A pulse compression and difference ranging method

Abstract

The purpose of the present invention is to realize a pulse compression and difference ranging method, improve the distance measurement accuracy of digital pulse compression, break through the limitation of range gate accuracy, and have higher distance measurement accuracy than amplitude weighted method under most distance deviation and signal-to-noise ratio (SNR) conditions.

Description

Technical Field

[0001] The invention belongs to the technical field of radar information processing, and in particular relates to a pulse compression and difference ranging method. Background Art

[0002] Radar measures the distance of a target by measuring the delay time of the target echo. Pulse compression radar transmits a signal with a sufficiently large pulse width and bandwidth, and can obtain a higher distance resolution by compressing the received signal pulse. In most modern radars, pulse compression usually gives priority to digital processing. In digital pulse compression processing, the radar's effective range is usually discretized into multiple range gates, and the accuracy of the range gates is limited by the sampling rate, which limits the radar's distance measurement accuracy. In current applications, adjacent range gates at the target position are usually weighted by amplitude to improve the accuracy of distance measurement. However, the measurement accuracy of this method is greatly affected by the offset of the target at the full range scale. With the increasing requirements for radar distance measurement accuracy in various industries, this method is increasingly difficult to meet the requirements. Therefore, further improving the accuracy of digital pulse compression distance measurement is an important task for modern radars. Summary of the invention

[0003] The purpose of the present invention is to realize a pulse compression and difference ranging method, improve the distance measurement accuracy of digital pulse compression, break through the limitation of range gate accuracy, and have higher distance measurement accuracy than amplitude weighted method under most distance deviation and signal-to-noise ratio (SNR) conditions. The main process of this technology is: divide the reference signal into left and right halves, calculate and store frequency domain pulse compression coefficients respectively, use these two sets of frequency domain pulse compression coefficients respectively, perform pulse compression on radar receiving AD signal, then perform sum and difference operations on the two sets of pulse compression results respectively, use the sum operation output signal to detect the target, calculate the imaginary part of the difference operation and sum operation ratio at the target, look up the table to obtain the corresponding distance deviation, and finally compensate the distance deviation for the target position, so as to achieve accurate ranging.

[0004] A pulse compression and difference ranging method, the specific steps are as follows:

[0005] Step 1: Create and store distance deviation table

[0006] Step 1.1: Construct signal s(n)

[0007] Assume that the bandwidth of the radar transmitted baseband signal s(t) is B and the time width is t p , with sampling rate f s The reference signal s(t) is sampled to obtain a discrete signal s(n).

[0008] Step 1.2: Set the right half of the discrete signal sequence s(n) to zero to generate s L(n), set the left half of s(n) to zero, and generate s R (n).

[0009] Step 1.3, use s L (n) and s R (n) is used as the reference signal, and it is flipped, conjugated, windowed, and fast Fourier transformed (FFT) to construct the frequency domain pulse compression coefficient H L With H R .

[0010] Step 1.4: Distance quantification unit is R △ =c / (2f s ), set a positive integer α, and set the distance αR △ to (α+0.5)R △ Split into distance scale R 1 ,R 2 ,…,R N , perform the following calculations for each distance scale

[0011] Step 1.4.1: Construct the noise-free received signal s in (t), and sample and pulse compress it to obtain s outL With s outR

[0012] Set the current distance scale to R i , 1≤i≤N is a positive integer, then the delay is τ=2R i / c, where c is the speed of light, to construct a noise-free receiving signal:

[0013]

[0014] After sampling, the frequency domain pulse compression coefficient H is used respectively. L With H R To perform pulse compression:

[0015]

[0016] in, is the hadamard product of the matrix, and IFFT{·} is the inverse fast Fourier transform.

[0017] Step 1.4.2: outL With s outR Sum and difference

[0018] s + =s outL +s outR (4)

[0019] s - =soutL -s outR (5)

[0020] Obviously + is the conventional pulse compression output signal, and its peak index is n d .

[0021] Step 1.4.3. Calculate n d The absolute value p(i) of the imaginary part of the ratio of the difference signal to the sum signal, and R i The corresponding distance deviation q(i)

[0022] difference signal - and signal + In n d Orthogonal, s - (n d ) / s + (n d )When the imaginary part is positive, it means that the point is on the left side of the target position, and when it is negative, it means that the point is on the right side of the target position. d The absolute value of the imaginary part of the ratio of the difference signal to the sum signal is:

[0023] p(i)=|imag{s - (n d ) / s + (n d )}| (6)

[0024] Among them, imag{·} is the imaginary part.

[0025] Calculate R i The corresponding distance deviation q(i):

[0026] q(i)=R i -αR △ (7)

[0027] Step 1.5: For all distance scales R 1 ,R 2 ,…,R N After calculation, p and q are matched one by one to make a distance deviation table and store it for subsequent use.

[0028] Step 2: Pulse compression and differential ranging method

[0029] Step 2.1: Calculate and store the frequency domain pulse compression coefficient H under specific radar parameters in advance according to the method in step 1.3. L With H R .

[0030] Step 2.2: Let the received AD signal be u(n), n=1,2,…,N s, N s is the number of sampling points, and the domain pulse compression coefficient H is used respectively. L With H R Pulse compression for u(n)

[0031]

[0032] Step 2.3: Sum and subtract the pulse compression results (8) and (9) respectively.

[0033] u + =u outL +u outR (10)

[0034] u - =u outL -u outR (11)

[0035] Step 2.4: u + Perform target detection and record the target index as n 0 , calculate u - (n 0 ) / u + (n 0 )The absolute value of the imaginary part:

[0036] p(n 0 )=|imag{u - (n 0 ) / u + (n 0 )}| (12)

[0037] Step 2.5, load the distance deviation table stored in step 1.5, according to p(n 0 ) values ​​in p, select the two closest values, denoted as p(n 1 ) and p(n 2 ), and record the corresponding distance deviation, respectively denoted as q(n 1 ) and q(n 2 ).

[0038] Step 2.6: Calculate the target distance R by linear interpolation

[0039]

[0040] Among them, R min is the minimum detection distance, when imag{u - (n 0 ) / u + (n 0 )}≥0, the ± in formula (13) is taken as +, when imag{u - (n0 ) / u + (n 0 )}<0, ± in formula (13) is -.

[0041] The beneficial effects of the present invention are:

[0042] The target distance calculated by this method can break through the accuracy limitation of the radar range gate and has good distance measurement root mean square error (RMSE) performance. Figure 3 This is a performance comparison chart of the root mean square error corresponding to the deviation of the target position from the integer distance quantization unit under the parameters of Example 1 and Example 2, when the distance quantization unit is 60m and the target signal-to-noise ratio after pulse compression is about 17.41dB. It can be seen that the method in this paper has a lower RMSE under all distance offsets, and is better than the three-point amplitude weighted method when the distance offset is greater than about 14m. Figure 4 for Figure 3 The enlarged image of our method in the figure shows that the distance accuracy of our method when the distance offset is low is better than that when the distance offset is large. Figure 5 Under the parameters of Example 1 and Example 2, when the distance offset is uniformly distributed between -30m and 30m, the distance measurement RMSE under different target SNRs after pulse compression using the distance deviation table of Example 1 is shown. It can be seen that as the SNR increases, the accuracy of the method in this paper increases, which is significantly better than the three-point amplitude weighted method. BRIEF DESCRIPTION OF THE DRAWINGS

[0043] Figure 1 Example 2 Schematic diagram of the real part of AD signal (the theoretical SNR value after pulse compression is 18dB)

[0044] Figure 2 Example 2 Pulse compression and signal schematic diagram (theoretical SNR value after pulse compression is 18dB)

[0045] Figure 3 RMSE of distance measurement under different distance offsets (the theoretical value of SNR after pulse compression is 18dB)

[0046] Figure 4 The RMSE of distance measurement under different distance offsets (the theoretical value of SNR after pulse compression is 18dB)

[0047] Figure 5 Distance RMSE under different pulse pressure SNR DETAILED DESCRIPTION

[0048] The technical solution provided by the present invention will be described in detail below in conjunction with specific embodiments. It should be understood that the following specific implementation methods are only used to illustrate the present invention and are not used to limit the scope of the present invention.

[0049] Example 1: Establishing and storing a distance deviation table

[0050] (1) The baseband signal transmitted by the radar is selected to have a bandwidth of B = 2 MHz and a time width of t p = 20μs zero intermediate frequency linear frequency modulation signal:

[0051]

[0052] With sampling rate f s =2.5MHz for the reference signal s(t) at -t p / 2≤t≤t p / 2 time, and obtain the discrete signal sequence s(n), n=1,2,…,51, and the total number of sampling points M=51.

[0053] (2) Use the left and right halves of the discrete signal sequence s(n) to construct two sequences:

[0054]

[0055] (3) Use s L (n) and s R (n) is used as the reference signal to construct the frequency domain pulse compression coefficient. By conjugating, windowing, and fast Fourier transforming, the frequency domain pulse compression coefficient can be obtained:

[0056]

[0057] Wherein, W is a Hamming window with a length of 51, and the length of FFT is set to L=512.

[0058] (4) The distance quantification unit is R △ =c / (2f s )=60m, respectively set the distance R i =15000+0.5(i-1)(m), i=1,2,…,61, corresponding delay is τ i =2R i / c, construct a noise-free receiving signal:

[0059]

[0060] f s =2.5MHz at time -t p / 2≤t≤L / f s -t p / 2 sampling, get AD signal s in (n), n=1,2,…,L, respectively using the frequency domain pulse compression coefficient H L With H R , for inTo perform pulse compression:

[0061]

[0062] (5) For s outL With s outR Find the sum and difference separately:

[0063] s + =s outL +s outR (twenty two)

[0064] s - =s outL -s outR (twenty three)

[0065] Notes + The index of the maximum peak is n d .

[0066] (6) Calculate n d The absolute value of the imaginary part of the ratio of the difference signal to the sum signal:

[0067] p(i)=|imag{s - (n d ) / s + (n d )}|,i=1,2,…,61 (24)

[0068] Calculate R i The corresponding distance deviation q(i):

[0069] q(i)=R i -αR △ =0.5(i-1),i=1,2,…,61 (25)

[0070] P and q are matched one by one, and a 61-point distance deviation table (as shown in Table 1) is prepared and stored.

[0071] Table 1 61-point distance deviation table

[0072]

[0073] Embodiment 2: Pulse compression and differential ranging based on received AD signals

[0074] (1) Assume that the baseband signal transmitted by the radar is a zero-IF linear frequency modulation signal with a bandwidth of B = 2 MHz and a pulse width of t p = 20 μs, calculate and store the normalized frequency domain pulse compression coefficient H L With H R .

[0075] (2) Assume the sampling rate of the receiver is fs =2.5MHz, the number of sampling points is 512. The receiver starts receiving sampling after the pulse transmission is completed, and the received AD signal is recorded as u(n), n=1,2,…,512, the target distance is set to 17.2125km, the background noise is Gaussian white noise, and the target SNR after pulse compression is 17.41dB (the theoretical value of SNR after pulse compression is 18dB). Figure 1 The real part data of the AD signal is displayed, and it can be seen that the sampling point indexes of the pulse are 237 to 288. The domain pulse compression coefficient H is used respectively. L With H R For u(n) pulse compression:

[0076]

[0077] (3) Sum and subtract the pulse compression results:

[0078] u + =u outL +u outR (28)

[0079] u - =u outL -u outR (29)

[0080] (4) Search u + The points that meet the detection threshold are given by Figure 1 As shown, the peak occurs at index n 0 = 237. Select peak value n 0 = 237 as a target, calculate u - (n 0 ) / u + (n 0 )The absolute value of the imaginary part:

[0081] p(n 0 )=|imag{u - (n 0 ) / u + (n 0 )}|=0.049973 (30)

[0082] (5) Load the distance deviation table stored in Example 1, p(n 0 ) is 0.049973, and the closest two values ​​are p(n 1 )=0.047994 and p(n 2 )=0.054862, and the corresponding distance deviations are q(n 1 )=3.5m and q(n 2 )=4m.

[0083] (6) Assume R min =ct p / 2=3000m,due to imag{u - (n 0 ) / u + (n 0 )}=-0.049973<0, the target distance R can be calculated according to formula (13):

[0084]

[0085] It can be seen that compared with the actual target distance of 17.2125km, the error is 3.9m.

[0086] The above description is only the best specific implementation mode of the present invention, but the protection scope of the present invention is not limited thereto. Any changes or substitutions that can be easily thought of by any technician familiar with the technical field within the technical scope disclosed by the present invention should be covered within the protection scope of the present invention.

[0087] The contents not described in detail in the specification of the present invention belong to the common knowledge of the professionals in this field.

Claims

1. A pulse compression and differential ranging method, It is characterized in that The steps of this method are as follows: Step 1: Create and store distance deviation table Step 1.1: Construct signal s(n) Assume that the bandwidth of the radar transmitted baseband signal s(t) is B and the time width is t p , with sampling rate f s Sample the reference signal s(t) to obtain a discrete signal s(n); Step 1.2: Set the right half of the discrete signal sequence s(n) to zero to generate s L (n), set the left half of s(n) to zero, and generate s R (n); Step 1.3, use s L (n) and s R (n) is used as the reference signal, and it is flipped, conjugated, windowed, and fast Fourier transformed to construct the frequency domain pulse compression coefficient H L With H R ; Step 1.4: Distance quantification unit is R △ =c / (2f s ), where c is the speed of light, f s As the sampling rate, set a positive integer α and set the distance αR △ to (α+0.5)R △ Split into distance scale R 1 ,R 2 ,…,R N , perform the following calculation steps for each distance scale respectively 1.4.1, construct the noise-free received signal s in (t), and sample and pulse compress it to obtain s outL With s outR ; Step 1.4.2: outL With s outR Sum and difference s + =s outL +s outR (4) s - =s outL -s outR (5) Obviously + is the conventional pulse compression output signal, and its peak index is n d ; Step 1.4.

3. Calculate n d The absolute value p(i) of the imaginary part of the ratio of the difference signal to the sum signal, and R i The corresponding distance deviation q(i); Step 1.5: For all distance scales R 1 ,R 2 ,…,R N After calculation, p and q are matched one by one to make a distance deviation table and store it for subsequent use; Step 2: Pulse compression and differential ranging method Step 2.1: Calculate and store the frequency domain pulse compression coefficient H under specific radar parameters in advance according to the method in step 1.

3. L With H R ; Step 2.2: Let the received AD signal be u(n), n=1,2,…,N s , N s is the number of sampling points, and the domain pulse compression coefficient H is used respectively. L With H R Pulse compression for u(n) Step 2.3: Sum and subtract the pulse compression results (8) and (9) respectively. in + =in outL +in outR (10) in - =in outL -in outR (11) Step 2.4: u + Perform target detection and record the target index as n 0 , calculate u - (n 0 ) / u + (n 0 )The absolute value of the imaginary part: p(n 0 )=|imag{u - (n 0 ) / u + (n 0 )}| (12) Step 2.5, load the distance deviation table stored in step 1.5, according to p(n 0 ) values ​​in p, select the two closest values, denoted as p(n 1 ) and p(n 2 ), and record the corresponding distance deviation, respectively denoted as q(n 1 ) and q(n 2 ); Step 2.6: Calculate the target distance R by linear interpolation Among them, R min is the minimum detection distance, when imag{u - (n 0 ) / u + (n 0 )}≥0, the ± in formula (13) is taken as +, when imag{u - (n 0 ) / u + (n 0 )}<0, ± in formula (13) is -.

2. The method according to claim 1, It is characterized in that In step 1.4.1, let the current distance scale be R i , 1≤i≤N is a positive integer, then the delay is τ=2R i / c, where c is the speed of light, to construct a noise-free receiving signal: After sampling, the frequency domain pulse compression coefficient H is used respectively. L With H R To perform pulse compression: in, is the hadamard product of the matrix, and IFFT{·} is the inverse fast Fourier transform.

3. The method according to claim 1, It is characterized in that In step 1.4.3, the difference signal s - and signal + In n d Orthogonal, s - (n d ) / s + (n d ) When the imaginary part is positive, it means that the point is on the left side of the target position, and when it is negative, it means that the point is on the right side of the target position; record n d The absolute value of the imaginary part of the ratio of the difference signal to the sum signal is: p(i)=|imag{s - (n d ) / s + (n d )}| (6) Among them, imag{·} is the imaginary part; Calculate R i The corresponding distance deviation q(i): q(i)=R i -αR △ (7)。

Images