CN114660560B - Pulse repetition interval sorting method based on equivalent DTOA density curve - Google Patents
Pulse repetition interval sorting method based on equivalent DTOA density curve Download PDFInfo
- Publication number
- CN114660560B CN114660560B CN202210242468.6A CN202210242468A CN114660560B CN 114660560 B CN114660560 B CN 114660560B CN 202210242468 A CN202210242468 A CN 202210242468A CN 114660560 B CN114660560 B CN 114660560B
- Authority
- CN
- China
- Prior art keywords
- equivalent
- time difference
- dtoa
- pri
- curve
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Images
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01S—RADIO DIRECTION-FINDING; RADIO NAVIGATION; DETERMINING DISTANCE OR VELOCITY BY USE OF RADIO WAVES; LOCATING OR PRESENCE-DETECTING BY USE OF THE REFLECTION OR RERADIATION OF RADIO WAVES; ANALOGOUS ARRANGEMENTS USING OTHER WAVES
- G01S7/00—Details of systems according to groups G01S13/00, G01S15/00, G01S17/00
- G01S7/02—Details of systems according to groups G01S13/00, G01S15/00, G01S17/00 of systems according to group G01S13/00
- G01S7/41—Details of systems according to groups G01S13/00, G01S15/00, G01S17/00 of systems according to group G01S13/00 using analysis of echo signal for target characterisation; Target signature; Target cross-section
- G01S7/418—Theoretical aspects
Landscapes
- Engineering & Computer Science (AREA)
- Computer Networks & Wireless Communication (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Radar, Positioning & Navigation (AREA)
- Remote Sensing (AREA)
- Radar Systems Or Details Thereof (AREA)
Abstract
The invention belongs to the technical field of radar signal PRI sorting, and particularly relates to a pulse repetition interval sorting method based on an equivalent DTOA density curve, which comprises the steps of obtaining arrival time differences among radiation source pulse signals, and establishing an arrival time difference sequence change curve through magnitude sorting; constructing an equivalent arrival time difference density curve which is equivalent to the arrival time difference probability density distribution curve and is used for representing an extreme value of the arrival time difference probability density distribution curve by using the sequence change curve; and extracting pulse repetition intervals PRIs by detecting the peak value of the equivalent arrival time difference density curve according to a preset threshold value. The invention changes angles, solves the sorting problem of the PRI jittering staggered signals by obtaining the equivalent DTOA density curve with similar characteristics to the DTOA probability density distribution curve, improves the sorting precision of the pulse signals and the radar reconnaissance technical level, and is convenient for the application of actual scenes.
Description
Technical Field
The invention belongs to the technical field of radar signal PRI sorting, and particularly relates to a pulse repetition interval sorting method based on an equivalent DTOA density curve.
Background
The most central technology of the current radar reconnaissance equipment is to correctly sort and identify pulse signals of each radiation source from a high-density and large-quantity pulse stream. Sorting of Pulse Repetition Interval (PRI) is mainly performed by using a Sequence Difference Histogram (SDIF) algorithm and a PRI transformation method. When the SDIF and PRI conversion methods are used to process PRI jitter, DTOA (Differential Time Of Arrival) is distributed in a plurality Of PRI bins near the real PRI bin, which causes the frequency in the real PRI bin to be lower than the threshold. For PRI jitter, the classic improvement strategy of SDIF and PRI transformation methods is to use overlapping windows to perform histogram statistics. However, this strategy has some problems, one of which is that the peak value of the position where the real PRI is located is widened, which results in low resolution of PRI jitter; secondly, the setting of the jitter rate and the length of the overlapping window has a relatively strict corresponding relationship, so that the interleaving conditions of the pulse trains with different jitter rates cannot be effectively processed; thirdly, under the condition that the PRI fixed pulse train and the PRI jittering pulse train are staggered, the peak value broadening enables the PRI estimation precision of the PRI fixed pulse train to be seriously reduced; in practical situations, the threshold coefficient is often required to be continuously adjusted manually, and the time and the efficiency of the threshold coefficient are difficult to meet the practical requirements.
Disclosure of Invention
Therefore, the invention provides a pulse repetition interval sorting method based on an equivalent DTOA density curve, which is based on deeper essence, solves the sorting problem of the PRI jittering staggered signals by acquiring the equivalent DTOA density curve with similar characteristics to the DTOA probability density distribution curve, improves the sorting precision of the pulse signals and the radar reconnaissance technical level, and is convenient for practical scene application.
According to the design scheme provided by the invention, the pulse repetition interval sorting method based on the equivalent DTOA density curve comprises the following steps:
acquiring arrival time difference between radiation source pulse signals, and establishing an arrival time difference sequence change curve between the signals through magnitude sorting;
constructing an equivalent arrival time difference density curve which is equivalent to the arrival time difference probability density distribution curve and is used for representing an extreme value of the arrival time difference probability density distribution curve by using the sequence change curve;
and extracting pulse repetition intervals PRIs by detecting the peak value of the equivalent arrival time difference density curve according to a preset threshold value.
As the pulse repetition interval sorting method based on the equivalent DTOA density curve, further, in establishing the arrival time difference sequence change curve, firstly, the arrival time difference values are reordered from small to large; and then, establishing a sequence change curve of the arrival time difference by taking the serial number as an abscissa and the numerical value of the arrival time difference as an ordinate.
As the pulse repetition interval sorting method based on the equivalent DTOA density curve of the present invention, further, in constructing the equivalent arrival time difference density curve, first, the arrival time difference value is preprocessed, the preprocessing at least includes: calculating the difference value of the numerical value of the adjacent arrival time difference, and subtracting the difference value obtained by calculation from the maximum value of the arrival time difference; then, an equivalent arrival time difference density curve is obtained by performing mean filtering on the preprocessing result.
As the pulse repetition interval sorting method based on the equivalent DTOA density curve, the point position for performing mean value filtering on the preprocessing result is further obtained according to the set adjustable parameter and the length of the staggered pulse train.
As the pulse repetition interval sorting method based on the equivalent DTOA density curve of the present invention, further, the equivalent arrival time difference density curve is represented as C 2 ＝{( dtoa i 1, 2.. L-1}, wherein dtoa (i)) | i =1,2 i Representing the ith time difference of arrival value; the mean filtering process is represented as:
As the pulse repetition interval sorting method based on the equivalent DTOA density curve, further, in the pulse repetition interval PRIS, a detection threshold is determined by using a self-adaptive strategy, and a peak value area of the equivalent arrival time difference density curve is detected by using a continuous threshold.
As the pulse repetition interval sorting method based on the equivalent DTOA density curve, the detection threshold T is further determined by a self-adaptive strategy p The process of (a) is represented as:
wherein λ is an adjustable parameter.
As the pulse repetition interval sorting method based on the equivalent DTOA density curve of the present invention, further, the extraction process of the pulse repetition interval PRIs is represented as:
therein, index i For peak position index, D (-) is the sequence of arrival time difference values, and L is the weighted window length parameter.
As the pulse repetition interval sorting method based on the equivalent DTOA density curve, further, when a plurality of radiation sources exist, the arrival time difference of a preset grade is accumulated and calculated firstly by combining the cumulative difference histogram algorithm, and the corresponding equivalent arrival time difference density curve is established; and then extracting pulse repetition interval PRIs, and performing pulse sequence retrieval on the pulse repetition interval PRI fixation and the pulse repetition interval PRI jitter by using a direct sequence retrieval method.
As the pulse repetition interval sorting method based on the equivalent DTOA density curve, further, in the direct sequence retrieval method, after a plurality of pulses are continuously retrieved from a reference pulse, the pulses are retrieved from the head and the tail of the last retrieved pulse to two sides simultaneously until the pulses cannot be retrieved at the preset multiple pulse repetition interval, and when the length of the retrieved pulse sequence is larger than or equal to the preset value, the pulse sequence retrieval is determined to be successful.
The invention has the beneficial effects that:
different from the method for obtaining the approximate shape of the DTOA probability density distribution curve by accumulating the DTOA frequency in the statistical interval in the histogram method, the invention starts from deeper essence, extracts PRIs and converts the PRIs into extracting extreme points of the DTOA probability density distribution curve, solves the sorting problem of the PRI jittering staggered signals by extracting the extreme points of the DTOA probability density distribution curve and further adopting a self-adaptive threshold, solves the problem of transition dependence on manual work, improves the sorting efficiency, the sorting accuracy and the radar detection level and has better application prospect.
Description of the drawings:
FIG. 1 is a schematic flow chart of a pulse repetition interval sorting method based on an equivalent DTOA density curve in an embodiment;
FIG. 2 is a schematic diagram showing the DTOA sequence variation in the example;
FIG. 3 is a graph showing the equivalent DTOA density under different conditions in the examples;
FIG. 4 is a schematic diagram of a PRI estimation accuracy variation curve under different algorithms in the embodiment;
FIG. 5 is a graph showing the equivalent DTOA density under different pulse loss rate conditions in the example;
FIG. 6 is a comparison of sorting performance of different algorithms under the PRI jitter staggered burst experiment in the example;
FIG. 7 is a schematic diagram showing the comparison of sorting performance of different algorithms in the experiment of pulse trains staggered by the fixed PRI and the jittering PRI in the embodiment;
FIG. 8 is a comparison of sorting performance of different algorithms for PRI dithering implementation in the examples.
The specific implementation mode is as follows:
in order to make the objects, technical solutions and advantages of the present invention clearer and more obvious, the present invention is further described in detail below with reference to the accompanying drawings and technical solutions.
Aiming at the problem of sorting the PRI jittering staggered radar signals, the traditional PRI sorting method usually adopts an overlapping window strategy, but the strategy cannot effectively cope with complex staggered conditions such as various different jittering rate pulse string staggered conditions, PRI fixed and PRI jittering pulse string staggered conditions and the like; meanwhile, the problem that the threshold coefficient needs to be manually debugged when PRIs are extracted also needs to be solved. To this end, an embodiment of the present invention provides a pulse repetition interval sorting method based on an equivalent DTOA density curve, as shown in fig. 1, which includes the following steps:
s101, obtaining arrival time difference between radiation source pulse signals, and establishing a sequence change curve of the arrival time difference between the signals through magnitude sorting;
s102, constructing an equivalent arrival time difference density curve which is equivalent to the arrival time difference probability density distribution curve and is used for representing an extreme value of the arrival time difference probability density distribution curve by using the sequence change curve;
s103, extracting pulse repetition intervals PRIs by detecting the peak value of the equivalent arrival time difference density curve according to a preset threshold value.
And starting from deeper essence, extracting PRIs and converting into extracting extreme points of a DTOA probability density distribution curve, and extracting the extreme points of the DTOA probability density distribution curve.
In a Sequence Difference Histogram (SDIF) algorithm, a Difference between Arrival Times (TOA) Of two adjacent pulses is calculated to obtain a Difference Histogram and calculate a threshold, then sub-harmonic detection is carried out, and if only one value exceeds the detection threshold, the value is taken as a possible PRI to carry out Sequence search; otherwise, calculating the difference value histogram of the next stage. For two-level and above histograms, if more than one peak exceeds the threshold, a sequence search is performed from the minimum pulse interval corresponding to the peak exceeding the threshold after sub-harmonic inspection. The thresholds of which are shown below
T 1 (τ)＝x(E-τ)exp(-τ/gN) (1)
Wherein E is the total number of pulses, and C is the number of levels of the difference histogram; g is a normal number less than 1; n is the total scale value of the pulse interval on the histogram; the constant x is determined experimentally. The SDIF has the advantages of small calculation amount and easy realization, but the method is easily interfered by sub-harmonics, and the original algorithm of the SDIF has no sorting capability on jittering PRI.
In the PRI transformation method, the range of PRI [ tau ] to be investigated min ,τ max ]Divided into a plurality of cells, called PRI boxes, centred at τ k K =1, 2.. K, width b k ，t n (N =1, 2.., N-1) is the pulse arrival time, and the discrete form of the PRI spectrum is shown below
The threshold form is shown below
Wherein T is the total observation time, C k For autocorrelation integration without complex phase, ρ is the pulse density, and α, β, γ are three adjustable coefficients. The PRI transform method is advantageous in that it can eliminate interference of sub-harmonics, but the amount of calculation thereof is significantly increased compared to SDIF, and the original PRI transform algorithm has no sorting capability for jittered PRI. The modified PRI transform algorithm has the ability to sort jittered PRIs, but due to the overlapping windows, the PRI resolution problem is also caused.
In the embodiment of the scheme, different from a histogram method, an approximate shape of a DTOA probability density distribution curve is obtained by accumulating DTOA frequency in a statistical interval, and an equivalent DTOA density curve with similar characteristics to the DTOA probability density distribution curve is utilized, so that the equivalent DTOA density curve can represent extreme points of the DTOA probability density distribution curve, the problem to be solved by PRI sorting is better fitted, and the application in a practical scene is facilitated. Further, in establishing a time difference of arrival sequence change curve, firstly, reordering the time difference of arrival values from small to large; and then, establishing a sequence change curve of the arrival time difference by taking the serial number as an abscissa and the numerical value of the arrival time difference as an ordinate.
The DTOA between the pulses is calculated and rearranged from small to large as shown below
D＝{dtoa i |pri min ≤dtoa i ≤dtoa i+1 ≤pri max ,i＝1,...,l-1} (4)
Wherein l is the length of D, [ pri ] min ,pri max ]Indicating a preset reasonable range of PRIs. The DTOA order change curve can be expressed as
C 1 ＝{(i,dtoa i )|dtoa i ∈D,i∈In,In＝{1,2,...,l}} (5)
Assuming a radiation source pulse repetition interval pri and a time of arrival measurement error δ · pri, C 1 As shown in FIG. 2, the distribution of D will map completely to C 1 When D is most sparsely distributed, C 1 The change is fastest, and D is most densely distributed at the position of the real PRISet, correspond to C 1 The variation is most gradual. This feature is essentially consistent with the histogram method in terms of characterizing true PRIs, and is fully applicable to extracting PRIs.
Further, in constructing the equivalent arrival time difference density curve, first, the arrival time difference value is preprocessed, where the preprocessing at least includes: calculating the difference value of the numerical values of the adjacent arrival time differences, and subtracting the difference value obtained by calculation from the maximum value of the arrival time differences; then, an equivalent arrival time difference density curve is obtained by performing mean filtering on the preprocessing result. Extracting PRIs and converting the PRIs into extreme points for extracting a DTOA probability density distribution curve, wherein the DTOA sequence change curve is very simple to construct, after one-dimensional DTOA is reordered according to the numerical value, the curve is obtained by adding the serial number of the one-dimensional DTOA and expanding the serial number to two dimensions, and the slope extreme point of the curve and the extreme points of the DTOA probability density distribution curve have a corresponding relation; the equivalent DTOA density curve is obtained by preprocessing the DTOA sequence variation curve, and the preprocessing process comprises the following steps: the derivative (difference), inversion (negative), translation (maximum added) are obtained, and the peak of the curve corresponds exactly to the slope extreme point of the DTOA sequence variation curve. Further, the point location for performing mean filtering on the preprocessing result is obtained according to the set adjustable parameter and the length of the staggered pulse train.
And constructing a density curve equivalent to the DTOA probability density distribution curve by using the DTOA sequential variation curve, namely an equivalent DTOA density curve, and representing the extreme value of the DTOA probability density distribution curve in a peak form. The specific process can be described as follows:
first, a difference calculation is performed for D, and the result is shown below
DD＝{dtoa i+1 -dtoa i |i＝1,2,...,l-1} (6)
Then the DD is negated, plus its maximum ddtoa max To obtain IDD as shown below
IDD＝{ddtoa max -dtoa i+1 +dtoa i |i＝1,2,...,l-1} (7)
Finally, carrying out T on the IDD f = η n point mean filter, η being adjustable parameter, n being the length of the interleaved burstThe resultant SIDD was as follows
The resulting equivalent DTOA density curve is shown below: c 2 ＝{(dtoa i ,SIDD(i))|i＝1,2,...l-1}
As a pulse repetition interval sorting method based on an equivalent DTOA density curve in the embodiment of the present invention, further, in extracting pulse repetition intervals PRIs, a detection threshold is determined by using an adaptive strategy, and a peak region of an equivalent arrival time difference density curve is detected by using a continuous threshold.
Detecting a peak value area by adopting a continuous threshold value T c Detection threshold T p The adaptive strategy is adopted to be determined by SIDD, and can be expressed as follows:
in the formula, λ (generally set to 0.9) is an adjustable parameter. Peak area is peak i With a position index of
j＝1,2,...,l i ，l i Is peak i K, k is the number of peak areas, peak area peak i As shown below
Peak area peak i Index of peak-to-peak position of i As shown below
The window length of the weighting window is T w ＝2L+1 (general order T) w ＝T c ) The PRIs obtained by weighting calculation with SIDD as weight can be shown as follows
In the embodiment of the scheme, by using the idea of SDIF, the DTOA sequential variation curve is established after the current-level difference is sequentially calculated, and then the equivalent DTOA density curve is established. Because the equivalent DTOA density curve adopts a constant form threshold, and the numerical value of the equivalent DTOA density curve does not have a strict corresponding relation with the numerical value of the probability density distribution curve, sub-harmonic detection is not needed. In addition, the SDIF originally specializes the first-level difference, i.e., sequence search is performed only when more than one value exceeds a threshold, and the SDIF is not applicable.
Further, when a plurality of radiation sources exist, firstly, the arrival time difference of a preset level is accumulated and calculated by combining an accumulated difference histogram algorithm, and a corresponding equivalent arrival time difference density curve is established; and then extracting pulse repetition interval PRIs, and performing pulse sequence retrieval on the pulse repetition interval PRI fixation and the pulse repetition interval PRI jitter by using a direct sequence retrieval method. Further, in the direct sequence retrieval method, after a plurality of pulses are continuously retrieved from the reference pulse, the pulses are retrieved from the head and the tail of the last retrieved pulse to two sides simultaneously until the pulses cannot be retrieved at the preset multiple pulse repetition interval, and when the length of the retrieved pulse sequence is larger than or equal to the preset value, the pulse sequence retrieval is determined to be successful.
When a plurality of radiation sources exist, a plurality of false PRI exist in the equivalent DTOA density curve corresponding to the first-order difference value, and the false PRI is smaller than the real PRI. Therefore, the idea of Cumulative Difference Histogram (CDIF) algorithm and SDIF algorithm can be combined to calculate C cumulatively first c,max (general order C c,max = 3), and then establishing an equivalent DTOA density curve thereof, and performing PRIs extraction and sequence retrieval. If the sequence retrieval is successful, repeating the above steps, otherwise, from C c,max +1 calculation of sequential differences until the maximum number of stages C is reached s,max 。
In the direct sequence retrieval method, according to the actual use situation, 4 pulses can be continuously retrieved from the reference pulse, namely the PRI can be considered to exist, and then the PRI is retrieved from the head and the tail of the PRI to two sides simultaneously. The condition of the retrieval termination can be set as follows according to the actual situation: pulses cannot be retrieved at 1 to 5 times PRI, and the condition that the pulse sequence retrieval is successful may be set to a pulse sequence length of 10 or more.
In the case of interleaving between the PRI fixed and PRI dithered bursts, a large search window must be used for sequence searching for PRI dithering, but the search window is too long and is easily interfered by other radiation source pulses, resulting in search interruption or errors. Therefore, in the embodiment of the present invention, a small search window may be first set to perform a sequence search on PRI-fixed, and then a large search window is used to perform a sequence search on PRI-jittered, where the length of the small search window is denoted as w 1 The length of the large search window is marked as w 2 。
To verify the validity of the protocol, the following further explanation is made with reference to the test data:
the observation time is 10e4us, and the pulse loss rate is P 1 Jitter rate of P 2 The interleaved bursts may be arranged as shown in the following table:
TABLE 1 staggered burst setup
The equivalent DTOA density curves under different conditions are shown in fig. 3. Although the jitter rate P is shown in FIGS. 3 (a-c) 2 Increasing, as well as the PRI fixed and PRI spread, there is always a distinct peak where the PRI jitters. The peak value of the equivalent DTOA density curve and the maximum value point of the DTOA probability density distribution curve have a strict corresponding relation, and the scheme in the embodiment of the scheme has good sorting performance on the PRI jittering pulse sequence besides the sorting capacity on the PRI fixation and the PRI dispersion. In addition, because the overlapping window strategy is not adopted, the corresponding problem of the strategy is avoided. As shown in FIG. 3 (d-f), on the equivalent DTOA density curve, although the pulse loss rate P is 1 And the peak value of the position of the real PRI is basically in the same level regardless of the size of the real PRI and the length of the pulse sequence, so that the peak value detection can be carried out by setting a threshold in a constant form, and the advantage of sub-harmonic waves in threshold detection can be eliminated. In addition, on the equivalent DTOA density curve, the real PRI is always located at the position of the maximum value, so that the peak detection threshold can be set by adopting a self-adaptive strategy according to the maximum value and the mean value of the equivalent DTOA density curve. The peak level of the PRI jitter is a little lower than that of the PRI in the conventional system, so when the adaptive threshold coefficient λ is set, if the PRI is intentionally fixed and the PRI disparity is extracted before the PRI jitter is extracted, the threshold coefficient λ needs to be set to be larger.
Taking three cases as examples, the PRI fixed interleaved burst, the PRI dithered interleaved burst, and the PRI fixed and PRI dithered interleaved burst compare the existing algorithm with the scheme implementation algorithm of the present case:
1. PRI fixed interleaved burst
Two groups of comparison algorithms are set in the experiment, namely SDIF algorithm which is recorded as SDIF-1, and PRI transformation method which is recorded as PRIT-1, and the scheme implementation method is recorded as SIDD. The observation duration was 10e4us, PRI fixed staggered burst setting is shown in Table 2
TABLE 2 PRI fixed interleaved burst setup
Let pri min ＝5us,pri max =2000us, sequence search window length w =4us c,max ＝3，C s,max =7,PRI box length is 0.4us. SDIF-1 has a threshold coefficient of x =0.04, g =0.2, prit-1 has a threshold of α =0.03, β =0.015, γ =3, sidd has threshold related coefficients set to λ =0.9, η =0.05, t c ＝T w ＝9。
The Monte Carlo experiments were performed 1000 times each at each pulse loss rate, for a total of 18000 Monte Carlo experiments.
TABLE 3 simulation experiment results
As shown in Table 3, accuracy I is estimated for PRI for a PRI fixed-interlace burst 3 In other words, SDIF-1, PRIT-1 and SIDD are almost indistinguishable and do not follow the pulse loss rate P 1 May vary. As shown in FIGS. 4 (a-c), P 1 When = 0-50%, then I 1 、I 2 And I 4 In other words, SIDD has slight differences from SDIF-1 and PRIT-1, but the performance is generally consistent. In addition, in P 1 And when the content is 0 to 30 percent, the performances of the three components are very excellent. As shown in Table 3, at sort time I 5 Above, SIDD is [0.0046,0.0057]s, SDIF-1 is [0.0282,0.0697 ]]s, PRIT-1 is [0.7098,2.1613]And s. As shown in FIG. 4 (d), the time efficiency of SIDD is much better than SDIF-1 and PRIT-1. In conclusion, the performance of the algorithm SIDD realized by the scheme is superior to SDIF-1 and PRIT-1.
As shown in FIGS. 4 (a-c), P 1 If the content is 0 to 30%, the performance of SDIF-1 and PRIT-1 is excellent. The reason is that on the premise of not increasing the sorting time, the detection thresholds of SDIF-1 and PRIT-1 are set at a lower level as far as possible through manual debugging in the experiment, so that the real PRI still meets the threshold passing requirement when the pulse loss rate is increased progressively, and the better sorting performance is maintained.
In a simulation experiment, the threshold coefficient can be debugged through the evaluation index to approach the optimal threshold as much as possible, but in an actual situation, a reconnaissance party adopts a non-cooperative working mode and is hardly easy to adjust to the proper threshold through manual debugging. On the contrary, the SIDD does not have the problem related to the threshold setting, as shown in fig. 5, the SIDD adopts the adaptive constant threshold, and the PRIs extraction process always only locks the real PRI and its sub-harmonics no matter how the pulse loss rate changes, and it can be ensured that the real PRI is extracted first only by performing the sequence retrieval in the order from small to large. Therefore, the adaptive constant threshold of SIDD is effective in dealing with different pulse loss rate situations when dealing with PRI fixed interleaved bursts and has a significant advantage in time efficiency compared to SDIF-1 and PRIT-1.
2. PRI dithered interleaved burst
Two groups of comparison experiments are set in the experiment, wherein the first group adopts SDIF algorithm of a novel window and is recorded as SDIF-2, and the second group adopts modified PRI transformation algorithm and is recorded as PRIT-2. The observation duration was 10e4us, and the PRI jittered staggered burst setting is shown in Table 4
TABLE 4 PRI jitter interleave burst setup
Let pri min ＝5us,pri max =2000us, sequence search window length w = pri · P 2 ，C c,max ＝3，C s,max =7,pri bin length 0.4us. The threshold coefficient of SDIF-2 is x =0.12, g =0.2, PRIT-2 threshold coefficient is alpha =0.15, beta =0.12, gamma =5, the threshold-related coefficient of SIDD is lambda =0.9, eta =0.1, T c ＝T w ＝5。
Ideally, the crossover window length is set to pri P, and the comparative experiments are labeled SDIF-2-2 and PRIT-2-2, respectively. However, when a plurality of different jitter rates coexist, SDIF-2 and PRIT-2 cannot distinguish and process different jitter rates, and usually only the overlap window length can be set to pri · P 2,max ，P 2,max =30% indicates the expected ability to handle PRI jitter, compare experiments are noted SDIF-2-1 and PRIT-2-1, respectively.
The shaking ratio was 1000 times, and 20000 Monte Carlo experiments were performed in total, and the results are shown in Table 5
Table 5 experimental results of simulation experiment
As shown in FIG. 6, there was a significant drop in SDIF-2-1 and PRIT-2-1 sorting performance compared to SDIF-2-2 and PRIT-2-2. As shown in Table 5, SDIF-2-1 and PRIT-2-1 also showed a significant decrease in time efficiency. In summary, the lower the jitter rate, the more pronounced the decrease. The main reason is that the overlap window length of SDIF-2-1 and PRIT-2-1 is set to pri P 2,max Resulting in an impact on the performance of handling small jitter rates. SIDD does not have the problem, and the sorting performance of the SIDD shows a strict descending trend along with the increase of jitter like SDIF-2-2 and PRIT-2-2. As shown in FIG. 6, from I 1 ～I 4 In view, the performances of the SIDD and the SDIF-2-2 are basically consistent and are superior to those of PRIT-2-2. As shown in Table 5, at sort time I 5 In this context, SIDD is [0.0059,0.0063]s, SDIF-2-2 is [0.0647,0.0886]s, PRIT-2-2 is [0.4332,1.2179 ]]The time efficiency of s, SIDD is far superior to SDIF-2-2 and PRIT-2-2. In conclusion, the SDIF-2-1 and PRIT-2-1 cannot effectively process various staggered conditions with different jitter rates, the SIDD has the processing capability of the conditions, and the sorting performance of the SIDD is far better than that of the SDIF-2-1 and PRIT-2-1. In view of time efficiency, the sorting performance of SIDD is superior to SDIF-2-2 and PRIT-2-2.
3. PRI fixed and PRI dithered interleaved burst
The comparison experiment was performed as in the previous PRI dithered staggered burst experiment, with an observation duration of 10e4us, and with PRI fixed and PRI dithered staggered burst settings as shown in the following table
TABLE 6 PRI dithered staggered burst setup
For the situation of interleaving of pulse strings of PRI fixed pulse trains and PRI jittering pulse trains, the scheme of the scheme adopts a double-scale search window to carry out sequence search and uses a small search window w 1 Search PRI fixation using a large search window w 2 ＝pri·P 2 PRI jitter is retrieved. PRI fixed estimation considering peak broadeningAccuracy decreases, in SDIF-2 and PRIT-2 experiments, w 1 =10us, threshold coefficient is tested with PRI jitter interleaved burst. In the SIDD experiment, when PRI is fixed sorted, w 1 And =2us, setting the threshold coefficient to be identical to the PRI fixed staggered pulse train experiment, and setting the threshold coefficient to be identical to the PRI jittered staggered pulse train experiment when the PRI is jittered and sorted.
Each shaking ratio was 1000 times, 12000 Monte Carlo experiments were performed in total, and the results are shown in Table 7
Table 7 experimental results of simulation experiment
As shown in FIG. 7, from the viewpoint of the sorting performance fixed for PRI, SIDD is superior to SDIF-2 and PRIT-2 in any evaluation index, which indicates that the sorting performance of SIDD is far superior to SDIF-2 and PRIT-2 in the present case. As shown in fig. 7 (c), the main reason is that the overlap window strategy is adopted by SDIF-2 and PRIT-2, which results in significant deterioration of estimation accuracy for PRI fixation, and further significant deterioration of other evaluation indexes. As shown in Table 7, the estimated accuracy of SDIF-2 was [0.69,2.27]%, and the estimated accuracy of PRIT-2 was [1.51,6.46]%, which was drastically deteriorated as compared to 0.03%. The SIDD does not have the problem, the estimation precision of the SIDD is always kept about 0.03 percent, and the estimation precision is basically consistent with the experimental result of the PRI fixed staggered pulse train condition. This indicates that SIDD has good sorting performance for PRI fixation in the current case.
As shown in FIG. 8, from the sorting performance on PRI jitter, the sorting performance of SDIF-2 and PRIT-2 is significantly lower than SIDD. For any evaluation index, SIDD is superior to PRIT-2. As shown in FIGS. 8 (a) and 8 (c), I 1 And I 3 In other words, SDIF-2 substantially coincides with SIDD as shown in FIGS. 8 (b) and 8 (d), but I of SDIF-2 2 And I 4 Much like SIDD. The cause of this disease, SDIF-2 and PRIT-2 do not work well for PRI fixed sorting, which makes the sequence retrieval process of PRI jitter very vulnerable to interference of the remaining PRI fixed pulses. SIDD is ideal for sorting performance of PRI dithering because it has performed efficient sequence search for PRI fixing. In addition, the sorting time for SDIF-2 was [0.1638,0.2873 ]]sorting time for s, PRIT-2 was [0.4581,1.8033]sorting times for s, SIDD were [0.0075,0.0088]The time efficiency of s, i.e., SIDD, is much better than SDIF-2 and PRIT-2.
Therefore, the algorithm SIDD realized by the scheme can effectively process the pulse trains of the PRI fixed pulse train and the PRI jittered pulse train, and the SIDD is superior to SDIF-2 and PRIT-2 no matter the PRI jitters or the PRI is fixed.
Unless specifically stated otherwise, the relative steps, numerical expressions, and values of the components and steps set forth in these embodiments do not limit the scope of the present invention.
Based on the foregoing method and/or system, an embodiment of the present invention further provides a server, including: one or more processors; a storage device for storing one or more programs which, when executed by the one or more processors, cause the one or more processors to implement the method described above.
Based on the above method and/or system, the embodiment of the invention further provides a computer readable medium, on which a computer program is stored, wherein the program, when executed by a processor, implements the above method.
In all examples shown and described herein, any particular value should be construed as merely exemplary, and not as a limitation, and thus other examples of example embodiments may have different values.
It should be noted that: like reference numbers and letters refer to like items in the following figures, and thus, once an item is defined in one figure, it need not be further defined and explained in subsequent figures.
Finally, it should be noted that: although the present invention has been described in detail with reference to the foregoing embodiments, it should be understood by those skilled in the art that the following descriptions are only illustrative and not restrictive, and that the scope of the present invention is not limited to the above embodiments: any person skilled in the art can modify or easily conceive the technical solutions described in the foregoing embodiments or equivalent substitutes for some technical features within the technical scope of the present disclosure; such modifications, changes or substitutions do not depart from the spirit and scope of the embodiments of the present invention, and they should be construed as being included therein. Therefore, the protection scope of the present invention shall be subject to the protection scope of the claims.
Claims (10)
1. A pulse repetition interval sorting method based on an equivalent DTOA density curve is characterized by comprising the following contents:
acquiring arrival time differences among radiation source pulse signals, and establishing an arrival time difference sequence change curve among the signals through magnitude sorting;
constructing an equivalent arrival time difference density curve which is equivalent to the arrival time difference probability density distribution curve and used for representing an extreme value of the arrival time difference probability density distribution curve by using the sequential variation curve;
and extracting pulse repetition intervals PRIs by detecting the peak value of the equivalent arrival time difference density curve according to a preset threshold value.
2. The equivalent DTOA density profile-based pulse repetition interval sorting method of claim 1, wherein the time difference of arrival sequence variation profile is established by first reordering the time difference of arrival values in order of decreasing to increasing; and then, establishing a sequence change curve of the arrival time difference by taking the serial number as an abscissa and the numerical value of the arrival time difference as an ordinate.
3. The method of claim 1 or 2, wherein in constructing the DTOA density profile, the time difference of arrival values are first preprocessed, the preprocessing at least comprises: calculating the difference value of the numerical values of the adjacent arrival time differences, and subtracting the difference value obtained by calculation from the maximum value of the arrival time differences; then, an equivalent arrival time difference density curve is obtained by performing mean filtering on the preprocessing result.
4. The method of claim 3, wherein the point locations for performing mean filtering on the pre-processing results are obtained according to the set adjustable parameters and the interleaving burst length.
5. The equivalent DTOA density curve-based pulse repetition interval sorting method of claim 3 wherein the equivalent time difference of arrival density curve is represented as C 2 ＝{(dtoa i 1, 2.. L-1}, wherein dtoa (i)) | i =1,2 i Representing the ith time difference of arrival value; the mean filtering process is represented as:
IDD (j) denotes the jth preprocessing result, T f N is the length of the interleaved burst, and l is the total length of the time difference of arrival value sequence.
6. The equivalent DTOA density curve-based pulse repetition interval sorting method of claim 5, wherein in extracting the pulse repetition intervals PRIs, an adaptive strategy is used to determine a detection threshold, and a continuous threshold is used to detect a peak region of the equivalent time difference of arrival density curve.
7. The equivalent DTOA density curve-based pulse repetition interval sorting method of claim 6, wherein an adaptive strategy determines the detection threshold T p The process of (a) is represented as:
wherein λ is an adjustable parameter.
8. The equivalent DTOA density profile-based pulse repetition interval sorting method of claim 6, wherein the pulse repetition interval PRIs extraction process is represented as:
wherein, index i For peak position index, D (-) is the sequence of arrival time difference values, and L is the weighted window length parameter.
9. The equivalent DTOA density curve-based pulse repetition interval sorting method as recited in claim 1, wherein when a plurality of radiation sources are present, a cumulative difference histogram algorithm is combined to cumulatively calculate the arrival time difference of a preset level and establish a corresponding equivalent arrival time difference density curve; and then extracting pulse repetition interval PRIs, and performing pulse sequence retrieval on the pulse repetition interval PRI fixation and the pulse repetition interval PRI jitter by using a direct sequence retrieval method.
10. The method of claim 9, wherein in the direct sequence search method, after a plurality of pulses are consecutively searched from the reference pulse, the search is performed from the beginning and the end of the last searched pulse to both sides simultaneously until the pulse cannot be searched at the preset multiple pulse repetition interval, and when the length of the searched pulse sequence is greater than or equal to the preset value, the pulse sequence search is determined to be successful.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210242468.6A CN114660560B (en) | 2022-03-11 | 2022-03-11 | Pulse repetition interval sorting method based on equivalent DTOA density curve |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210242468.6A CN114660560B (en) | 2022-03-11 | 2022-03-11 | Pulse repetition interval sorting method based on equivalent DTOA density curve |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN114660560A CN114660560A (en) | 2022-06-24 |
| CN114660560B true CN114660560B (en) | 2022-12-06 |
Family
ID=82029023
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210242468.6A Active CN114660560B (en) | 2022-03-11 | 2022-03-11 | Pulse repetition interval sorting method based on equivalent DTOA density curve |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114660560B (en) |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| EP0452023A2 (en) * | 1990-04-12 | 1991-10-16 | Raytheon Company | Method and apparatus for pulse sorting |
| US7184493B1 (en) * | 2002-02-05 | 2007-02-27 | Alliant Techsystems Inc. | Pulse sorting apparatus for frequency histogramming in a radar receiver system |
| CN110764063A (en) * | 2019-10-15 | 2020-02-07 | 哈尔滨工程大学 | Radar signal sorting method based on combination of SDIF and PRI transformation method |
| CN114019505A (en) * | 2021-11-09 | 2022-02-08 | 中国人民解放军海军航空大学 | Radar signal sorting method and system based on PRI interval information |
| CN114089285A (en) * | 2022-01-24 | 2022-02-25 | 安徽京淮健锐电子科技有限公司 | Signal sorting method based on first-order pulse repetition interval PRI |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US7916814B2 (en) * | 2007-05-30 | 2011-03-29 | Bandspeed, Inc. | Method and apparatus for real-time pulse parameter estimator |
-
2022
- 2022-03-11 CN CN202210242468.6A patent/CN114660560B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| EP0452023A2 (en) * | 1990-04-12 | 1991-10-16 | Raytheon Company | Method and apparatus for pulse sorting |
| US7184493B1 (en) * | 2002-02-05 | 2007-02-27 | Alliant Techsystems Inc. | Pulse sorting apparatus for frequency histogramming in a radar receiver system |
| CN110764063A (en) * | 2019-10-15 | 2020-02-07 | 哈尔滨工程大学 | Radar signal sorting method based on combination of SDIF and PRI transformation method |
| CN114019505A (en) * | 2021-11-09 | 2022-02-08 | 中国人民解放军海军航空大学 | Radar signal sorting method and system based on PRI interval information |
| CN114089285A (en) * | 2022-01-24 | 2022-02-25 | 安徽京淮健锐电子科技有限公司 | Signal sorting method based on first-order pulse repetition interval PRI |
Non-Patent Citations (4)
| Title |
|---|
| "A Fast and Real-time PRI Transform Algorithm for Deinterleaving Large PRI Jitter Signals";Yin Xi et al.;《Proceedings of the 37th Chinese Control Conference》;20180727;全文 * |
| "一种脉冲重复间隔复杂调制雷达信号分选方法";李英达 等;《电子与信息学报》;20131031;第35卷(第10期);全文 * |
| "基于到达时间差直方图的信号分选算法研究";杨翔 等;《电子与信息学报》;20151130;第37卷(第11期);全文 * |
| "基于改进DSets的无参数雷达信号分选算法";刘鲁涛 等;《中国舰船研究》;20210831;第16卷(第4期);全文 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN114660560A (en) | 2022-06-24 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110764063B (en) | Radar signal sorting method based on combination of SDIF and PRI transformation method | |
| Milojević et al. | Improved algorithm for the deinterleaving of radar pulses | |
| CN110865357B (en) | Laser radar echo signal noise reduction method based on parameter optimization VMD | |
| FI107081B (en) | Procedure and arrangement for determining the number of partial discharges | |
| CN108549078B (en) | Cross-channel combination and detection method for radar pulse signals | |
| CN110929842B (en) | Accurate intelligent detection method for non-cooperative radio signal burst time region | |
| CN111104398A (en) | Detection method and elimination method for approximate repeated record of intelligent ship | |
| CN114355298B (en) | Radar composite modulation pulse signal identification method | |
| CN108089100A (en) | The detection method of small current neutral grounding system arc light resistance earth fault | |
| CN114660560B (en) | Pulse repetition interval sorting method based on equivalent DTOA density curve | |
| CN117111016B (en) | Real-time pulse analysis method and system based on channelization in complex electromagnetic environment | |
| CN112198481B (en) | PRI jitter radar signal sorting method under condition of pulse loss aliasing | |
| CN112698274A (en) | Radar signal sorting and pulse sequence extraction system based on hierarchical PRI transformation | |
| US8175829B2 (en) | Analyzer for signal anomalies | |
| CN110632563B (en) | Intra-pulse frequency coding signal parameter measuring method based on short-time Fourier transform | |
| CN115980689A (en) | Point cloud detection-based radiation source signal sorting method, device, equipment and medium | |
| CN106446548A (en) | Dynamic structural mutation detection method based on wavelet analysis | |
| CN112016045A (en) | Data processing method of nanosecond pulse power meter | |
| CN112379357B (en) | Signal discrimination method based on secondary processing of pulse signal estimated time parameter | |
| CN113702919B (en) | Method and device for estimating PRI value and extracting pulse sequence | |
| CN112763989B (en) | CDIF-based jitter signal sorting method | |
| CN112765557B (en) | Ship high-resolution range profile power conversion method and system | |
| CN112924831B (en) | Ultrahigh frequency partial discharge positioning time delay estimation method | |
| CN115372923B (en) | Multi-dimensional combination-based radar sea slow target detection method | |
| CN117310636B (en) | Fixed pulse repetition interval measurement method, device and medium |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |