to the strong target is inverted by using the CLEAN algorithm and is eliminated from the original pulse pressure output. And finally, performing MTD processing on the residual pulse pressure echoes again and sending the residual pulse pressure echoes to a CFAR module for repeated detection until all targets are detected.

Example 1 main parameters of the system: the system adopts a single-base-station radar model and has working frequency range

Width of pulse

Duty ratio D =10%, range resolution is

，cIs the speed of light. The length of the pulse code sequence is N =3328, and the radar pulse width is

. Pulse repetition interval

. The system adopts a monostatic model, namely the radar receiver is turned on only after the pulse transmission is finished. Receiver sampling frequency

。

Let 5 targets independent of each other exist in the space, and the distance, speed, RCS of the 5 targets are set to [420m,5m/s,40 respectivelym ²]、[425m,10m/s,50m ²]、[435m,20m/s,70m ²]、[650m,8m/s,90m ²]、[670m,25m/s,100m ²]。

The radar transmits 16 pulses in sequence, and the distance between each pulse signal transmitted is

Relative velocity of

Object of (2)pAnd (3) reflecting, finally reaching a receiver and performing down-conversion treatment to obtain the form as follows:

wherein the content of the first and second substances,

representing transmitter, receiver gain and free space channel attenuation, respectively.G _pRepresenting objectspThe gain in the reflection of the signal is,f _cis the frequency of the carrier signal and,

respectively round trip delay and doppler shift.

The STF is processed by a cross correlator to obtain the cross correlation value of the sequence and each sequence in the pseudo-random sequence set SR(k) Selecting to replace the STF sequence in the matched filter coefficient with the m sequence with the same length by comparing the peak side lobe ratio PSLR of the cross-correlation valueq ₂₀₄₈Thereby obtaining the mismatched filter coefficients in the non-shielding distance sectionh1。

Will be provided withC(k) The CEF code sequence in (a) is obtained as a filter coefficienth2. When the target is located in the front occlusion range, the echo pulse can only receive part of the STF, and by directly setting the pulse pressure filter in the range to the CEF sequence, the grating lobe caused by STF cross-correlation with the CEF can be eliminated as much as possible.

Table 2 shows a JCR system pulse pressure filter construction method based on IEEE802.11ad preamble:

TABLE 2 preamble sequence of Length 3328 and constructed segmented pulse pressure Filter coefficients

The radar receiver receives echo signals in sequence according to a set distance window, and baseband sampling signals of each pulse are obtained by sampling after passing through RF devices such as an LNA (low noise amplifier), a mixer and the likex(n) Then, the pulse pressure is segmented and the results are spliced to obtain the pulse pressure outputy(n) At the same time, for comparison, will alsox(n) Obtaining an output through a matched filterg(n). The pulse pressure output comparison results of the 1 st pulse echo signal are shown in fig. 5 and 6.

Combining 16 sets of pulse pressure data into a matrix

Where each row of the matrix represents 32768 sample points of a single pulse echo. Coherent accumulation of all echo pulse pressure data, i.e. on a matrix

Each column of the RDM matrix is subjected to 256-point FFT to obtain the RDM matrix

. And then the matrix Y is subjected to square rate detection and then is sent to a two-dimensional FGOS-CACFAR module.

The two-dimensional FGOS-CACFAR module combines the unit average CFAR and the order statistics CFAR by matching Doppler units in a sliding reference windowiCorresponding toU _iA reference range unit sample

One-dimensional OSCAGO CFAR processing is carried out to obtain the estimated values of noise and clutter power in each Doppler unit

. Then the matrix is divided into

Calculated for each row in a sliding reference window

Combined into a one-dimensional vector

Then, the vector is processed by CA-CFAR, i.e. the arithmetic mean value of each sample element in the vector is calculated, thereby obtaining the average power level of the background noise plus the clutter

. Will be provided with

Multiplication by

The threshold T corresponding to the cell to be detected in the reference window is obtained.

Traversing each element of the RDM matrix by sliding the reference window from left to right and from top to bottom, and repeating the calculation to obtain a threshold matrix

. Finally, carrying out target discrimination on each element of the two matrixes, and if a plurality of targets are detected, selecting a point with the highest peak value as an output targetpAnd from the targetpObtaining the target parameter estimated value by the coordinate value of the point

(ii) a Then, an ideal original echo signal corresponding to the target is constructed according to the formula (2), the echo signal is output from the pulse pressure after passing through a mismatch filter to obtain a corresponding ideal echo pulse pressurey(n) And (4) subtracting.

And finally, the residual pulse pressure output of each pulse is subjected to MTD processing again and is sent to a CFAR module for successive detection until all targets are detected. The results of each iteration of the algorithm are shown in fig. 7-11.

Example 1 simulation analysis:

from the comparison results of fig. 5 and 6, it can be seen that compared with the output result of the matched pulse pressure filter using the original transmission sequence, the pulse pressure main-side lobe ratio can be improved by 10dB by replacing the STF in the filter coefficients with the m sequence with the same length. As can be seen from fig. 7-11, after the five rounds of algorithm processing are performed on the original pulse pressure output, 5 targets (RCS are also different) respectively set at different distances of [420, 425, 435, 520, 550] m are correctly detected one by one, and the ideal echo signals corresponding to the targets are also directly removed in the original echo in each iteration. On the other hand, the multi-target detection method combines and improves the two-dimensional FGOS-CACFA algorithm, thereby greatly improving the detection efficiency and accuracy of the adjacent target. From FIG. 12 it is clear that three close-proximity targets located at [420, 425, 435] m can be effectively identified by the two-dimensional FGOS-CACFAR algorithm.

Example 2:

embodiment 1 of the present invention is directed to a JCR system based on IEEE802.11ad, but may also be applied to other JCR systems based on IEEE802.11 series communication protocols. For example, when the JCR system based on the IEEE802.11 ax protocol adopts the scheme, the mismatch filter design may be performed on the Legacy preamble and the HE preamble in the HE-PPDU field, respectively, where the L-STF field may be replaced with an m-sequence of equal length corresponding to the coefficient part of the matched filter. On the other hand, considering that the HE-STF and HE-LTF fields are newly added training fields of the protocol, the pseudo-random sequence also needs to be selected according to the actual autocorrelation performance of each part of the sequence. Similarly, the system also adopts a target detection scheme based on a CLEAN algorithm at a receiving end, namely, the two-dimensional FGOS-CACFAR algorithm provided by the invention is used for carrying out target detection on the echo signal, a corresponding ideal pulse pressure signal is constructed based on the target parameter information, then the ideal pulse pressure signal is eliminated in the original echo pulse pressure signal, and then the residual echo pulse pressure is subjected to MTD processing and CFAR detection again until all targets are detected.

The invention designs a pulse radar communication integrated system by taking a preamble sequence in an IEEE802.11 protocol physical layer communication frame as a phase coding sequence. The invention provides a multi-target detection scheme based on CLEAN thought, which aims to realize target-by-target detection from echoes by designing an echo signal processing algorithm. Aiming at the problem that the pulse pressure sidelobe of a JCR system based on an IEEE802.11 protocol lead code is high, the scheme utilizes a pseudo-random sequence to reduce the influence of a repeating period sequence STF in a conventional lead code on the non-periodic autocorrelation performance of an overall coding sequence. Meanwhile, the problem of high target echo sidelobe in the shielding area is solved at a receiving end by utilizing segmented pulse pressure. The multi-target detection method provided by the invention combines the CLEAN algorithm and the two-dimensional FGOS-CA CFAR algorithm, and makes up for the defects of error detection, missing detection and the like easily occurring when the traditional CLEAN algorithm faces the situation of an adjacent target.

Claims

1. A communication radar joint processing method is characterized by comprising the following steps:

step 1: performing phase coding modulation on a lead code sequence in an IEEE802.11 protocol communication frame, and meanwhile, emptying a data field part in the frame so as to construct and transmit a pulse radar signal;

step 2: segmenting the acquired radar echo data into two parts including a front shielding part and a non-shielding part;

and step 3: designing corresponding pulse pressure filter coefficients for each section of echo data, and splicing the output after segmenting pulse pressure;

and 4, step 4: sending the pulse pressure output to a two-dimensional FGOS-CA CFAR module for strong target detection after MTD moving target detection to obtain target information;

and 5: inverting the ideal echo signal corresponding to the strong target by utilizing a CLEAN algorithm, and removing the ideal echo signal from the original pulse pressure output;

step 6: performing MTD processing on the residual pulse pressure echoes again and sending the residual pulse pressure echoes to a CFAR module for successive detection;

and 7: and repeating the steps 4-5 until all the targets are detected.

2. The communication radar joint processing method according to claim 1, wherein the step 1 specifically comprises:

the data field portion of the communication frame is first zeroed out, followed by the remaining preamble sequenceC(k) BPSK or MPSK modulation is carried out, the base band modulation scheme is determined by the physical layer of the communication protocol, then single carrier modulation or OFDM modulation is carried out according to the carrier modulation scheme adopted by the JCR system, thereby obtaining the pulse radio frequency signal, and the pulse radio frequency signal is transmittedThe phase encoded pulse train baseband signal form of (a) is as follows:

，

represents the complex envelope of the nth pulse,

。

3. the communication radar joint processing method according to claim 2, wherein the step 2 specifically comprises:

no shielding:

front shielding:

4. The communication radar joint processing method according to claim 3, wherein the step 3 specifically comprises:

step 3.1: initializing system parameters and encoding the preamble sequenceC(k) The STF in (1) is divided into two parts including a repeated sequence partC _stf1And non-repetitive sequence portionsC _stf2(ii) a Will be provided withC _stf1Obtaining the cross-correlation value between the partial sequence and each sequence in the pseudo-random sequence set S through a cross-correlation selectorR _c(k) If the length of each sequence is L, the form is as follows:

wherein S = { m sequence, Gold sequence, Kasami sequence, Oppermann sequence },l=0,1,2,…,L-1；

whereinP _sThe peak value is the highest peak value,P _mthe highest sidelobe value;

then selecting a sequence S corresponding to the PSLR highest value from the pseudo-random sequence set, and matching the coefficient of the pulse pressure filterh _cPart of the repetitive sequence of (1)C _stf1Replacing the pseudo-random sequence with the same length to obtain the mismatched pulse pressure filterh1；

Step 3.3: will be provided withC(k) The LTF code sequence in (1) is used as a filter coefficient to obtain mismatchingPulse pressure filterh2；

5. The communication radar joint processing method according to claim 4, wherein the step 4 specifically comprises:

In which N is_FFTThe FFT point number adopted in the phase-coherent accumulation process is represented, U is a single-pulse sampling point number, and the RDM matrix is sent into a two-dimensional FGOS-CACFAR module after square rate detection;

step 4.2: the module firstly processes from the fast time dimension, namely the distance dimension, of the RDM matrix and carries out Doppler unit processing on each row in a two-dimensional reference windowiCorresponding toU _iA reference distance unit

，

The calculation formula is as follows:

wherein Q is the number of reference units processed by the CACFAR algorithm,

as a Doppler unitiThe corresponding qth reference range unit,

is left overU _i-k ordered statistics obtained by processing Q reference cells by OSCFAR algorithm;

the decision threshold T is a threshold multiplication factor

Threshold offset

And

jointly determining:

wherein the content of the first and second substances,

related to false alarm probability:

，Y ₀the event that clutter and noise only exist in the unit to be detected is represented, so that the event can be obtained through calculation through a Monte Carlo experiment; is provided withY ₀The method aims to reduce the influence of using a window function and limited reference unit quantity, and simultaneously improve the accuracy of detecting a single target in each subsequent iteration, wherein the value range is 1.27-1.52;

step 4.3: carrying out target judgment on the RDM matrix according to the threshold value obtained by calculation, selecting the point with the highest peak value as an output target from the RDM matrix if a plurality of targets are detected, and obtaining a target parameter estimation value according to the coordinate value of the point

。

6. The communication radar joint processing method according to claim 5, wherein the step 5 specifically comprises:

wherein the content of the first and second substances,

are respectively the targetpCorresponding to the amplitude, the round trip delay and the Doppler frequency shift of the theoretical echo signal, wherein P is the target number;s(t) is a transmit signal baseband waveform;

step 5.2: to pair

；

Step 5.3: from pulse pressure outputy(n) Minus

The result is used as input data for the next iteration.

7. The joint processing method for communication radar as claimed in claim 4, wherein the pulse pressure filter design method is applied to a communication radar integrated system based on IEEE802.11ad protocol, and the matched pulse pressure filter coefficients are selectedh _cThe STF sequence in (1) is replaced by m sequence with the same lengthq ₂₀₄₈Thereby obtaining the mismatched pulse pressure filter in the non-shielding distance sectionh1; meanwhile, the CEF coding sequence is used as a filter coefficient to obtain a mismatched pulse pressure filterh2; mismatched pulse pressure filterh1 andh2 the construction is as follows:

lead code

；

；

。