PRI sequence design method based on nonlinear changes of Staggered SAR system

Abstract

The present invention relates to a method for designing a PRI sequence based on a nonlinearly changing PRI sequence of a StaggeredSAR system. The purpose of the present invention is to solve the limitation problem of the blind area distribution of the linearly changing PRI sequence in the existing StaggeredSAR system. The process is: one: based on the StaggeredSAR system, the blind area position corresponding to the changing PRI sequence is calculated to form a blind area distribution map corresponding to the PRI sequence; two: quantitatively describe the discontinuity of the blind area position of the blind area distribution map; three: quantitatively describe the uniformity of the blind area position of the blind area distribution map; four: define the target criterion function of the nonlinearly changing PRI sequence; five: use a genetic algorithm to calculate the target criterion function for each PRI sequence and optimize it to obtain the PRI sequence corresponding to the optimal target criterion function value. The present invention is used in the field of microwave remote sensing technology.

Description

Technical Field

[0001] The invention relates to the field of microwave remote sensing technology, in particular to a method for designing a PRI sequence based on nonlinear changes of a Staggered SAR system. Background Art

[0002] Staggered SAR can achieve high-resolution wide-swath SAR imaging by using a variable PRI sequence combined with the pitch-dimensional multi-beam technology. PRI sequence design can solve the blind spot problem of wide swaths, but also affects the reconstruction effect of SAR imaging, and has gradually become a hot topic for domestic and foreign scholars. At present, the research on the sequence design of Staggered SAR system focuses on linear changes. The advantage of linear changes is that the position distribution of blind spots can be quantitatively calculated by establishing a linear change model, but there are certain limitations. It is mainly manifested in its blind spot distribution map, where there will be many continuous pulse missing (such as slow-changing linear PRI change sequence) or the position of pulse missing has a strong periodicity (such as fast-changing linear PRI sequence and fine linear PRI sequence). This will affect the effect of data recovery, resulting in high side lobes near the main lobe or far away from the main lobe. This phenomenon is more prominent at low oversampling rates. In fact, if the PRI change interval is determined, the PRI value of each pulse in the sequence can be arbitrarily selected within the interval, or even randomly changed. However, at this time, there will be more continuous pulse missing at a specific distance dimension in the blind spot distribution map, and there will be very few pulse missing data at some distance dimensions. This will result in significant differences in range-dimension imaging performance across the swath. Summary of the invention

[0003] The purpose of the present invention is to solve the limitation problem of blind area distribution of linearly changing PRI sequences in the existing Staggered SAR system, and to propose a design method of nonlinearly changing PRI sequences based on the Staggered SAR system.

[0004] The PRI sequence design method based on the nonlinear change of the Staggered SAR system is carried out in the following steps:

[0005] Step 1: Based on the Staggered SAR system, calculate the blind area position corresponding to the changing PRI sequence and form a blind area distribution map corresponding to the PRI sequence;

[0006] Step 2: Considering the requirements of the blind spot distribution map for the discontinuity of the blind spot position, quantitatively describe the discontinuity of the blind spot position of the blind spot distribution map;

[0007] The requirements of the blind spot distribution map for the discontinuity of the blind spot position are:

[0008] If a pulse is missing at a certain slant range unit, the adjacent pulses of the missing pulse are not expected to be missing;

[0009] Step 3: Based on the result of step 2, considering the requirement of the blind spot distribution map for the uniformity of the blind spot position, quantitatively describe the uniformity of the blind spot position of the blind spot distribution map;

[0010] The requirements of the blind spot distribution map for the uniformity of the blind spot position are:

[0011] In the blind area distribution diagram I(i,j), the proportion of blind areas can be evenly distributed along the horizontal axis j, that is, the element value of the kth row of P(k,j) can satisfy the uniform distribution;

[0012] Step 4: Based on the quantitative description of the discontinuity of the blind area position of the blind area distribution map in step 2 and the quantitative description of the uniformity of the blind area position of the blind area distribution map in step 3, define the target criterion function of the nonlinearly changing PRI sequence;

[0013] Step 5: Based on the target criterion function obtained in step 4, a genetic algorithm is used to calculate and optimize the target criterion function for each PRI sequence to obtain the PRI sequence corresponding to the optimal target criterion function value.

[0014] The present invention includes the following beneficial effects:

[0015] The present invention breaks through the limitation of linear change and makes full use of the significant advantages of non-uniform sampling in data recovery in subsequent signal processing. The present invention proposes for the first time a non-linear change PRI sequence design scheme based on the Staggered SAR system, defines the optimization criteria for sequence design, and uses a genetic algorithm to design a non-linear change PRI sequence in combination with the characteristics of the sequence;

[0016] 1. The present invention introduces nonlinear PRI variation sequence into the Staggered SAR system, breaking through the limitation of linear variation PRI sequence, and its blind area distribution has significant advantages in terms of uniformity and discontinuity of blind area position.

[0017] 2. The target criterion function of the blind area distribution map corresponding to the PRI sequence is defined, which provides a quantitative measurement standard for the discontinuity and uniformity of the blind area distribution map.

[0018] 3. The results of comparative experiments show that the nonlinear change PRI sequence proposed in the sequence design of the present invention has better index performance. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] Figure 1 A flowchart of the nonlinear PRI sequence generation method of the present invention;

[0020] Figure 2Design a flow chart for nonlinear sequence optimization of staggered SAR system;

[0021] Figure 3(a) shows the PRI variation pattern of the random variation sequence;

[0022] Figure 3(b) is the blind area distribution diagram of the random change sequence;

[0023] Figure 3(c) is a graph showing the percentage of missing pulses for a random variation sequence;

[0024] Figure 3(d) is a graph showing the relationship between pulse missing intervals of a random variation sequence;

[0025] Figure 3(e) is a graph showing the relationship between the pulses without missing intervals in a random variation sequence;

[0026] Transmitted pulse in one period is the number of transmitted pulses, Slant range (km) is the slant range (km), Transmitted pulse is the transmitted pulse, Percentage of lost pulses is the percentage of lost pulses, Pulse separation is the pulse separation, The interval without lost pulses is the missing pulse interval, PRIs is the PRI sequence (PRI0 to PRI M-1 );

[0027] Figure 4(a) shows the PRI variation rule of the linear fine variation sequence;

[0028] Figure 4(b) is the blind area distribution diagram of the linear fine change sequence;

[0029] Fig. 4(c) is a graph showing the percentage of missing pulses for a linear fine variation sequence;

[0030] FIG4( d ) is a graph showing the relationship between pulse missing intervals of a linear fine variation sequence;

[0031] Fig. 4(e) is a relationship diagram of pulse-free intervals of a linear fine-variation sequence;

[0032] Transmitted pulse in one period is the number of transmitted pulses, Slant range (km) is the slant range (km), Transmitted pulse is the transmitted pulse, Percentage of lost pulses is the percentage of lost pulses, Pulse separation is the pulse separation, The interval without lost pulses is the missing pulse interval, PRIs is the PRI sequence (PRI0 to PRIM-1 );

[0033] Figure 5(a) shows the PRI variation rule of the nonlinear variation sequence;

[0034] Figure 5(b) is the blind area distribution diagram of the nonlinear change sequence;

[0035] Fig. 5(c) is a graph showing the percentage of missing pulses for a nonlinear variation sequence;

[0036] FIG5( d ) is a graph showing the relationship between pulse missing intervals of a nonlinear variation sequence;

[0037] Fig. 5(e) is a relation diagram of pulse-free intervals of a nonlinear variation sequence;

[0038] Transmitted pulse in one period is the number of transmitted pulses, Slant range (km) is the slant range (km), Transmitted pulse is the transmitted pulse, Percentage of lost pulses is the percentage of lost pulses, Pulse separation is the pulse separation, The interval without lost pulses is the missing pulse interval, PRIs is the PRI sequence (PRI0 to PRI M-1 );

[0039] Figure 6 It is a curve graph showing the relationship between the objective criterion function value and the number of iterations. The number of iteration is the number of iterations, and the Objective function is the objective criterion function. DETAILED DESCRIPTION

[0040] In order to make the above-mentioned objects, features and advantages of the present invention more clearly understood, the following Figures 1 to 6 The present invention is further described in detail with specific embodiments, wherein Figure 1 The present invention is a flowchart of the PRI sequence design method based on the nonlinear change of the Staggered SAR system.

[0041] Specific implementation method 1: This implementation method is based on the nonlinear change PRI sequence design method of the Staggered SAR system, and is carried out in the following steps:

[0042] Step 1: Based on the Staggered SAR system, calculate the changing PRI sequence (PRI0 to PRI M-1 ) corresponding to the blind area position, forming a blind area distribution map corresponding to the PRI sequence;

[0043] Step 2: Considering the requirements of the blind spot distribution map for the discontinuity of the blind spot position, quantitatively describe the discontinuity of the blind spot position of the blind spot distribution map;

[0044] The requirements of the blind spot distribution map for the discontinuity of the blind spot position are:

[0045] If a pulse is missing at a certain slant range unit, the adjacent pulses of the missing pulse are not expected to be missing;

[0046] Step 3: Based on the result of step 2, considering the requirement of the blind spot distribution map for the uniformity of the blind spot position, quantitatively describe the uniformity of the blind spot position of the blind spot distribution map;

[0047] The requirements of the blind spot distribution map for the uniformity of the blind spot position are:

[0048] In the blind area distribution diagram I(i,j), the proportion of blind areas (pulse loss) can be evenly distributed along the horizontal axis j (slant distance), that is, the element value of the kth row of P(k,j) can satisfy the uniform distribution;

[0049] Step 4: Based on the quantitative description of the discontinuity of the blind area position of the blind area distribution map in step 2 and the quantitative description of the uniformity of the blind area position of the blind area distribution map in step 3, define the target criterion function of the nonlinearly changing PRI sequence;

[0050] Step 5: Based on the target criterion function obtained in step 4, the design model of the nonlinear PRI sequence is transformed into an optimization problem. The target criterion function is calculated and optimized for each PRI sequence using a genetic algorithm to obtain the optimal target criterion function value (criterion cost function The PRI sequence corresponding to the smallest value).

[0051] The specific process is as follows:

[0052] As a heuristic algorithm based on population evolution, the genetic algorithm exchanges information, selects the best and discards the worst through methods such as crossover and mutation between individuals in the population. It is an adaptive global optimization probability search algorithm. The present invention uses a genetic algorithm to optimize the objective function criterion defined in the present invention, and evolves from generation to generation to produce better and better nonlinear change PRI sequence individuals.

[0053] The initial population is composed of randomly changing PRI sequence individuals. The individuals are evaluated according to the value of the defined target criterion function, and the individuals with low target criterion function values ​​are selected. The genetic operators of natural genetics are used for combined crossover and mutation to generate PRI sequence individuals representing new nonlinear changes.

[0054] This process will lead to the population being more adaptable to the environment than the previous generation, just like natural evolution. When the target criterion function value no longer decreases, or the number of iterations reaches the preset value, the iteration is terminated, and finally a nonlinear PRI sequence with the optimal target criterion function value is obtained.

[0055] Specific implementation method 2: This implementation method is different from the specific implementation method 1 in that in step 1, based on the Staggered SAR system, the changing PRI sequence (PRI0 to PRI M-1 ) corresponding to the blind area position, forming a blind area distribution map corresponding to the PRI sequence; the specific process is:

[0056] In the Staggered SAR mode, there are M pulses in the transmit and receive sequence, and the mth pulse has a corresponding pulse repetition period of PRI. m , m=0,1,…,M-1;

[0057] The Staggered SAR is a staggered synthetic aperture radar;

[0058] Defining PRI max and PRI min They correspond to the maximum and minimum pulse repetition periods among the M pulses respectively;

[0059] ΔPRI=PRI max -PRI min Indicates the variation range of the pulse repetition period PRI;

[0060] The emission time of the mth pulse is t m The expression is

[0061]

[0062] In the formula, PRI p is the pulse repetition period of the pth pulse, and p is the pulse number;

[0063] For the mth transmitted pulse, the arrival time of the mth pulse echo is the same as that of the (m+n)th transmitted pulse, which causes the mth pulse to be lost in the nth blind area. The target echo falling in the nth blind area cannot be received because the transmission and reception cannot be simultaneous, which is expressed as

[0064]

[0065] Among them, t r is the echo delay time, PRI i is the pulse repetition period of the ith pulse, T p is the pulse width;

[0066] According to formula (2), the echo delay time tr The range of t r =2R0 / c to solve the range of slant distance R0 corresponding to the blind area, so as to obtain all blind area positions of the mth transmitting pulse and form a blind area distribution map;

[0067] Where c is the speed of light.

[0068] The blind area map reflects the distribution of the pulse loss positions of the target echo within the surveying band. The PRI sequence determines the blind area distribution map, and the position of the pulse loss in the blind area distribution map directly affects the effect of data recovery.

[0069] The other steps and parameters are the same as those in the first embodiment.

[0070] Specific implementation method three: This implementation method is different from specific implementation methods one or two in that, in step two, the requirement of the blind spot distribution map on the discontinuity of the blind spot position is considered, and the discontinuity of the blind spot position of the blind spot distribution map is quantitatively described;

[0071] The requirements of the blind spot distribution map for the discontinuity of the blind spot position are:

[0072] If a pulse is missing at a certain slant range unit, the adjacent pulses of the missing pulse are not expected to be missing;

[0073] The specific process is:

[0074] The variation rule of the PRI sequence uniquely determines the distribution of the blind area. The purpose of the design of the present invention is to hope that the blind area can be evenly distributed within the width of the surveying and mapping band.

[0075] The requirement of the blind zone distribution diagram for the discontinuity of the blind zone position is that if a pulse is missing at a certain slant range unit, the adjacent pulses of the missing pulse are not expected to be missing.

[0076] The blind spot distribution map is regarded as image I(i,j), i=1,2,…M, j=1,2,…N;

[0077] Where, M is the number of pulses; N is the number of slant range units (the number of slant ranges R0); (i, j) is the pixel point of the blind area distribution map;

[0078] The image abscissa j represents the slant range position of the blind area distribution diagram, and the image ordinate i represents the serial number of the transmission pulse causing the blind area in the blind area distribution diagram;

[0079] The coordinate position (i, j) of the blind spot distribution map corresponds to the image pixel unit (i, j), and when the i-th transmission pulse has a blind spot in the j-th slant range unit, the image I (i, j) pixel point value I (i, j) = 1; when the i-th transmission pulse has no blind spot in the j-th slant range unit, the image I (i, j) pixel point value I (i, j) = 0;

[0080] Then the total number of blind spots in image I(i,j) is

[0081]

[0082] The missing discontinuity measure can be implemented using linear spatial filtering.

[0083] In order to measure the number of blind spots in the neighborhood of a single blind spot (the i-th transmitted pulse appears in the j-th slant range unit as a single blind spot, that is, the pixel point corresponding to I(i,j)=1) in the blind spot distribution diagram, a spatial filter w(i,j) is defined:

[0084]

[0085] The response of the filter g(i,j) is the sum of the products of the filter coefficients w(d,s) (the filter w(i,j) is a matrix, and the values ​​w(d,s) in the matrix are called coefficients) and the image pixels surrounded by the filter:

[0086]

[0087] Where w(d,s) is the filter coefficient of the dth row and sth column, d is the row offset of pixel (i,j), s is the column offset of pixel (i,j), a is the maximum row offset of the filter, -a is the minimum row offset of the filter, b is the maximum column offset of the filter, -b is the minimum column offset of the filter, N a =(2a+1), N b =(2b+1), N a is the total number of rows of the filter, N b is the total number of columns of the filter;

[0088] Here, the design of the spatial filter only needs to consider the image ordinate y (the serial number of the transmitted pulse causing the blind spot), and the image abscissa x (the slant distance position where the blind spot appears) is irrelevant to the continuity of the missing pulse. Therefore, the values ​​of a and b are a=0. is a positive integer;

[0089] The range of the filter response is g(i,j)=u, u=0,1,…,N a N b -1;

[0090] The filter response value can reflect the number of blind areas near the pixel position where the blind area currently appears. The larger the filter response value, the more blind areas appear around, that is, the more pulses are lost.

[0091] g(i,j)=0 means that at the pixel point (i,j) where the blind spot appears, the cells (i±d,j±s) adjacent to the pixel point (i,j) where the blind spot appears will not appear, d=-a,…,0,…,a,s=-b,…,0,…,b;

[0092] The discontinuity of pulse loss hopes that the filter response value is as small as possible. For example, in an ideal case, for a column of data (data with the same slant distance), for all pixels in the blind area, g(i,j)=0, and g(i,j)=1 does not exist, ...N a N b -1. This ideal situation is actually difficult to meet, but it provides an optimization direction for sequence design. Therefore, it is necessary to further consider the statistical characteristics of the blind continuous distribution of g(i,j).

[0093] In order to measure the statistical characteristics of the continuous distribution of the blind area of ​​each column of data g(i,j) (same slant range unit), H(k,j) is defined to represent the number of occurrences of pixels in the jth column of the filter response that satisfy the conditions g(i,j)=u and I(i,j)==1, u=k-1=0,1,…N a N b -1 is the filter response value;

[0094] Then we have:

[0095]

[0096] Among them, the operator Indicates the number of pixels in the jth column of pixel data that satisfy the condition g(*,j)==k and the condition I(i,j)==1;

[0097] We can use the frequency of occurrence of elements in H(k,j) (H(k,j) is a matrix, the frequency of occurrence of each element in the matrix) to represent the probability distribution P(k,j), then:

[0098]

[0099] P(k,j) can measure the distribution of blind spots in the vicinity of the pixels where blind spots appear in the original image I(i,j), so as to measure the discontinuity of blind spot distribution;

[0100] In the formula, ε is an infinitesimal quantity;

[0101] When k = 1, the value of P(k, j) is large (k is the kth row of P(k, j) and is related to u); when k = 2, ..., N a N b When , the value of P(k,j) is small; at this time, the requirement of blind area distribution discontinuity can be guaranteed.

[0102] The other steps and parameters are the same as those in the first or second embodiment.

[0103] Specific implementation method 4: This implementation method is different from any one of specific implementation methods 1 to 3 in that, in step 3, based on the result of step 2, the uniformity of the blind spot position of the blind spot distribution map is quantitatively described by considering the requirement of the blind spot distribution map on the uniformity of the blind spot position;

[0104] The requirements of the blind spot distribution map for the uniformity of the blind spot position are:

[0105] In the blind area distribution diagram I(i,j), the proportion of blind areas (pulse loss) can be evenly distributed along the horizontal axis j (slant distance), that is, the element value of the kth row of P(k,j) can satisfy the uniform distribution;

[0106] The specific process is:

[0107] The uniformity here is reflected in the fact that the blind area distribution Q is uniformly distributed, that is, the probability distribution is the same when j=1, 2, ...N.

[0108] Taking into account the requirement of uniformity, we can let the expected probability distribution Q(k,j) be:

[0109]

[0110] (k is the kth row of Q(k,j), related to u)

[0111] The other steps and parameters are the same as those in Specific Embodiments 1 to 3.

[0112] Specific implementation method 5: This implementation method is different from any one of specific implementation methods 1 to 4 in that, in step 4, based on the quantitative description of the discontinuity of the blind area position of the blind area distribution map in step 2 and the quantitative description of the uniformity of the blind area position of the blind area distribution map in step 3, the target criterion function of the nonlinearly changing PRI sequence is defined; the specific process is:

[0113] The present invention defines a criterion to measure the difference between the expected probability distribution Q(k, j) and the existing probability distribution P(k, j), and uses it as a criterion function for improving the genetic algorithm.

[0114] In order to accurately measure the difference between the expected probability distribution Q(k,j) and the existing probability distribution P(k,j), with the help of the concept of relative entropy (asymmetric measure of the difference between two probability distributions), a target criterion function for measuring the blind spot distribution characteristics is defined:

[0115]

[0116] This criterion function can be used to measure the difference between the existing probability distribution P(k, j) and the expected probability distribution Q(k, j);

[0117] Criterion Cost Function The smaller the value, the smaller the difference, the closer the existing probability distribution is to the expected distribution, which means that the number of consecutive missing pulses in the blind area distribution is smaller, and the blind area distribution is more uniform along the distance dimension, which is conducive to the reconstruction of the azimuth data in subsequent signal processing.

[0118] The other steps and parameters are the same as those in Specific Embodiments 1 to 4.

[0119] The following examples are used to verify the beneficial effects of the present invention:

[0120] Embodiment 1:

[0121] according to Figure 2 The Staggered SAR nonlinear sequence design process shown in the figure realizes the optimization of the sequence. The parameters are set as PRI min =0.538ms,PRI max =0.685ms,R 0max =1097km,R 0max =868km, T p = 20us and the number of pulse cycles M = 273. The initial population is composed of multiple random PRI sequence individuals. The M PRI values ​​of each sequence individual can be in [PRI min ,PRI max ] are selected arbitrarily. The generated PRI sequence is evaluated using the optimization objective function criterion proposed in the present invention. The size parameter of the spatial filter can be set to N a =1 and N b =5 to measure the discontinuity and uniformity of the blind area distribution. The improved mutation operator can easily generate individuals with smaller target criterion function values. The optimization process based on the genetic algorithm searches along the direction of individuals with smaller target criterion functions. Finally, the optimized PRI periodic nonlinear change sequence is obtained as follows Figure 5(a) , 5(b) , 5(c), 5(d), and 5(e). To further illustrate the effectiveness of the optimized periodic nonlinear PRI sequence, we also give the results of the PRI random variation sequence and the PRI periodic linear fine variation sequence, as shown in Figure 3(a) , 3(b) , 3(c), 3(d), 3(e) and Figure 4(a) , 4(b), 4(c), 4(d), and 4(e). The PRI random change sequence selects one of the PRI sequence individuals in the initial population. The PRI periodic linear fine change sequence is composed of a long sequence spliced ​​from several fast-changing PRI sequences.

[0122] The discontinuity and uniformity of the blind area distribution can be intuitively reflected in the blind area distribution diagram. As shown in Figure 3(b) for the blind area distribution diagram of the random PRI sequence, there are many continuously lost pulses in some slant ranges, while there are few lost pulses in other slant ranges, which will lead to large differences in imaging performance in different slant ranges of the entire strip. As shown in Figures 3(b) and 4(b), the blind area distribution of the PRI periodic linear fine variation sequence and the PRI periodic nonlinear sequence both meet the discontinuity requirement, that is, there will be no continuous pulse loss in all slant ranges of interest. However, for the PRI periodic linear fine variation sequence, due to the existence of the linear variation law, the distribution of the pulse loss position still maintains regularity and periodicity, which will lead to high side lobes in the azimuth pulse response after data reconstruction. The blind area corresponding to the proposed PRI periodic nonlinear sequence is discontinuous, discrete, and uniformly distributed, which will bring potential advantages to subsequent reconstruction, that is, after reconstruction, no high side lobes will appear near the main lobe or far away from the main lobe in the azimuth pulse response. And the uniformity of the blind area distribution reduces the sensitivity of the azimuth pulse response to the target position.

[0123] The blind zone results of different PRI sequences are further analyzed, and the simulation results also give the relationship diagram of missing pulse percentage, pulse missing interval and pulse non-missing interval. It can be seen from Figure 3(c), Figure 4(c) and Figure 5(c) that the percentage of pulse missing of the random PRI sequence is much higher than that of the other two PRI sequences. In addition, compared with the results of Figure 3(c) and Figure 4(c), for the PRI periodic nonlinear sequence, the percentage of missing pulses in Figure 4(c) is more evenly distributed along the slant distance unit. Figure 3(d), Figure 4(d) and Figure 5(d) respectively give the maximum, average and minimum pulse separation. In Figure 3(d), the maximum pulse separation degree of different slant distances varies greatly. However, in Figure 4(d) and Figure 5(d), the difference in the distribution of pulse separation along the slant distance is significantly reduced. It can be seen from Figure 3(e), Figure 4(e) and Figure 5(e) that the average time interval without pulse missing and the minimum time interval without pulse missing of the PRI periodic linear fine change sequence and the PRI periodic nonlinear change sequence are also much lower than those of the random PRI sequence. The maximum pulse-free missing time interval of the PRI period nonlinear variation sequence is also much lower than that of the PRI period linear fine variation sequence.

[0124] The objective function criterion can quantitatively measure the discontinuity and uniformity of the blind area. Figure 6The curve represents the relationship between the target criterion function value and the number of iterations. The smaller the target criterion function value, the better the discontinuity and uniformity of the blind area distribution corresponding to the PRI sequence. Figure 5(a) , 5(b) , 5(c), 5(d), and 5(e), it can be seen that the objective function value of the random PRI sequence is as high as 8.934. For the PRI periodic linear fine change sequence, the value of the objective function is 0.2561, as shown by the dotted line mean in the figure. However, for the proposed PRI periodic nonlinear change sequence, the value of the objective function can be significantly reduced to 0.029. The results show that the PRI periodic nonlinear change sequence has significant advantages over other PRI sequences in terms of the uniformity and discontinuity of its blind area distribution.

[0125] The present invention may also have many other embodiments. Without departing from the spirit and essence of the present invention, those skilled in the art may make various corresponding changes and modifications based on the present invention, but these corresponding changes and modifications should all fall within the scope of protection of the claims attached to the present invention.

Claims

1. A PRI sequence design method based on the nonlinear change of the Staggered SAR system, characterized by: The specific process of the method is: Step 1: Based on the Staggered SAR system, calculate the blind area position corresponding to the changing PRI sequence and form a blind area distribution map corresponding to the PRI sequence; Step 2: Considering the requirements of the blind spot distribution map for the discontinuity of the blind spot position, quantitatively describe the discontinuity of the blind spot position of the blind spot distribution map; The requirements of the blind spot distribution map for the discontinuity of the blind spot position are: If a pulse is missing at a certain slant range unit, the adjacent pulses of the missing pulse are not expected to be missing; Step 3: Based on the result of step 2, considering the requirement of the blind spot distribution map for the uniformity of the blind spot position, quantitatively describe the uniformity of the blind spot position of the blind spot distribution map; The requirements of the blind spot distribution map for the uniformity of the blind spot position are: In the blind area distribution diagram I(i,j), the proportion of blind areas can be evenly distributed along the horizontal axis j, that is, the element value of the kth row of P(k,j) can satisfy the uniform distribution; Step 4: Based on the quantitative description of the discontinuity of the blind area position of the blind area distribution map in step 2 and the quantitative description of the uniformity of the blind area position of the blind area distribution map in step 3, define the target criterion function of the nonlinearly changing PRI sequence; Step 5: Based on the target criterion function obtained in step 4, a genetic algorithm is used to calculate and optimize the target criterion function for each PRI sequence to obtain the PRI sequence corresponding to the optimal target criterion function value.

2. The PRI sequence design method based on nonlinear changes of Staggered SAR system according to claim 1, characterized in that: In the step 1, based on the Staggered SAR system, the blind area position corresponding to the changed PRI sequence is calculated to form a blind area distribution map corresponding to the PRI sequence; the specific process is: In the Staggered SAR mode, there are M pulses in the transmit and receive sequence, and the mth pulse has a corresponding pulse repetition period of PRI. m , m=0,1,…,M-1; The Staggered SAR is a staggered synthetic aperture radar; Defining PRI max and PRI min They correspond to the maximum and minimum pulse repetition periods among the M pulses respectively; ΔPRI=PRI max -PRI min Indicates the variation range of the pulse repetition period PRI; The emission time of the mth pulse is t m The expression is In the formula, PRI p is the pulse repetition period of the pth pulse, and p is the pulse number; For the mth transmitted pulse, the arrival time of the mth pulse echo is the same as that of the (m+n)th transmitted pulse, which causes the mth pulse to be lost in the nth blind area. The target echo falling in the nth blind area cannot be received because the transmission and reception cannot be simultaneous, which is expressed as Among them, t r is the echo delay time, PRI i is the pulse repetition period of the ith pulse, T p is the pulse width; According to formula (2), the echo delay time t r The range of t r =2R0 / c to solve the range of slant distance R0 corresponding to the blind area, so as to obtain all blind area positions of the mth transmitting pulse and form a blind area distribution map; Where c is the speed of light.

3. The PRI sequence design method based on nonlinear changes of Staggered SAR system according to claim 2, characterized in that: In the step 2, the requirement of the blind spot distribution map on the discontinuity of the blind spot position is taken into consideration, and the discontinuity of the blind spot position of the blind spot distribution map is quantitatively described; The requirements of the blind spot distribution map for the discontinuity of the blind spot position are: If a pulse is missing at a certain slant range unit, the adjacent pulses of the missing pulse are not expected to be missing; The specific process is: The blind spot distribution map is regarded as image I(i,j), i=1,2,…M, j=1,2,…N; Where, M is the number of pulses; N is the number of slant range units; (i, j) is the pixel point of the blind area distribution map; The image abscissa j represents the slant range position of the blind area distribution diagram, and the image ordinate i represents the serial number of the transmission pulse causing the blind area in the blind area distribution diagram; The coordinate position (i, j) of the blind spot distribution map corresponds to the image pixel unit (i, j), and when the i-th transmission pulse has a blind spot in the j-th slant range unit, the image I (i, j) pixel point value I (i, j) = 1; when the i-th transmission pulse has no blind spot in the j-th slant range unit, the image I (i, j) pixel point value I (i, j) = 0; In order to measure the number of blind spots in the neighborhood of a single blind spot in the blind spot distribution map, a spatial filter w(i,j) is defined: The response of a filter g(i,j) is the sum of the products of the filter coefficients w(d,s) and the image pixels enclosed by the filter: Where w(d,s) is the filter coefficient of the dth row and sth column, d is the row offset of pixel (i,j), s is the column offset of pixel (i,j), a is the maximum row offset of the filter, -a is the minimum row offset of the filter, b is the maximum column offset of the filter, -b is the minimum column offset of the filter, N a =(2a+1), N b =(2b+1), N a is the total number of rows of the filter, N b is the total number of columns of the filter; The value of a,b is a=0, is a positive integer; The range of the filter response is g(i,j)=u, u=0,1,…,N a N b -1; g(i,j)=0 means that at the pixel point (i,j) where the blind spot appears, the cells (i±d,j±s) adjacent to the pixel point (i,j) where the blind spot appears will not appear, d=-a,…,0,…,a,s=-b,…,0,…,b; Define H(k,j) to represent the number of occurrences of pixels in the jth column of the filter response data that satisfy the conditions g(i,j)=u and I(i,j)==1, u=k-1=0,1,…N a N b -1 is the filter response value; Then we have: Among them, the operator Indicates the number of pixels in the jth column of pixel data that satisfy the condition g(*,j)==k and the condition I(i,j)==1; The probability distribution P(k,j) is represented by the frequency of occurrence of elements in H(k,j) (H(k,j) is a matrix, the frequency of occurrence of each element in the matrix), then: In the formula, ε is an infinitesimal quantity; When k = 1, the value of P(k, j) is large; when k = 2, ..., N a N b When , the value of P(k,j) is small; at this time, the requirement of blind area distribution discontinuity can be guaranteed.

4. The PRI sequence design method based on nonlinear changes of Staggered SAR system according to claim 3 is characterized by: In the step 3, based on the result of the step 2, the uniformity of the blind spot positions of the blind spot distribution map is quantitatively described taking into account the requirement of the blind spot distribution map on the uniformity of the blind spot positions; The requirements of the blind spot distribution map for the uniformity of the blind spot position are: In the blind area distribution diagram I(i,j), the proportion of blind areas can be evenly distributed along the horizontal axis j, that is, the element value of the kth row of P(k,j) can satisfy the uniform distribution; The specific process is: Let the expected probability distribution Q(k,j) be:

5. The PRI sequence design method based on nonlinear changes of Staggered SAR system according to claim 4 is characterized by: In step 4, based on the quantitative description of the discontinuity of the blind area position of the blind area distribution map in step 2 and the quantitative description of the uniformity of the blind area position of the blind area distribution map in step 3, the target criterion function of the nonlinearly changing PRI sequence is defined; the specific process is: Define a target criterion function to measure the blind area distribution characteristics This criterion function can be used to measure the difference between the existing probability distribution P(k,j) and the expected probability distribution Q(k,j).

Images