A method for frequency domain multiplexing pulse compression of broadband multi-carrier signals

Abstract

The present invention discloses a frequency domain multiplexing pulse compression method for broadband multi-carrier signals, comprising: performing multi-phase extraction filtering on broadband multi-carrier signals to obtain baseband signals of echo signals; moving each sub-carrier signal to zero frequency, and performing smooth filtering through a designed sliding average filter to output echo data of each sub-carrier; performing a pulse compression on the output of the sliding average filter to obtain a range image of the sub-carrier; performing a sum operation on the sub-carrier range image to obtain a secondary pulse compression result, which is a high-resolution range image of the echo signal. The present invention realizes low resolution of the target by performing a primary pulse compression on the sub-carrier based on sub-carrier separation, and realizes high resolution of the target by performing secondary pulse compression processing on the sub-carrier pulse compression result. The digital pulse compression of the multi-carrier signal is realized by using orthogonal transformation based on multi-phase extraction filtering, CORDIC algorithm, and sliding average filter, and a rough range image and a fine range image of the target can be obtained.

Description

Technical Field

[0001] The invention belongs to the technical field of signal processing, and relates to a digital pulse compression technology for multi-carrier broadband signals, and in particular to a broadband multi-carrier signal frequency domain multiplexing pulse compression method. Background Art

[0002] Existing radars use waveforms modulated by a single carrier (such as frequency modulation or phase coded signals) for search, tracking, imaging and other modes, and different modes use waveforms with different resolutions; radar time resources are dynamically allocated between several modes according to time slots. In the case of large airspace search and multi-target tracking and guidance, the radar is easily saturated with multiple targets and loses important targets; from the perspective of low interception and anti-interference, waveforms modulated by a single carrier are easy to be intercepted and identified, easy to be interfered by noise aiming and lose target detection capabilities, and easy to be interfered by deception and unable to extract target points and maintain target tracking; from the perspective of target detection, the power and aperture resources of existing radars cannot be optimized and allocated according to the target distance. Multi-carrier phase coded (MCPC) signals (also known as radar OFDM signals) have the characteristics of both narrowband signals and broadband signals, and have the advantages of high spectrum utilization, frequency diversity, time diversity, broadband measurement, low intercept probability, and short-time Doppler frequency measurement. These advantages determine the huge potential of multi-carrier phase coded signals in future radars. Multi-carrier phase coded signals are synthetic waveforms modulated by multiple subcarrier multiphase codes. When used in radar, the problem of transmitting waveform modulation needs to be solved. It is different from the pulse compression of existing radar waveforms. First, the Doppler frequency destroys the orthogonality of the subcarriers, and the Doppler frequency estimation and calibration are required before pulse compression. Second, compared with the research on OFDM waveforms in the communication field, multi-carrier phase coded signals (later referred to as OFDM signals) have not made much progress in the research of radar. The main bottleneck lies in the pulse compression processing of multi-carrier signals. The traditional pulse compression method cannot effectively utilize the subcarrier signals because it directly matches the echo signal. Therefore, digital pulse compression of multi-carrier signals has become a bottleneck technology for radars to use multi-carrier signals. Summary of the invention

[0003] Purpose of the invention: In order to overcome the technical problems of digital pulse compression of radar multi-carrier broadband signals in the prior art, a method for frequency domain multiplexing pulse compression of broadband multi-carrier signals is provided. The low resolution of the target is achieved by primary pulse compression of sub-carriers based on sub-carrier separation, and secondary pulse compression is achieved by summing the sub-carrier pulse compression results to obtain a fine range image of the target. Different sub-carriers obtained by sub-carrier separation can be used for searching and tracking different targets. A single sub-carrier is used as a narrowband search signal, and a group of sub-carriers is used as a tracking signal. The low resolution result of the same target can be used for subsequent speed ambiguity resolution. The secondary pulse compression processing of the same target synthesizes broadband signals to achieve high resolution of the target. Based on orthogonal transformation, CORDIC algorithm, and sliding average filter of multi-phase extraction filtering, digital pulse compression of multi-carrier signals is achieved to obtain a coarse range image and a fine range image of the target.

[0004] Technical solution: To achieve the above object, the present invention provides a broadband multi-carrier signal frequency domain multiplexing pulse compression method, comprising the following steps:

[0005] S1: Perform polyphase extraction filtering on the broadband multi-carrier signal to obtain the baseband signal of the echo signal;

[0006] S2: Move each subcarrier signal to zero frequency, and smooth it through a designed sliding average filter to output the echo data of each subcarrier;

[0007] S3: Perform pulse compression on the output of the sliding average filter to obtain the range image of the subcarrier;

[0008] S4: Perform phase rotation on each subcarrier range image respectively, and then perform sum operation to obtain a secondary pulse pressure result, which is a high-resolution range image of the echo signal.

[0009] Furthermore, the baseband signal in step S1 is obtained by designing a multi-carrier broadband signal receiving module, performing intermediate frequency sampling on the echo signal, sending the sampled data to the FPGA, and using multi-phase extraction filtering technology to implement orthogonal processing of the intermediate frequency signal to obtain the baseband signal of the echo signal.

[0010] Furthermore, the step S1 specifically includes the following steps:

[0011] A1: Design a multi-carrier broadband signal receiving module, including a receiving antenna, a limiting module, a radio frequency bandpass filter (BPF0), a low noise amplifier (LNA), a mixer, an intermediate frequency bandpass filter (BPF1), an analog-to-digital conversion ADC device and an FPGA connected in sequence;

[0012] A2: Design a multi-carrier broadband signal waveform, where the frequency interval Δf between adjacent carriers is the symbol time width t bThe reciprocal of the multi-carrier wideband signal bandwidth of N carriers is equal to NΔf, and each carrier adopts phase coding modulation;

[0013] The coding phase in the present invention takes the P4 complementary code coding phase as an example, but it is applicable to all phase coding, and the pulse width is Mt b , the pulse repetition period is (N+1)Mt b , the effective echo time is NMt b ;

[0014] P4 complementary code encoding phase:

[0015]

[0016] In the formula, It represents the nth subcarrier and the mth symbol coding phase, n=0,1,…,N-1, m=0,1,…,M-1; N is the number of carriers, NM coding phases are generated by cyclic shift, and the number of symbols for each carrier is M.

[0017] A3: According to the bandpass sampling theorem, the intermediate frequency is 3NΔf, the sampling frequency is 4NΔf, the sampling data is sent to the FPGA for four-phase representation, and the four-phase signal is subjected to digital orthogonal processing based on multi-phase extraction filtering to obtain the baseband signal of the echo signal.

[0018] Furthermore, the step S2 specifically includes the following steps:

[0019] B1: Generate N subcarrier baseband negative frequency phases using N accumulators;

[0020] B2: Send the baseband signal of the echo signal and the subcarrier baseband negative frequency phase obtained in step B1 to the CORDIC module to move each subcarrier to zero frequency;

[0021] B3: Design a sliding average filter. For N subcarriers, the coefficients of the sliding average filter are N ones.

[0022] The subcarriers moved to zero frequency are filtered out by a sliding average filter. Since the sliding average filter only needs delay addition, it is easy to implement and can be realized by simple addition and shifting without consuming multiplier resources.

[0023] B4: Send the output of the CORDIC module in step B2 to the sliding average filter designed in step B3, and output the echo data of each subcarrier. Since the sampling frequency is NΔf, the number of data points in each subcarrier is NM.

[0024] Furthermore, the step S3 specifically includes: performing matched filtering on the sliding average filter output of the n'th subcarrier to obtain the range image of the subcarrier.

[0025] Furthermore, the step S4 is specifically as follows: sending the obtained N matched filter outputs to N CORDIC modules to eliminate the fixed phase change caused by delay on each subcarrier; summing the outputs obtained by the CORDIC modules to obtain a secondary pulse compression result, which is a high-resolution range image of the echo signal.

[0026] Furthermore, the result of performing digital orthogonal processing based on multi-phase extraction filtering on the four-phase signal in step A3 is expressed as:

[0027]

[0028] Where h[4i+i′] represents the i′th phase of the decimation filter, where I is an integer multiple of 4, i = 0, 1, …, I / 4-1, i′ = 0, 1, 2, 3, s R,IF [4k+k'] represents the k'th phase input of the decimation filter, k = 0, 1, 2, ..., N 2 M-1, k′=0,1,2,3. R,B [4k+k′] is the k′th phase output of the decimation filter.

[0029] Furthermore, the step B1 is specifically as follows:

[0030] The negative frequency phase of the n′th subcarrier can be expressed as:

[0031]

[0032] Where k = 0, 1, ..., N 2 M-1, n′=0,1,…,N-1.

[0033] Further, the expression of the primary pulse pressure method in step S3 is:

[0034]

[0035] In the formula, R 1,n′ [4k] is the frequency domain pulse compression output of the n′th subcarrier, k = 0, 1, ..., N 2 M-1,s R,MM,n' [4(k-m')] is the n'th subcarrier echo signal after smoothing filter, h MF1,n' [m'] is the matched filter coefficient of the n'th subcarrier and is the conjugate flip of the n'th subcarrier code phase.

[0036] The present invention provides a digital pulse compression method for realizing low resolution of a target through primary pulse compression of a subcarrier based on subcarrier separation and realizing high resolution of a target through secondary pulse compression processing of a subcarrier pulse compression result. The core technology of the method is that a digital orthogonal transformation based on multi-phase extraction filtering is adopted in FPGA, and an accumulator and a CORDIC algorithm are used to move an intermediate frequency signal to a baseband; a sliding average filter is designed to filter and output each subcarrier signal; a pulse compression of a subcarrier signal is realized by matching filtering of a code element of the same subcarrier with a subcarrier filtering output signal, so as to obtain a rough range image of the target; and secondary pulse compression is realized by summing up all subcarrier pulse compression results, so as to obtain a fine range image of the target.

[0037] The present invention utilizes different subcarriers obtained by subcarrier separation to search and track different targets. A single subcarrier is used as a narrowband search signal, and a group of subcarriers is used as a tracking signal. The low-resolution result of the same target can be used for subsequent velocity ambiguity resolution. Secondary pulse compression processing of the same target synthesizes broadband signals to achieve high resolution of the target. Based on orthogonal transformation of multi-phase extraction filtering, CORDIC algorithm, and sliding average filter, digital pulse compression of multi-carrier signals is achieved to obtain coarse range images and fine range images of the target.

[0038] Beneficial effects: Compared with the prior art, the present invention has the following advantages:

[0039] 1. The conversion from intermediate frequency signal to baseband signal is realized by using orthogonal transformation processing technology based on multi-phase extraction filtering.

[0040] 2. Use the accumulator and CORDIC algorithm to move each subcarrier to zero frequency.

[0041] 3. Use a sliding average filter to separate the subcarriers. Since the coefficient of the sliding average filter is 1, the implementation is very simple and only requires simple addition and shifting, saving a lot of multiplier resources.

[0042] 4. Taking advantage of the particularity of multi-carrier signals, only M points of the sliding average filter output are needed for one pulse compression, saving a lot of multiplier resources.

[0043] 5. The adder is used to sum the primary pulse compression results of each subcarrier to achieve secondary pulse compression, which can obtain a detailed range image of the target. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] Figure 1 It is a hardware receiving block diagram of a multi-carrier broadband signal echo signal of the present invention;

[0045] Figure 2 It is a hardware implementation block diagram of orthogonal transformation based on polyphase decimation filtering of the present invention;

[0046] Figure 3 It is a block diagram of the hardware implementation of the subcarrier separation processing of the present invention;

[0047] Figure 4 It is a hardware implementation block diagram of a pulse compression process of the present invention;

[0048] Figure 5 It is a hardware implementation block diagram of the secondary pulse compression processing of the present invention;

[0049] Figure 6 is the phase-frequency characteristic of the low-pass decimation filter designed by the present invention;

[0050] Figure 7 is the amplitude-frequency characteristic of the low-pass decimation filter designed by the present invention;

[0051] Figure 8 This is the spectrum diagram of the intermediate frequency signal obtained by AD sampling of the present invention

[0052] Fig. 9 is a spectrum diagram of the output signal of the polyphase decimation filter of the present invention;

[0053] Fig.10 It is a comparison diagram of the pulse compression output of the separated subcarrier and the ideal subcarrier signal under the single target of the present invention;

[0054] Fig.11 It is a comparison diagram of the pulse compression output of the secondary pulse compression under a single target of the present invention and the ideal subcarrier signal;

[0055] Fig.12 It is a comparison diagram of the ideal results of two adjacent targets and the output of a pulse compression of the method proposed by the present invention;

[0056] Fig.13 It is a comparison diagram of the ideal result of two adjacent targets and the secondary pulse compression output of the method proposed in the present invention;

[0057] Fig.14 It is a pulse compression fuzzy function diagram of the present invention;

[0058] Fig.15 It is the secondary pulse compression fuzzy function diagram of the present invention;

[0059] Fig.16 This is a graph showing the relationship between the maximum peak sidelobe ratio of the secondary pulse compression output of the present invention and the target speed. DETAILED DESCRIPTION

[0060] The present invention is further explained below in conjunction with the accompanying drawings and specific embodiments. It should be understood that these embodiments are only used to illustrate the present invention and are not used to limit the scope of the present invention. After reading the present invention, various equivalent forms of modifications to the present invention by those skilled in the art all fall within the scope defined by the claims attached to this application.

[0061] The present invention provides a broadband multi-carrier signal frequency domain multiplexing pulse compression method, comprising the following steps:

[0062] S1: Perform polyphase extraction filtering on the broadband multi-carrier signal to obtain the baseband signal of the echo signal;

[0063] S2: Move each subcarrier signal to zero frequency, and smooth it through a designed sliding average filter to output the echo data of each subcarrier;

[0064] S3: Perform pulse compression on the output of the sliding average filter to obtain the range image of the subcarrier;

[0065] S4: Perform phase rotation on each subcarrier range image respectively, and then perform sum operation to obtain a secondary pulse pressure result, which is a high-resolution range image of the echo signal.

[0066] The specific process of the above scheme is as follows:

[0067] Step 1: Design a multi-carrier broadband signal receiving module, perform intermediate frequency sampling on the echo signal, send the sampled data to the FPGA, use multi-phase extraction filtering technology to implement orthogonal processing of the intermediate frequency signal, and obtain the baseband signal of the echo signal, such as Figure 1 As shown, the specific steps are as follows:

[0068] (1a) Design a multi-carrier broadband signal receiving module, including a receiving antenna, a limiting module, a radio frequency bandpass filter (BPF0), a low noise amplifier (LNA), a mixer, an intermediate frequency bandpass filter (BPF1), an analog-to-digital conversion ADC device and an FPGA connected in sequence;

[0069] (1b) Design a multi-carrier broadband signal waveform, where the frequency interval Δf between adjacent carriers is the symbol time width t b The bandwidth of the multi-carrier broadband signal of N carriers is equal to NΔf; each carrier adopts phase coding modulation, the coding phase is coded with P4 complementary code, and the pulse width is Mt b , the pulse repetition period is (N+1)Mt b , the effective echo time is NMt b .

[0070] The P4 code consists of discrete phases of a linear frequency modulation signal at specific time intervals. The P4 code phase sequence is described as:

[0071]

[0072] Where m = 0, 1, ..., M-1; N is the number of subcarriers, and the P4 complementary code phase is generated by cyclic shift

[0073]

[0074] In the formula, Represents the nth subcarrier, mth symbol coding phase, n = 0, 1, ..., N-1, m = 0, 1, ..., M-1.

[0075] (1c) According to the bandpass sampling theorem, the intermediate frequency is 3NΔf, the sampling frequency is 4NΔf, the sampling data is sent to FPGA for four-phase representation, and the four-phase signal is subjected to digital orthogonal processing based on multi-phase extraction filtering.

[0076] Assuming there is only one target with an initial distance of R and a speed of v, the discrete intermediate frequency subcarrier echo signal of the lth pulse can be expressed as:

[0077]

[0078] Where τ = 2R / c represents the echo delay, and the duration of each coding phase is t b =1 / Δf,t s =1 / f s =t b / 4N represents the sampling interval, and it can be found that the number of points in each code element is 4N. c = 3 × 10 8 m / s is the speed of light in vacuum, n=0,1,…,N-1, N is the number of subcarriers, Δf is the subcarrier frequency interval, k=0,1,…,4N 2 M-1, μ n is the target echo strength of the nth subcarrier signal, w n is the real weighting coefficient of the nth subcarrier, represents the Doppler frequency on the nth subcarrier, is the coding phase of the mth symbol of the nth subcarrier signal. Represents the phase change of the nth subcarrier caused by the path.

[0079] s IF [k] can be expressed in four phases as:

[0080]

[0081]

[0082]

[0083]

[0084] Where k = 0, 1, 2, ..., N 2 M-1,

[0085] I and Q demodulation requires an intermediate frequency local oscillator, whose in-phase component and orthogonal component are:

[0086]

[0087]

[0088] The AD sampled signal is multiplied by the in-phase component and quadrature component of the intermediate frequency local oscillator to obtain:

[0089] s I,mix [4k+0]=s R,IF [4k+0]s IF,I [4k+0]=s R,IF [4k+0]

[0090] s I,mix [4k+1]=s R,IF [4k+1]s IF,I [4k+1]=0

[0091] s I,mix [4k+2]=s R,IF [4k+2]s IF,I [4k+2]=-s R,IF [4k+2]

[0092] s I,mix [4k+3]=s R,IF [4k+3]s IF,I [4k+3]=0

[0093] s Q,mix [4k+0]=s R,IF [4k+0]s IF,Q [4k+0]=0

[0094] s Q,mix [4k+1]=s R,IF [4k+1]s IF,Q [4k+1]=s R,IF [4k+1]

[0095] s Q,mix [4k+2]=s R,IF [4k+2]s IF,Q [4k+2]=0

[0096] sQ,mix [4k+3]=s R,IF [4k+3]s IF,Q [4k+3]=-s R,IF [4k+3]

[0097] It can be found that after I and Q demodulation, the 0th phase of the I output signal is equal to the 0th phase of the input signal, the 1st phase and the 3rd phase are equal to 0, and the 2nd phase is equal to the negative 2nd phase input signal; the 0th phase and the 3rd phase of the Q output signal are equal to 0, the 1st phase is equal to the 1st phase of the input signal, and the 3rd phase is equal to the negative 3rd phase input signal.

[0098] After I and Q demodulation, the obtained signal contains baseband components and intermediate frequency components, so low-pass filtering is required. The low-pass filter coefficient is h[i], i = 0, 1, 2, ..., I-1, where I is an integer multiple of 4, so the filter coefficient can also be divided into four phases, h[4i+i'], i = 0, 1, ..., I / 4-1, i' = 0, 1, 2, 3. The filtering process is as follows.

[0099] In-phase section:

[0100]

[0101] Orthogonal part:

[0102]

[0103]

[0104] After I, Q demodulation and low-pass filtering, the signal is moved to the baseband. Since the signal bandwidth is NΔf, only one phase of the four-phase baseband signal needs to be taken. In this embodiment, the 0th phase is taken. According to the analysis of I, Q demodulation and low-pass filtering, the I and Q components of the 0th phase baseband signal after extraction can be obtained respectively:

[0105]

[0106]

[0107]

[0108] From the above analysis, it can be found that when outputting the in-phase component of the 0th phase of the baseband signal, only the 0th and 2nd phases of the intermediate frequency signal and the 0th and 2nd phases of the filter multiplied by them are needed, and the 2nd phase input signal needs to be inverted; when outputting the orthogonal component of the 0th phase of the baseband signal, only the 1st and 3rd phases of the intermediate frequency signal and the 3rd and 1st phases of the filter multiplied by them are needed, and the 3rd phase of the input signal needs to be inverted. The specific implementation is as follows Figure 2 shown.

[0109] Step 2: Design the subcarrier separation module in FPGA. Figure 3 , the specific steps are as follows:

[0110] (2a) using N accumulators to generate N subcarrier baseband negative frequency phases;

[0111] The n'th subcarrier baseband negative frequency phase can be expressed as:

[0112]

[0113] Where n'=0,1,…,N-1,k=0,1,…,N 2 M-1, k'=0,1,2,3. Take the 0th phase among the four phases and get:

[0114]

[0115] (2b) sending the baseband signal in step (1d) and the phase in step (2a) to the CORDIC module, shifting each subcarrier to zero frequency, and removing the phase effect of the filter while shifting to zero frequency according to the amplitude-phase characteristics of the low-pass filter;

[0116] The CORDIC algorithm converts the vector circular rotation pattern into an iterative shift-and-add algorithm. The equation for each iteration is expressed as:

[0117] x i'+1 =x i' -d i' (2 -i' y i' )

[0118] y i'+1 =y i' +d i' (2 -i' x i )

[0119] z i'+1 =z i' -d i' θ i'

[0120] In the formula, x i' ,y i' , z i' represents the data before the i'+1th iteration; x i'+1 ,y i'+1 , z i'+1 represents the data after the i+1th iteration; θ i' =arctan(2 -i' ), arctan(·) represents the inverse tangent function; the symbol d i'is a decision operator to determine the direction of rotation, and

[0121] After I' iterations, we get:

[0122] x I'+1 =K I' (x0cosz0-y0sinz0)

[0123] y I'+1 =K I' (y0cosz0+x0sinz0)

[0124] By setting x0 = s R,B,I [4k],y0=s R,B,Q [4k], It is possible to move the n'th subcarrier signal to zero frequency, where K I' is the stretch factor, is a fixed phase added by the low-pass filter to the n'th subcarrier, which can be known from the amplitude-phase characteristics of the low-pass filter. I' It is the same for any input and is therefore omitted in subsequent analysis.

[0125] (2c) Design a sliding average filter. For N subcarriers, the coefficients of the sliding average filter are N ones. Its implementation is very simple and only requires simple addition and shifting, without consuming multiplier resources.

[0126] (2d) The output of the CORDIC module in step (2b) is fed into the sliding average filter designed in step (2c) to output the echo data s of each subcarrier R,B,I,n' [4k] and R,B,Q,n' [4k];

[0127] According to the properties of I, Q demodulation, low-pass filter and extraction, we can know that the obtained R,B,I [4k] and s R,B,Q [4k]The resulting complex signal can be expressed as:

[0128]

[0129] in, It is the fixed phase added by the low-pass filter to the nth subcarrier minus the fixed phase added to the n'th subcarrier. According to the CORDIC algorithm, the n'th, n'=0, 1, ..., N-1th subcarrier signals after the shift can be expressed as:

[0130]

[0131] Where, τ'=τ / 4t s , The phase change of the RF and IF due to delay, the phase shift on each subcarrier is consistent. After passing through the sliding average filter, we get:

[0132]

[0133] Obviously, only when When R,MM,n' [4k]≠0.

[0134] Step 3: Design subcarrier pulse compression. The primary pulse compression and secondary pulse compression refer to Figure 4 and Figure 5 , the specific steps are as follows:

[0135] (3a) performing matched filtering on the sliding average filter output of the n'th subcarrier;

[0136] The matched filter coefficients are:

[0137]

[0138] in

[0139] Perform matched filtering on the output of the sliding average filter (one pulse pressure):

[0140]

[0141] When k=(M+1)N-2+τ' DI =k DI and 4f d,n' t s =f d,n' / NΔf≈0,

[0142]

[0143] in Therefore, the final output of a pulse pressure is:

[0144]

[0145] h MF1,n' [i]s R,MM,n' [4(ki)] is the multiplication of two complex numbers. In actual implementation, 4 multipliers and 2 adders are required. For details, refer to Figure 4 .

[0146] (3b) The N matched filter outputs obtained in (3a) are sent to N CORDIC modules to eliminate the fixed phase variation caused by the delay on each subcarrier;

[0147] Obviously, the specific echo time of the target is unknown, but the time when the radar starts processing data is known. When the signal completely enters the radar, the first sampling moment of the signal is 4τ' DI t s The last sampling time of the signal is 4(τ' DI +NM-1)t s , which is also the delay used when the subcarrier is moved in (2d). However, since the phase of the signal does not change, all sampling points in (2d) will have a fixed phase deviation related to the subcarrier frequency and the actual echo time when moving. The purpose of (3b) is to eliminate this phase deviation and compensate for it. Since the phase offset of all sampling points of the same subcarrier is consistent, it is sufficient to compensate for the fixed phase, that is, the phase compensation of the kth pulse output of the n'th subcarrier is

[0148] The pulse compression output of the n'th subcarrier is set to x0 = Re {R 1,n′ [4(NM-1)]}, y0=Im{R 1,n′ [4(NM-1)]}, in Then the output after passing through the CORDIC module is:

[0149]

[0150] (3c) performing secondary pulse compression on the output obtained by the CORDIC module in (3b);

[0151] When the scattering coefficients and weights of different carriers are the same, the sum of the N subcarrier output data is obtained:

[0152]

[0153] Although in the previous analysis τ′ DF ∈(-1,0], but in practice only τ′ needs to be considered DF ∈[-1 / 2,0], because when the sampling deviation here is too large, it means that the next sampling error will become smaller, so the actual possible range of the peak value is [NN / π,NM / 2].

[0154] In order to verify the actual effect of the pulse compression method provided by the present invention, the following simulation experiments and analyses are performed based on the above scheme:

[0155] 1. Experimental conditions: The signal form is a multi-carrier phase coded signal, the subcarriers are uniformly weighted, the coding phase uses P4 code, the subcarrier frequency interval Δf = 6.25MHz, the number of subcarriers N = 32, the bandwidth is B = 200MHz, and the time width is T PD = 20.48us, sampling rate is f s =800MHz.

[0156] 2. Simulation content:

[0157] Simulation 1: Design a low-pass decimation filter based on the following parameters: passband cutoff frequency 100MHz, stopband cutoff frequency 300MHz, passband ripple 0.01dB, stopband attenuation 60dB; Figure 6 and Figure 7 The phase-frequency response and amplitude-frequency response of the designed filter are given respectively.

[0158] Simulation 2: Based on the filter designed above, perform orthogonal transformation. Figure 8 and Fig. 9 The spectrum diagrams of the intermediate frequency signal after AD sampling and the baseband signal after extraction are given respectively.

[0159] Simulation 3: Based on the above orthogonal transform output, subcarrier separation and pulse compression are performed. Fig.10 The theoretical results of the 0th subcarrier pulse compression and the results of the first pulse compression of the separated subcarrier of this patent are given. Fig.11 The secondary pulse compression results of the ideal and the inventive schemes are given.

[0160] Simulation 4: Assume that the echo delay difference between two adjacent targets is 5t s =5 / 4NΔf, Fig.12 The ideal result of two adjacent target echoes and the result of a pulse compression of the 0th subcarrier of the scheme of the present invention are given. Fig.13 The ideal results of two adjacent target echoes and the secondary pulse compression results of the scheme of the present invention are given.

[0161] Simulation 5: Assuming the carrier frequency f c =10GHz, Fig.14 The fuzzy function diagram of a pulse compression of the present invention is given. Fig.15 The fuzzy function diagram of the secondary pulse compression of the present invention is given. Fig.16 The curve of the maximum peak sidelobe ratio (PSLR) of the secondary pulse compression output of the present invention varying with speed is given.

[0162] 3. Analysis of simulation results:

[0163] from Figure 6 and Figure 7It can be seen that the amplitude of the low-pass filter is basically unchanged within the passband, and the phase changes linearly. Since the OFDM signal is the cumulative sum of multiple single-frequency signals, the impact of the low-pass filter on each subcarrier can be directly calculated.

[0164] from Figure 8 and Fig. 9 It can be seen that the spectrum of the signal obtained after AD sampling is symmetrical, and after down-conversion, low-pass filtering and four-fold decimation, the sampling rate of the signal becomes NΔf, the starting and ending frequencies are -NΔf / 2 and NΔf / 2, and the amplitude fluctuates slightly with the fluctuation of the amplitude-frequency response of the filter.

[0165] from Fig.10 It can be seen that the single subcarrier pulse compression result of the present invention is basically consistent with the ideal single subcarrier pulse compression result. Fig.11 It can be seen that the secondary pulse pressure results of the present invention are also basically consistent with the direct pulse pressure results.

[0166] from Fig.12 It can be seen that since the echo delay difference 5 / 4NΔf of the two targets is less than the subcarrier delay resolution 1 / Δf, the two targets cannot be effectively separated. Fig.13 It can be seen that since the echo delay difference 5 / 4NΔf of the two targets is greater than the broadband delay resolution 1 / NΔf, the two targets can be effectively separated.

[0167] Under this simulation condition, f d / (Δf / N) = 0.1 represents a speed of about 276.4 m / s, which can be obtained from Fig.14 and Fig.15 and see, in f d / (Δf / N)=0.3, that is, when the speed is about 829.2 m / s, the peak value of the secondary pulse pressure result is attenuated by about 0.2, and the primary pulse pressure is attenuated by about 0.05. Therefore, the multi-carrier broadband signal encoded based on P4 cyclic code has better Doppler tolerance.

[0168] from Fig.16 It can be seen that as the target speed increases, the PSLR will gradually increase.

[0169] The simulation results demonstrate the effectiveness of the multi-carrier broadband signal digital pulse compression method proposed in the present invention.

Claims

1. A broadband multi-carrier signal frequency domain multiplexing pulse compression method, characterized in that: The steps include: S1: Perform polyphase extraction filtering on the broadband multi-carrier signal to obtain the baseband signal of the echo signal; S2: Move each subcarrier signal to zero frequency, and smooth it through a designed sliding average filter to output the echo data of each subcarrier; S3: Perform pulse compression on the output of the sliding average filter to obtain the range image of the subcarrier; S4: performing phase rotation on each subcarrier range image respectively, and then performing summation operation to obtain a secondary pulse pressure result, which is a high-resolution range image of the echo signal; The expression of the primary pulse pressure method in step S3 is: In the formula, R 1,n' [4k] is the frequency domain pulse compression output of the n'th subcarrier, k = 0, 1, ..., N 2 M-1,s R,MM,n' [4(k-m')] is the n'th subcarrier echo signal after smoothing filter, h MF1,n' [m'] is the n'th subcarrier matched filter coefficient, i.e., the conjugate flip of the n'th subcarrier code phase; Perform a phase rotation on the pulse compression result of the n'th subcarrier to obtain R' 1,n' [4k], the secondary pulse pressure method in step S4 is expressed as:

2. A broadband multi-carrier signal frequency domain multiplexing pulse compression method according to claim 1, characterized in that: The baseband signal in step S1 is obtained by designing a multi-carrier broadband signal receiving module, performing intermediate frequency sampling on the echo signal, sending the sampled data to the FPGA, and using multi-phase extraction filtering technology to implement orthogonal processing of the intermediate frequency signal to obtain the baseband signal of the echo signal.

3. A broadband multi-carrier signal frequency domain multiplexing pulse compression method according to claim 2, characterized in that: The step S1 specifically includes the following steps: A1: Design a multi-carrier broadband signal receiving module, including a receiving antenna, a limiting module, a radio frequency bandpass filter, a low noise amplifier, a mixer, an intermediate frequency bandpass filter, an analog-to-digital conversion ADC device and an FPGA connected in sequence; A2: Design a multi-carrier broadband signal waveform, where the frequency interval Δf between adjacent carriers is the symbol time width t b The reciprocal of the multi-carrier wideband signal bandwidth of N carriers is equal to NΔf, and each carrier adopts phase coding modulation; A3: According to the bandpass sampling theorem, the intermediate frequency is 3NΔf, the sampling frequency is 4NΔf, the sampling data is sent to the FPGA for four-phase representation, and the four-phase signal is subjected to digital orthogonal processing based on multi-phase extraction filtering to obtain the baseband signal of the echo signal.

4. The method for frequency domain multiplexing pulse compression of broadband multi-carrier signals according to claim 1, characterized in that: The step S2 specifically includes the following steps: B1: Generate N subcarrier baseband negative frequency phases using N accumulators; B2: Send the baseband signal of the echo signal and the subcarrier baseband negative frequency phase obtained in step B1 to the CORDIC module to move each subcarrier to zero frequency; B3: Design a sliding average filter. For N subcarriers, the coefficients of the sliding average filter are N ones. B4: Send the output of the CORDIC module in step B2 to the sliding average filter designed in step B3, and output the echo data of each subcarrier.

5. The method for frequency domain multiplexing pulse compression of broadband multi-carrier signals according to claim 1, characterized in that: The step S3 specifically includes: performing matched filtering on the sliding average filter output of the n'th subcarrier to obtain the range image of the subcarrier.

6. A broadband multi-carrier signal frequency domain multiplexing pulse compression method according to claim 5, characterized in that: The step S4 is specifically as follows: sending the obtained N matched filter outputs to N CORDIC modules to eliminate the fixed phase change caused by delay on each subcarrier; summing the outputs obtained by the CORDIC modules to obtain a secondary pulse compression result, which is a high-resolution range image of the echo signal.

7. A broadband multi-carrier signal frequency domain multiplexing pulse compression method according to claim 3, characterized in that: The result of performing digital orthogonal processing based on multi-phase extraction filtering on the four-phase signal in step A3 is expressed as: Where h[4i+i'] represents the i'th phase of the decimation filter, where I is an integer multiple of 4, i = 0, 1, ..., I / 4-1, i' = 0, 1, 2, 3, s R,IF [4k+k'] represents the k'th phase input of the decimation filter, k = 0, 1, 2, ..., N 2 M-1, k'=0,1,2,3,s R,B [4k+k'] is the k'th phase output of the decimation filter.

8. A broadband multi-carrier signal frequency domain multiplexing pulse compression method according to claim 4, characterized in that: The step B1 is specifically as follows: The negative frequency phase of the n'th subcarrier can be expressed as: Where k = 0, 1, ..., N 2 M-1, n'=0,1,…,N-1.

Images