Pulse pair blur removal method based on statistical histogram of moving platform

Abstract

The present invention discloses a motion platform-based statistical histogram pulse pair defuzzification method, which primarily addresses the high computational complexity and low real-time performance issues of existing statistical histogram pulse pair defuzzification methods. The implementation scheme comprises the following steps: providing a primary motion platform and multiple auxiliary motion platforms, measuring the arrival time differences of received target pulses, and pairing them within a time difference window to form a time difference pair dataset; calculating a time difference boundary threshold based on the movement distance of the auxiliary motion platforms within the observation time; classifying the time difference pairs in the time difference pair dataset that meet the time difference boundary threshold using a statistical histogram method; setting a quantity threshold based on the number of target pulses, and using the quantity threshold to screen the classified results, selecting true time difference pairs and eliminating false time difference pairs, thereby completing pulse pair defuzzification. The present invention effectively reduces the system computational complexity and improves the system's real-time performance, and can be applied to aerospace and electronic warfare.

Description

Technical Field

[0001] The present invention belongs to the technical field of passive positioning, and further relates to a pulse pair ambiguity elimination method, which can be applied to aerospace and electronic warfare. Background Art

[0002] Pulse pair ambiguity is a matching ambiguity phenomenon caused by the high-repetition-rate signal of the target radiation source. This pulse pair ambiguity not only wastes system computing resources but also makes it impossible to locate the target. Therefore, eliminating pulse pair ambiguity has become a key issue that needs to be solved urgently.

[0003] In 1999, Yang Lin, Sun Zhongkang, and others analyzed the causes of pulse pair ambiguity and proposed a multi-level statistical histogram method to eliminate pulse pair ambiguity. Their implementation involves: 1) buffering and pairing pulses to obtain time difference pairs and forming a statistical histogram of these time difference pairs; 2) selecting the time difference pair corresponding to the largest peak in the statistical histogram as a true target time difference pair and eliminating the pulses that formed this time difference pair; and 3) repeating steps 1) and 2) for the remaining pulses until no more pulses remain. Using the multi-level statistical histogram method to eliminate pulse pair ambiguity requires a high computational load, high requirements for time difference measurement accuracy, and a low success rate for eliminating pulse pair ambiguity.

[0004] To address these shortcomings, in 2004, Li Tao, Jiang Wenli, and others made improvements. Their implementation scheme is as follows: 1) Using a statistical histogram method, a table of pulse overlaps between time difference peaks is generated, time difference peaks are extracted, and fuzzy time difference pairs and true time difference pairs are obtained based on the time difference peaks; 2) Based on the relationship between the fuzzy time difference pairs and the true time difference pairs, possible time differences are calculated within the time difference window; 3) The fuzzy time difference pairs and the true time difference pairs are used for positioning to obtain fuzzy positioning points and true positioning points. Based on the difference in the divergence characteristics between the true and fuzzy positioning points, the true positioning points are retained and the fuzzy positioning points are eliminated, thereby achieving pulse pair ambiguity removal. Although this method has low requirements for time difference measurement accuracy and a high success rate in removing pulse pair ambiguity, it requires multiple observations and can only remove pulse pair ambiguity during the positioning process, resulting in a waste of computing resources and reducing the real-time performance of the multi-station time difference positioning system. Summary of the Invention

[0005] The purpose of the present invention is to address the deficiencies of the above-mentioned prior art and propose a method for eliminating ambiguity of pulse pairs based on a statistical histogram of a moving platform, so as to reduce the computational complexity, lower the requirements for the time difference measurement accuracy, and improve the real-time performance of the time difference positioning system.

[0006] To achieve the above objectives, the technical solutions of the present invention include the following:

[0007] 1) Set up a main motion platform and M auxiliary motion platforms. During the observation time, the main and auxiliary motion platforms measure the arrival time difference of the received target pulses and pair the time differences within the time difference window to form a time difference pair dataset;

[0008] 2) According to the movement distance Δd of the auxiliary motion platform during the observation time, the time difference boundary threshold G is derived and calculated:

[0009] G=Δd / c

[0010] Where c represents the speed of light;

[0011] 3) Using the statistical histogram method, the time difference pairs in the time difference pair data set that meet the time difference boundary threshold are classified into one category. The formula is as follows:

[0012] |TDOA i -TDOA j |≤G

[0013] Among them, TDOA i and TDOA j Respectively represent the i-th time difference pair and the j-th time difference pair in the time difference pair data set;

[0014] 4) A quantity threshold g is set according to the target pulse number n, and the classified results are screened using the quantity threshold to select the true time difference pairs and eliminate the ambiguous time difference pairs, thus achieving pulse pair ambiguity elimination.

[0015] Compared with the prior art, the present invention has the following advantages:

[0016] First, the present invention derives a time difference boundary threshold value and uses the time difference boundary threshold value to classify and form a first-level histogram, thereby avoiding the scheme of classifying and forming a multi-level histogram in the prior art and effectively reducing the amount of calculation of the system;

[0017] Secondly, the present invention directly eliminates pulse pair ambiguity by using the time difference boundary threshold value, thereby avoiding the disadvantage of the prior art of eliminating pulse pair ambiguity through multiple positioning, effectively avoiding the waste of system resources, and improving the real-time performance of the system. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] Figure 1 It is an implementation flow chart of the present invention;

[0019] Figure 2 is a schematic diagram of the motion model in the present invention;

[0020] Figure 3 This is a graph showing the effect of the target pulse number on the success rate of blur removal using the present invention;

[0021] Figure 4This is a result diagram of the target pulse loss probability versus ambiguity elimination success rate simulated by the present invention. DETAILED DESCRIPTION

[0022] The embodiments and effects of the present invention are described in further detail below with reference to the accompanying drawings.

[0023] Reference Figure 1 , the implementation steps of this example are as follows:

[0024] Step 1: Set up a main motion platform and M auxiliary motion platforms, where M ≥ 2, and calculate the arrival time difference of the target pulse, such as Figure 2 shown.

[0025] 1.1) The main motion platform and M auxiliary motion platforms receive the pulse signal emitted by the target. Assume that the target pulse signals received by the main motion platform and the mth auxiliary motion platform are x(t) and y(t) respectively. m (t), and the cross-correlation function R of the two moving platforms is obtained xy (τ) is:

[0026] R xy (τ)=E[x(t+τ)y m (t)] <1>

[0027] Where E[·] represents the mathematical expectation, τ represents the relative movement time, and m = 1, 2, …, M;

[0028] 1.2) R xy (τ) is solved to obtain the cross-correlation curve, and the τ value at its maximum value is used as the time difference dτ between the target pulse arriving at the main motion platform and the mth auxiliary motion platform m .

[0029] In this example, it is assumed but not limited to M=3.

[0030] Step 2: Calculate the time difference window between the main and auxiliary motion platforms.

[0031] According to the position distance d between the main motion platform and the mth auxiliary motion platform m , calculate the time difference window w m :

[0032] w m =d m / c <2>

[0033] Where c represents the speed of light.

[0034] Step 3: Perform time difference pairing within the time difference window to construct a time difference pair dataset.

[0035] 3.1) For the k0 target pulses received by the main motion platform and the k target pulses received by the mth auxiliary motion platformm Target pulse mode <1> Get N m cross-correlation functions, where N m =k0*k m ;

[0036] 3.2) For N m Solve the cross-correlation function and obtain the time difference dτ m p :

[0037] dτ m p =[dτ1,dτ2,…,dτ Nm ] <3>

[0038] 3.3) Using the time difference window w m Time difference dτ m p To filter, follow -w m ≤dτ mi p ≤w m The time difference window w is selected based on the conditions m The time difference within the range is obtained to obtain the filtered time difference dτ m ;

[0039] 3.4) The time difference between the main motion platform and the M auxiliary motion platforms is dτ1, dτ2, ... dτ M ,according to Pair them in a way to form a time difference pair dataset D, where represents the Kronecker product.

[0040] Step 4: Derived and calculated the time difference boundary threshold value G according to the moving distance of the auxiliary motion platform.

[0041] Reference Figure 2 , assuming that when the main motion platform is at position s 01 , the auxiliary motion platform is at position s 02 When both receive the same target pulse, the target pulse arrival times are t 01 and t 02 After a time ΔT, the main motion platform and the auxiliary motion platform move to position A and position B respectively. At this time, the target emits the second pulse signal. When the main motion platform is at position s 11 , the auxiliary motion platform is in position s 12 The second pulse signal is received at t 11 and t 12 , the moving distances of the two motion platforms are Δd0 and Δd1 respectively, d1 and d2 represent the distances between the target and the main motion platform, and d1' and d2' represent the distances between the target and the auxiliary motion platform, then t11 and t 12 The calculation formula is as follows:

[0042]

[0043] By <4> And the triangular relationship shows that:

[0044]

[0045] Depend on <5> Available

[0046]

[0047] Pair <6> Deformation, add t on both sides 12 -t 11 have to

[0048]

[0049] According to the order in which the target pulses arrive at the main motion platform, the following relationship is obtained:

[0050] -(Δd0 / c+ΔT)<t 01 -t 11 <-ΔT <8>

[0051] General <8> Substitution <7> The following relationship is obtained:

[0052] (t 12 -t 11 )-(t 02 -t 01 )<Δd1 / c <9>

[0053] Imperative <9> Here, Δd1 / c=G, and G is called the time difference boundary threshold.

[0054] Step 5: Use the statistical histogram method to classify the time difference pairs in the time difference pair data set D that meet the time difference boundary threshold G into one category.

[0055] 5.1) Select a time difference pair from the time difference pair dataset D as the center point, create an index array corresponding to it, remove the time difference pair from the time difference pair dataset D, and add it to the set S, which is initially empty;

[0056] 5.2) Take the time difference pair TDOA from the time difference pair dataset D j , take the center point TDOA from the set S i and its corresponding index array a i , judge TDOA j With TDOA i Whether it meets the time difference boundary threshold G:

[0057] If TDOA j Meet | TDOA j -TDOA i |≤G, then add subscript index j to a i and TDOA j Remove from the time difference pair data set;

[0058] Otherwise, traverse the remaining center points in the set S and execute 5.3);

[0059] 5.3) TDOA j Compare with the remaining center points and the time difference boundary threshold G:

[0060] If TDOA j and center point TDOA k Meet | TDOA j -TDOA k |≤G, then add subscript index j to a k and TDOA j Remove from the time difference pair data set;

[0061] Otherwise, TDOA j Treat it as the center point, add it to the set S, and create the corresponding index array a j , and TDOA j Removed from the time difference pair dataset.

[0062] Step 6: Use the quantity threshold to filter the classified results to achieve pulse pair ambiguity elimination.

[0063] 6.1) Set the quantity threshold g according to the target number of pulses n:

[0064] g=0.75*n <10>

[0065] 6.2) According to the index array a of the qth center point in the set S q , calculate its length l q , where q = 1, 2, ..., Q, Q represents the number of central points in the set S;

[0066] 6.3) Traverse all the center points in the set S and calculate the length of the index array to obtain the length array L = [l1, l2, ... l q ,,l Q ];

[0067] 6.4) Traverse the length array L and compare each element in the length array L with the quantity threshold g:

[0068] If l q ≥g, then a qCorresponding to the real time difference pairs, and extracting the real time difference pairs from the time difference pair dataset;

[0069] Otherwise, a q The corresponding false time difference pairs are removed from the time difference pair data set to eliminate ambiguity.

[0070] The effect of the present invention can be further illustrated by the following simulation experiments:

[0071] 1. Simulation conditions

[0072] Condition 1: Set up a main motion platform and three auxiliary motion platforms. The arrival time measurement error of the main and auxiliary motion platforms is 20ns, and the synchronization error of the main and auxiliary motion platforms is 10ns. Set up a target radiation source of a frequency agile radar with four working modes. The number of target pulses received by the motion platform is 30, 50, 100, 200, and 300 respectively. According to the formula <10> The number thresholds were set respectively so that the target number of pulses would not be lost during the receiving process of the moving platform, and 100 independent simulation experiments were performed.

[0073] The position parameters of the motion platform are shown in Table 1.

[0074] The target position parameters are shown in Table 2.

[0075] The four operating modes of frequency agile radar are shown in Table 3:

[0076] Table 1 Motion platform position parameters (unit: km)

[0077] Sports platform name x y z Main motion platform -3212 5547 2749 Auxiliary motion platform 1 -2608 6071 2242 Auxiliary motion platform 2 -3503 5632 2167 Auxiliary motion platform 3 -4027 5028 2681

[0078] Table 2 Target location parameters (unit: km)

[0079] x y z Target -2969 5030 2551

[0080] Table 3 Working mode (unit: us)

[0081] Working Mode Pulse repetition period Working Mode 1 200 Working Mode 2 463 Working Mode 3 600 Working Mode 4 784

[0082] Condition 2: Set up a primary motion platform and three auxiliary motion platforms. The positional relationships are as shown in Table 1. The arrival time measurement error of the primary and auxiliary motion platforms is 20 ns, and the synchronization error is 10 ns. Set up a frequency-agile radar target emitter with four operating modes. The target positions and operating modes are shown in Tables 2 and 3, respectively. The number of target pulses received by the motion platform is 300, with a threshold of 225. The probability of target pulse loss during the motion platform reception is 1%, 3%, 5%, 7%, and 10%, respectively. 100 independent simulation experiments are conducted.

[0083] 2. Simulation content

[0084] Simulation 1: Under the above condition 1, the pulse pair fuzziness elimination simulation is performed for different target pulse numbers using the method of the present invention, and the influence curve of the target pulse number on the fuzziness elimination success rate is obtained, as shown in FIG. Figure 3 shown.

[0085] from Figure 3 As can be seen from the figure, when the number of pulses reaches 50, pulse pair ambiguity removal can be achieved with a high probability of 95%. When the number of pulses reaches 100, pulse pair ambiguity removal is guaranteed to be successful. This shows that the present invention can effectively reduce the system computation load and improve the system real-time performance when the target number of pulses is only 100.

[0086] Simulation 2: Under the above condition 2, the pulse pair ambiguity elimination simulation is performed for different pulse loss probabilities using the present invention, and the influence curve of the target loss probability on the ambiguity elimination success rate in the present invention is obtained, as shown in FIG. Figure 4 shown.

[0087] from Figure 4 It can be seen from the figure that even when the target pulse loss probability is 10%, the method of the present invention can still ensure that the pulse pair ambiguity is completely eliminated, indicating that the method of the present invention has good anti-pulse loss performance.

[0088] The above simulation results show that the present invention can effectively reduce the computational complexity of the system, improve the real-time performance of the system, and has good anti-pulse loss performance.

Claims

1. A method for eliminating blurring of pulse pairs based on statistical histogram of a motion platform, characterized in that: The steps include: 1) Set up a main motion platform and M auxiliary motion platforms. During the observation time, the main and auxiliary motion platforms measure the arrival time difference of the received target pulses and pair the time differences within the time difference window to form a time difference pair data set. This is achieved as follows: 1.1) According to the distance d between the main motion platform and the mth auxiliary motion platform m , get the time difference window w m : w m =d m / c 1.2) Measure the time difference between the target pulse received by the main motion platform and the target pulse received by the mth auxiliary motion platform to obtain the time difference dτ m p =[dτ1,dτ2,…,dτ N ], and use the time difference window w m Time difference dτ m p To filter, follow -w m ≤dτ mi p ≤w m The time difference window w is selected based on the conditions m The time difference within the range is obtained to obtain the filtered time difference dτ m ; 1.3) The time difference between the main motion platform and the M auxiliary motion platforms is dτ1, dτ2, ... dτ M ,according to , forming a time difference pair data set, where represents the Kronecker product; 2) According to the moving distance △d of the auxiliary motion platform during the observation time, the time difference boundary threshold G is derived and calculated: G=△d / c Where c represents the speed of light; 3) Using the statistical histogram method, the time difference pairs in the time difference pair data set that meet the time difference boundary threshold are classified into one category. The formula is as follows: |TDOA i -TDOA j |≤G Among them, TDOA i and TDOA j Respectively represent the i-th time difference pair and the j-th time difference pair in the time difference pair data set; 4) A quantity threshold g is set according to the target pulse number n, and the classified results are screened using the quantity threshold to select the true time difference pairs and eliminate the ambiguous time difference pairs, thus achieving pulse pair ambiguity elimination.

2. The method according to claim 1, characterized in that The arrival time difference of the target pulses received by the main and auxiliary motion platforms during the observation time in step 1) is achieved as follows: 1a) Assume that the target pulse signals received by the main motion platform and the mth auxiliary motion platform are x(t) and y(t) respectively. m (t), and the cross-correlation function R of the two moving platforms is obtained xy (τ) is: R xy (τ)=E[x(t+τ)y m (t)] Where E[·] represents the mathematical expectation, and τ represents the relative movement time; 1b) R xy (τ) is solved to obtain the cross-correlation curve, and the τ value at its maximum value is used as the time difference dτ between the target pulse arriving at the main motion platform and the mth auxiliary motion platform m .

3. The method according to claim 1, characterized in that In step 3), the time difference pairs that meet the time difference boundary threshold in the time difference pair data set are classified into one category using the statistical histogram method, which is implemented as follows: 3a) Select a time difference pair from the time difference pair dataset as the center point, create an index array corresponding to it, then remove the time difference pair from the time difference pair dataset and add it to the set S that is initially empty; 3b) Get the time difference pair TDOA from the time difference pair dataset j , take the center point TDOA from the set S i and its corresponding index array a i , judge TDOA j With TDOA i Whether it meets the time difference boundary threshold G: If TDOA j Meet | TDOA j -TDOA i |≤G, then add subscript index j to a i In TDOA j Remove from the time difference pair data set; Otherwise, traverse the remaining center points in the set S and execute 3c); 3c) Determine TDOA j Whether the time difference boundary threshold G is met with other center points; If TDOA j TDOA with center point k Meet | TDOA j -TDOA k |≤G, then add subscript index j to a k In TDOA j Remove from the time difference pair data set; Otherwise, TDOA j Consider it as the center point, add it to the set S, and create the corresponding index array a j , and TDOA j Removed from the time difference pair dataset.

4. The method according to claim 1, wherein In step 4), the quantity threshold g is set according to the target number of pulses n, and its formula is as follows: g=0.75*n.

5. The method according to claim 1, wherein Step 4) Filter the classified results using the quantity threshold, as follows: 4a) According to the index array a of the qth center point in the set S q , calculate its length l q , where q = 1, 2, ..., Q, Q represents the number of center points in the set S, traverse all center points in the set S and calculate the length of the index array, and get L = [l1, l2, ..., l q ,…,l Q ]; 4c) Traverse L and convert l q Compare with the quantity threshold g: If l q ≥g, then a q Corresponding to a real time difference pair, and extracting the real time difference pair from the time difference pair dataset; Otherwise, a q The corresponding fuzzy time difference pair is removed from the time difference pair dataset.

Images