CN115291181A

CN115291181A - Constant false alarm probability detection method, device and equipment based on phase characteristics

Info

Publication number: CN115291181A
Application number: CN202210870118.4A
Authority: CN
Inventors: 刘宁波; 关键; 董云龙; 王国庆; 黄勇; 丁昊; 曹政; 于恒力
Original assignee: Naval Aeronautical University
Current assignee: Naval Aeronautical University
Priority date: 2022-07-22
Filing date: 2022-07-22
Publication date: 2022-11-04

Abstract

The invention provides a constant false alarm probability detection method, a device and equipment based on phase characteristics, wherein the method comprises the following steps: acquiring a pulse echo obtained by detecting a target to be detected in a sea surface environment by a coherent radar; acquiring phase information of a plurality of continuous pulse echoes of a first distance unit of a reference unit in the pulse echoes; determining a linear relationship of the phase information of the first range bin based on the phase information of the plurality of successive pulse echoes; and under the condition that the linear relation is linearly related, deleting the pulse echo of the first distance unit, and carrying out target detection on the target to be detected through a constant false alarm probability detection method based on the rest reference units. The embodiment of the invention is used for solving the defects that when two or more targets or strong ground clutter interference and the like exist around the target to be detected, the CFAR detector cannot detect the target, and the target shielding phenomenon occurs in the prior art.

Description

Constant false alarm probability detection method, device and equipment based on phase characteristics

Technical Field

The invention relates to the technical field of target detection, in particular to a constant false alarm probability detection method, device and equipment based on phase characteristics.

Background

In a complex marine environment, marine radars such as shore-based radars and ship-based radars are inevitably influenced by sea surface scattering echoes, namely sea clutters, when military and civil targets such as ships, sea skimming aircrafts, channel buoys, fishing boats, small yachts, floating ice and the like are detected. Particularly under the working conditions of high-resolution radar and high sea conditions, the peak phenomenon frequently occurs in sea clutter, the overall energy is strong, false alarms are easily caused, and the detection of a marine target is seriously influenced. In order to avoid the loss caused by False detection and missing detection, various CFAR (Constant False Alarm probability) detectors are needed to reduce the occurrence of False alarms.

A common CFAR detector is a Mean Level (ML) class CFAR detector. In such detectors, their local interference power level estimation adopts an Averaging method, the most classical three Of which are Cell Averaging CFAR (CA-CFAR), cell Averaging large GO (green Of) -CFAR and Cell Averaging small SO (Smallest Of) -CFAR. However, the concept of unit-averaged CFAR is limited to two basic assumptions.

1. The targets are independent. The length of at least one reference window between the objects is such that there is no possibility of two objects being present within the reference window at the same time.

2. All interference data within the reference window is independently distributed and co-distributed with the interference within the cell containing the target, i.e. the interference is uniform.

In a complex marine environment, the actual situation often violates one or two conditions, when two or more targets or strong ground clutter interfere around a target to be detected, the target echo power of a reference unit may exceed the surrounding interference power, the clutter power estimation value is increased, and the threshold of the CFAR is raised, so that the CFAR detector cannot detect the target, and a target shielding phenomenon occurs.

Disclosure of Invention

The invention provides a constant false alarm probability detection method, a constant false alarm probability detection device and constant false alarm probability detection equipment based on phase characteristics, which are used for solving the defects that in the prior art, when two or more targets are arranged around a target to be detected, a CFAR (constant false alarm rate) detector cannot detect the target and a target shielding phenomenon occurs, and the detection performance of target detection is improved.

The invention provides a constant false alarm probability detection method based on phase characteristics, which comprises the following steps:

acquiring a pulse echo obtained by detecting a target to be detected in a sea surface environment by a coherent radar;

acquiring phase information of a plurality of continuous pulse echoes of a first distance unit of a reference unit in the pulse echoes;

determining a linear relationship of the phase information of the first range bin based on the phase information of the plurality of successive pulse echoes;

and under the condition that the linear relation is linearly related, deleting the pulse echo of the first distance unit, and carrying out target detection on the target to be detected through a constant false alarm probability detection method based on the rest reference units.

The invention also provides a constant false alarm probability detection device based on phase characteristics, which comprises:

the first acquisition module is used for acquiring a pulse echo obtained by detecting a target to be detected in a sea surface environment by the coherent radar;

the second acquisition module is used for acquiring phase information of a plurality of continuous pulse echoes of a first distance unit of the reference unit in the pulse echoes;

a linear relationship determination module for determining a linear relationship of the phase information of the first range bin based on the phase information of the plurality of continuous pulse echoes;

and the target detection module is used for deleting the pulse echo of the first distance unit under the condition that the linear relation meets a preset condition, and carrying out target detection on the target to be detected through a constant false alarm probability detection method based on the rest reference units.

The invention also provides an electronic device, which comprises a memory, a processor and a computer program stored on the memory and capable of running on the processor, wherein the processor executes the program to realize the phase feature-based constant false alarm probability detection method.

The present invention also provides a non-transitory computer readable storage medium having stored thereon a computer program which, when executed by a processor, implements a phase feature based constant false alarm probability detection method as described in any of the above.

The invention also provides a computer program product comprising a computer program which, when executed by a processor, implements a method for phase-signature-based constant false alarm probability detection as described in any of the above.

According to the constant false alarm probability detection method, device and equipment based on the phase characteristics, pulse echoes are obtained by detecting the target to be detected in the sea surface environment through the acquisition of the coherent radar; acquiring phase information of a plurality of continuous pulse echoes of a first distance unit of a reference unit in the pulse echoes; determining a linear relationship of the phase information of the first range bin based on the phase information of the plurality of successive pulse echoes; and under the condition that the linear relation is linearly related, deleting the pulse echo of the first distance unit, and carrying out target detection on the target to be detected through a constant false alarm probability detection method based on the rest reference units. When the coherent radar is adopted to continuously detect the target to be detected in the sea surface environment, the phase characteristics of land islands, strong interference targets or sea peaks are linear under the condition that certain confidence coefficient allows. Therefore, in the embodiment of the present invention, based on the phase information of a plurality of consecutive pulse echoes of a first range bin of a reference bin in the pulse echoes, a linear relationship of the phase information of the first range bin is determined; and under the condition that the linear relation is linearly related, deleting the pulse echo of the first distance unit, and carrying out target detection on the target to be detected through a constant false alarm probability detection method based on the rest reference units. Therefore, the strong interference targets or the strong sea clutter near the target to be detected are screened and then removed, the amplitude probability density function of the sea clutter is enabled to be more consistent with Rayleigh distribution, the power of the background clutter is not affected by the interference targets, the background clutter is more consistent with the precondition used by the CFAR detection method, and meanwhile, the detection threshold is prevented from being accidentally raised by the interference factors of the first distance units of which the linear relation is linearly related, so that the detection performance of target detection is improved, and missing detection is reduced.

Drawings

In order to more clearly illustrate the technical solutions of the present invention or the prior art, the drawings needed for the description of the embodiments or the prior art will be briefly described below, and it is obvious that the drawings in the following description are some embodiments of the present invention, and those skilled in the art can also obtain other drawings according to the drawings without creative efforts.

FIG. 1 is a schematic flow chart of a constant false alarm probability detection method based on phase characteristics according to the present invention;

FIG. 2 is a diagram illustrating the calculation principle of a conventional CA-CFAR method;

FIG. 3a is a diagram showing a linear correlation of phase characteristics;

FIG. 3b is a diagram illustrating a non-linear dependence of phase characteristics;

FIG. 4 is a schematic diagram illustrating the computational principles of the improved CA-CFAR method of the present invention;

FIG. 5 is a schematic diagram showing a comparison of the detection performance curves of an ML-CFAR detector, a CMLD-CFAR detector and the improved CA-CFAR detection method of the present invention for single target detection in a uniform single target background environment;

FIG. 6a is a schematic diagram showing 2 target signals of different simulated sizes;

FIG. 6b is a schematic representation of the ROC curve for the larger of two targets;

FIG. 7a shows a schematic diagram of simulating 4 target signals;

FIG. 7b is a schematic diagram of an SO-CFAR detector compared to the A-frame detection threshold of the improved CA-CFAR detection method of the present invention;

FIG. 8a is a schematic diagram illustrating the simulation of 4 target signals in a clutter edge environment;

FIG. 8b shows a schematic diagram of a GO-CFAR detector compared to the A-frame detection threshold of the improved CA-CFAR detection method of the present invention;

FIG. 9 is a diagram illustrating P-frames of scan data according to an embodiment of the present invention;

fig. 10a is a diagram illustrating a comparison of thresholds of a conventional CA-CFAR device and an improved CA-CFAR method according to an embodiment of the present invention;

FIG. 10b is a diagram illustrating a comparison of thresholds of a conventional SO-CFAR and the improved CA-CFAR method of the present invention;

FIG. 11 is a schematic structural diagram of a constant false alarm probability detection apparatus based on phase characteristics according to the present invention;

fig. 12 is a schematic structural diagram of an electronic device provided in the present invention.

Detailed Description

In order to make the objects, technical solutions and advantages of the present invention clearer, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings, and it is obvious that the described embodiments are some, but not all embodiments of the present invention. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.

Echo signals received by the radar not only contain target signals, but also are mixed with various clutter, noise and interference signals, and meanwhile, the radar itself may also influence detection. If a fixed threshold is used for detection, a great amount of false alarms or false alarms occur, so that the method needs to be adoptedThe adaptive threshold replaces the fixed threshold, and the CFAR method can be adaptively adjusted along with the background noise, clutter and interference of the detected point. Among them, CA-CFAR (cell average constant false alarm detection) is the most basic target detection algorithm, and its working principle is shown in fig. 2, please refer to fig. 2, and fig. 2 shows the calculation principle of the conventional CA-CFAR method. In fig. 2D is the current unit under test (CUT), whose amplitude value is to be compared with the adaptive threshold; since the power of the target may leak into the adjacent units, several units adjacent to the target are not used as the estimation of the background clutter and are used as the protection unit P, and the protection unit is not included when the background power is calculated; taking N reference units x on the left and right sides of a unit D to be detected _i (i =1, \8230;, N), the reference unit, the protection unit, and the unit to be detected are collectively called a CFAR processing window; then, the reference units on the left side and the right side are respectively summed and averaged to estimate the background clutter power Z of the unit to be detected; at a given expected false alarm probability P _FA And then, multiplying the background clutter power Z by the normalization factor alpha to obtain a detection threshold T, comparing the unit D to be detected with the detection threshold T, if the amplitude of the target to be detected is greater than the threshold value, determining that the detection unit is a target, otherwise, determining that the detection unit is not a target.

The sea clutter is formed by overlapping scattering signals of a large number of scattering units in an antenna beam area, so that the sea clutter can be approximately considered to be in Gaussian distribution, an amplitude probability density function also accords with Rayleigh distribution after clutter echoes are subjected to amplitude detection, and the probability density function is

According to the unit average constant false alarm principle diagram, the background clutter power Z of the unit to be detected is

Multiplying the obtained background clutter power Z of the unit D to be detected by the normalization factor alpha to obtain a threshold value T which is expressed as

The detection threshold value expression can be finally obtained by estimation as

Let c = α/N, in conjunction with equation (1), the probability density distribution of the threshold T can be calculated, expressed as

Integrating the probability density distribution of the obtained false alarm probability through a Nelman-Pearson criterion to obtain the false alarm probability, and expressing the final result as

For the above equation, for a given expected false alarm probability P _FA The desired normalization factor can be obtained by solving

Note that the false alarm probability P _FA Independent of the magnitude of the actual interference noise power, is only related to the neighboring cell samples N participating in the averaging and the normalization factor α. Thus, the cell average constant false alarm is characteristic of CFAR.

The concept of cell-averaged CFAR is limited to two basic assumptions.

In view of this, embodiments of the present invention provide a method, an apparatus, and a device for detecting a constant false alarm probability based on phase characteristics, so as to solve the problem in the prior art that when two or more targets are located around a target to be detected, a CFAR detector cannot detect the target, and a target occlusion phenomenon occurs, thereby improving the detection performance of target detection.

The method for detecting the constant false alarm probability based on the phase characteristics in the invention is described in the following with reference to fig. 1, and the method comprises the following steps:

step 100, obtaining a pulse echo obtained by detecting a target to be detected in a sea surface environment by a coherent radar.

The electronic equipment acquires a pulse echo obtained by detecting a target to be detected in a sea surface environment by the coherent radar.

The coherent radar can be various shore-based radars or ship-based radars. Coherent in coherent radar means that the initial phase between pulses is deterministic, i.e. the initial phase of the first pulse may be random, but the phase between the subsequent pulse and the first pulse is deterministic, and the randomness of the initial phase of the first pulse does not affect the subsequent signal detection, which is the basis for extracting doppler information.

When coherent radar continuously detects sea surface targets, the phase characteristics of land islands, strong interference targets or sea spikes are linear under the permission of a certain confidence. Therefore, the pulse echo obtained by detecting the target to be detected in the sea surface environment through the coherent radar is obtained, and the phase characteristics of the reference unit near the target to be detected can be obtained. The reference units with strong phase linearity can be screened conveniently according to the phase characteristics of the reference units near the target to be detected.

And 200, acquiring phase information of a plurality of continuous pulse echoes of a first distance unit of a reference unit in the pulse echoes.

The electronic equipment acquires phase information of a plurality of continuous pulse echoes of a first distance unit of the reference unit in the pulse echoes. The reference unit refers to distance units on the left side and the right side of the unit D to be detected in the CA-CFAR detection algorithm. The first distance unit refers to any one of the left and right sides of the unit D to be detected. In other words, in the embodiment of the present invention, the linear relationship between the phase information of all the range bins on the left and right sides of the unit D to be detected is determined.

For example, the electronic device acquires 10 or 16 continuous pulse echoes of a certain distance unit in the reference unit of the pulse echoes, and acquires the phase information of the 10 or 16 continuous pulse echoes.

Step 300, determining a linear relation of phase information of the first range bin based on the phase information of the plurality of continuous pulse echoes;

the electronics determine a linear relationship of the phase information of the first range bin based on the phase information of the plurality of successive pulse echoes. For example, the determining a linear relationship of the phase information of the first range bin based on the phase information of the plurality of continuous pulse echoes, step 300, comprises:

and 310, calculating the phase information of the multiple continuous pulse echoes by a unary linear regression method, and determining the linear relation of the phase information of the first distance unit.

In some embodiments, a plurality of measured data of the pulse echo of the coherent radar may be tested to find a suitable confidence level, and it may be determined whether the linear relationship of the phase information of the first range bin is linearly related within an allowable range of the confidence level.

In an embodiment, in step 310, the calculating the phase information of the multiple continuous pulse echoes by using a unary linear regression method to determine a linear relationship of the phase information of the first range bin specifically includes:

311, calculating the phase information of the multiple continuous pulse echoes through a regression function to obtain model statistics;

step 312, determining the linear relationship of the phase information of the first distance unit according to the magnitude of the model statistic and the magnitude of the significance probability value.

For convenience of calculation, in the embodiment of the present invention, a unitary linear regression method may be performed on the phase information of the multiple continuous pulse echoes by using a regression function in a mathematical computation tool Matlab, so that a model statistic R may be obtained ² . According to model statistic R ² And the significance probability value p to judge whether the phases of the continuous pulse echoes of the first distance unit have a linear relation or not.

In one embodiment, the determining the linear relationship of the phase information of the first distance unit according to the magnitude of the model statistic and the magnitude of the significance probability value in step 312 includes: and when the model statistic is larger than or equal to a first set threshold value and the significance probability value is smaller than or equal to a second set threshold value, determining that the phase information of the first distance unit is linearly related.

It should be noted that, the first set threshold and the second set threshold may be determined by testing a large amount of measured data of pulse echoes of a large amount of coherent radars.

In an embodiment of the present invention, model statistic R is selected ² When the default significance level with the value larger than 0.90 and the significance probability value p smaller than 0.05 is used as a basis for judging linear correlation, the threshold is reduced due to the elimination of excessive reference units with good linearity, false alarms are easily generated, and the detection performance is influenced; when selecting model statistic R ² When the default significance level with the value larger than 0.98 and the significance probability value p smaller than 0.05 is used as the basis for judging the linear correlation, the detection performance is improved and is better compared with that of the traditional CA-CFAR detector because the rejected reference units are fewerSmall, not significant; after a large number of actual measurement data tests, the actual condition factors are synthesized, and when the model statistic R ² When the value is greater than 0.95 and the significance probability value p is less than the default significance level of 0.05, the detection performance is improved, and the occurrence of false alarms is avoided as much as possible, so that whether the phase characteristics of the reference unit are linearly related or not can be judged according to the parameters. Fig. 3a and 3b show two sets of 10 consecutive pulse echoes of the same range bin for testing, where fig. 3a shows that the phase characteristics are linearly related, fig. 3b shows that the phase characteristics are nonlinearly related, and fig. 3b shows that the phase characteristics are linearly related. FIG. 3a solves for model statistic R ² The value 0.9981 is greater than 0.95 close to 1, the significance probability value p 0.0133 is less than the default significance level of 0.05, and therefore can be judged to be linearly related; FIG. 3b solves for model statistic R ² The value 0.8343 is less than 0.95 and the significance probability value p is a default significance level with 0.2507 greater than 0.05, and therefore can be judged to be non-linearly related.

In an embodiment of the present invention, the first set threshold is in the range of [0.95,0.98]. The first set threshold is preferably 0.95.

And 400, deleting the pulse echo of the first distance unit under the condition that the linear relation is linearly related, and carrying out target detection on the target to be detected through a constant false alarm probability detection method based on the rest reference units.

In coherent radar, if the phase linearity of the successive pulse echoes is good in the same range bin, there may be targets, strong clutter, interfering targets, or land clutter in this range bin. In order to obtain a more appropriate detection threshold, reference units with better linearity near the unit D to be detected need to be removed. Therefore, when the linear relation of the phase information of the pulse echo of the first range bin is linearly related, the pulse echo of the first range bin is deleted, and the target to be detected is detected by a constant false alarm probability detection method based on the rest of the reference bins.

In some embodiments, the step 400 of performing target detection on the target to be detected by a constant false alarm probability detection method based on the remaining reference units includes:

step 410, averaging the pulse echoes based on the rest reference units to obtain an updated background clutter power estimation value;

step 420, obtaining an updated threshold value based on the product of the updated background clutter power estimation value and the normalization factor;

and 430, performing target detection based on the comparison result of the updated threshold value and the pulse echo of the unit to be detected.

Referring to fig. 4, fig. 4 is a schematic diagram illustrating a calculation principle of the improved CA-CFAR method according to the present invention, which performs target detection on the target to be detected by a constant false alarm probability detection method based on the remaining reference units. In fig. 4, the selection rules of the unit D to be detected, the protection unit P, and the reference unit N are the same as those of the classical CA-CFAR method in fig. 2, except that the reference units are selected on the left and right sides. In the embodiment of the invention, the first distance unit of the reference unit with good linearity is screened out in the step 300, the pulse echo intensity of the first distance unit is changed into zero, the number m and N of the pulse echo intensities on the left side and the right side are counted in the process of sliding a window, then N reference units are summed and averaged, the average value at the moment is actually the average value of N- (m + N) reference units, namely a new background clutter power estimated value Z, the new background clutter power Z is multiplied by a normalization factor alpha to obtain a new threshold value, and finally the new threshold value is compared with a unit D to be detected to judge whether a target exists in the distance unit. For example, assuming that 32 reference units N are initially selected, and 16 reference units N are respectively provided on the left and right sides, and then strong targets or strong clutter with good linearity are removed under the judgment of the phase linearity of certain model statistics, the actual number of the reference units may become 24 at this time, and the background clutter power estimation values obtained by the remaining reference units are far more suitable than the background clutter power estimation values obtained by the 32 reference units.

According to the embodiment of the invention, the reference units with strong phase linearity are removed by judging the phase linearity condition of the continuous pulse echoes in the distance units of the same reference unit, so that the detection threshold with better adaptability of the unit to be detected is formed.

The following describes the detection performance of the constant false alarm probability detection method based on the phase characteristics, which is applied to the uniform single-target background environment, the multi-target background environment and the clutter edge environment respectively.

The constant false alarm probability detection method based on the phase characteristics is applied to the detection performance of the uniform single-target background environment

Referring to fig. 5, fig. 5 shows the comparison between the ML-CFAR detector (i.e. average constant false alarm detector) and the CMLD-CFAR detector (i.e. average level detector) in the uniform single-target background environment and the detection performance curve (ROC curve, which is translated into the receiver operating characteristic curve, ROC curve) of the improved CA-CFAR detection method according to the embodiment of the present invention for single-target detection. False alarm probability PFA =10 in uniform clutter background ^-6 In the case of the reference unit N =32, the detection performance curve (ROC curve) of the ML-CFAR detector, the CMLD-CFAR detector, and the improved CA-CFAR detection method of the embodiment of the present invention for single target detection. Wherein the ML-CFAR detector includes an average detector (i.e., CA-CFAR) and a maximum detector (i.e., GO-CFAR and SO-CFAR). Wherein, CA-CFAR represents a unit average constant false alarm detector, GO-CFAR represents a maximum selection constant false alarm detector, and SO-CFAR represents a minimum selection constant false alarm detector.

In FIG. 5, opt is the optimal detection performance curve corresponding to the specific signal-to-clutter ratio, and the other detection performance curves are, from top to bottom, the improved CA-CFAR detection method, CA-CFAR, CMLD-CFAR, GO-CFAR, and SO-CFAR of the embodiment of the present invention. It should be noted that, since there is only one target and the improved CA-CFAR detection method in the embodiment of the present invention only has a function of screening and rejecting strong clutter or strong interference targets, the improved CA-CFAR detection method in the embodiment of the present invention is consistent with the detection performance of the conventional CA-CFAR detector. When the target signal-to-noise ratio is close to 19dB, the detection probability reaches 80%, and the detection method for screening the reference unit by using the phase characteristics does not lose the original detection function; the CMLD-CFAR has smaller signal-to-noise ratio loss compared with the CA-CFAR because part of the reference units are deleted in the detection process; the SO-CFAR has better detection performance in multi-target detection, and has larger signal-to-noise ratio loss for single-target detection.

Therefore, under the background of a uniform single target, the improved constant false alarm probability detection method based on the phase characteristics of the embodiment of the invention has the same detection performance as the unit average constant false alarm, and is improved compared with other average constant false alarm detectors.

The constant false alarm probability detection method based on the phase characteristics is applied to the detection performance of the multi-target background environment respectively

In the detection of a multi-target background environment, a phenomenon that a target with a high amplitude shields a target with a low amplitude often occurs, so that the detection probability is reduced. Referring to fig. 6a, fig. 6a shows 2 simulated target signals with different sizes in the detection of the multi-target background environment, where fig. 6a shows two simulated targets with a relatively close size, the large target power value is 5dB higher than the small target power value, and the average clutter power is about 20dB, where the phase characteristic linearity of the position where the small target is located is better. Referring to FIG. 6b, FIG. 6b shows the false alarm probability 10 of the conventional CA-CFAR detector and the improved CA-CFAR detection method using phase-signature filtering reference cells according to the present invention ^-6 And the reference unit N =32 is used for detecting a target with a larger power value (the large target signal-to-noise ratio is gradually increased from 0dB to 30dB, and the small target signal-to-noise ratio is increased therewith).

FIG. 6b shows the ROC curve for the larger of the two targets. It can be seen from fig. 6b that when the signal-to-noise ratio of the target is less than 5dB, the detection performance of the two detectors is equivalent, and the detection performance of both detectors is enhanced with the improvement of the signal-to-noise ratio, and when the system requires that the detection probability reaches 50%, the target required by the conventional CA-CFAR detector reaches 17dB, whereas the improved CA-CFAR detection method of the present invention obtains a more suitable threshold by eliminating the reference unit where the small target is located, and the target can be detected only by reaching 14 dB. It can be seen that the improved CA-CFAR detection method of the present invention has superior detection performance over conventional CA-CFAR detectors.

Referring to fig. 7a, fig. 7a shows a simulation of 4 target signals, and fig. 7a shows that four targets close to each other are simulated in the 80 th, 88 th, 98 th and 104 th distance units of clutter, wherein the power values of the 80 th and 98 th distance units are large enough, and the phase linearity is good, and the targets can be filtered and removed by the constant false alarm probability detection method based on the phase characteristics of the present invention. In the present invention, the SO-CFAR detector with the best multi-target detection effect is selected from the mean value type constant false alarm rate detectors and compared with the improved CA-CFAR detection method of the present invention, and the result is shown in the a diagram of fig. 7b, please refer to fig. 7b, as shown in fig. 7 b: the SO-CFAR detector is capable of detecting most targets among many targets, but fails to detect because the value of the target power at the 88 th range bin is too small; the improved CA-CFAR detection method screens the reference units and successfully detects all targets. Therefore, the embodiment of the invention improves the CA-CFAR detection method, can show the advantages in a multi-target detection simulation experiment, and has better detection performance.

The constant false alarm probability detection method based on the phase characteristics is applied to the detection performance of the clutter edge environment respectively

A typical clutter edge environment is generally a place of a sea-land boundary, and energy mutation occurs in a distance dimension, so that a target at a low clutter edge is judged to be a high-power clutter, and detection omission is caused; it is also possible to mistake the edge clutter in the high power region for the target, resulting in a false alarm. In the embodiment of the invention, a clutter edge environment is simulated, and two targets are added in 88 th and 95 th distance units in a low-power area, wherein the left target has better phase linearity, please refer to fig. 8a, wherein fig. 8a shows that 4 target signals are simulated in the clutter edge environment; fig. 8b shows a comparison of a-picture detection thresholds. Referring to fig. 8b, fig. 8b shows a GO-CFAR detector with a better detection effect on clutter edge environment compared with the improved CA-CFAR detection method according to the embodiment of the present invention. At reference cell N =32, false alarm probability P _FA ＝10 ^-6 Two thresholds are obtained in this case. From the results, both detectors successfully detected the first target, but the GO-CFAR detector failed to successfully detect the second target due to the large target; the CFAR detection method improved by the embodiment of the invention reduces the threshold of the target position through the improvement of the selection of the reference unit, and successfully detects the second target. On the clutter sideIn the fringe environment, the improved algorithm and the GO-CFAR have better detection performance, but when a plurality of targets exist in the clutter fringe environment, the CA-CFAR detection method improved by the embodiment of the invention has better detection performance than the GO-CFAR detection method.

In summary, in the improved CFAR detector using the phase feature screening reference units according to the embodiments of the present invention, in the background of the uniform single-target clutter, since there is no reference unit to be removed, the detection performance is substantially the same as that of the conventional CA-CFAR detection, and the function of the original detector is not lost; the detection performance under the multi-target background and the clutter edge background is superior to that of other mean value type constant false alarm detectors.

To further verify the performance of the constant false alarm probability detection method based on phase characteristics according to the embodiment of the present invention, a set of scan data (a set of 2239 × 2224 matrices, each row represents azimuth/pulse, and each column represents range cell) is used for testing, where the P-plot of the scan data is shown in fig. 9, please refer to fig. 9, and fig. 9 shows the P-plot of the scan data according to the embodiment of the present invention. This set of scan data records information for the radar scan range 259.445 ° to 0 ° to 125.992 °.

In the embodiment of the invention, data of ten continuous echoes are arbitrarily selected for testing, a plurality of ships are arranged near the azimuth 71.781 degrees and the

distance units

1501, 1784 and 1791, the number of the reference units N is 64, four protection units are respectively selected at the left and the right due to the factors such as ship width, and the constant false alarm rate PFA is 10 ^-4 In the CA-CFAR detection method improved by the embodiment of the invention, the model statistic R ² Whether the value is greater than 0.95 and the significance probability value p is less than 0.05 are taken as criteria for linear judgment, and the test results are shown in fig. 10a and 10 b. Referring to fig. 10a, fig. 10a shows a threshold comparison between a conventional CA-CFAR device and an improved CA-CFAR method according to an embodiment of the invention. Near the 1501 distance unit, because the main target amplitude is large enough and the adjacent interference target amplitude is small, the target ship can be detected by the conventional CA-CFAR and the CA-CFAR detection method improved by the embodiment of the invention, but the CA-CFAR detection improved by the embodiment of the invention only needs a lower power value; near the

distance units

1784 and 1791 due to strong sea clutter near the primary targetOr the target is interfered strongly, the target shielding phenomenon is caused by adopting the traditional CA-CFAR detector, and the CA-CFAR detection method improved by the embodiment of the invention avoids the situation and successfully detects the target. Fig. 10b shows the test results of the improved CA-CFAR detection and the SO-CFAR with better multi-target detection effect according to the embodiment of the present invention, and fig. 10b shows the threshold comparison between the conventional SO-CFAR device and the improved CA-CFAR method according to the embodiment of the present invention. Both successfully detected the target, but a false alarm phenomenon may exist due to both lowering the threshold.

According to the embodiment of the invention, the detection background is more consistent with the use premise of the mean CFAR by deleting the distance unit where the strong clutter or the strong interference target with good phase linearity is located. As proved by analysis and verification of simulation data and actual measurement data, the improved CA-CFAR method can detect the target which cannot be detected by the traditional CA-CFAR under the multi-target environment and the clutter edge environment, effectively solves the problem of target shielding without losing the original detection performance, has good detection performance improvement compared with other mean-value CFAR, and shows the effectiveness of the improved CA-CFAR method.

According to the embodiment of the invention, pulse echoes obtained by detecting the target to be detected in the sea surface environment by the coherent radar are obtained; acquiring phase information of a plurality of continuous pulse echoes of a first distance unit of a reference unit in the pulse echoes; determining a linear relationship of the phase information of the first range bin based on the phase information of the plurality of successive pulse echoes; and under the condition that the linear relation is linearly related, deleting the pulse echo of the first distance unit, and carrying out target detection on the target to be detected through a constant false alarm probability detection method based on the rest reference units. When coherent radar is adopted to continuously detect the target to be detected in the sea surface environment, the phase characteristics of land islands, strong interference targets or sea peaks are linear under the condition that certain confidence degrees allow. Therefore, in the embodiment of the present invention, based on the phase information of a plurality of consecutive pulse echoes of a first range bin of a reference bin in the pulse echoes, a linear relationship of the phase information of the first range bin is determined; and under the condition that the linear relation is linearly related, deleting the pulse echo of the first distance unit, and carrying out target detection on the target to be detected through a constant false alarm probability detection method based on the rest reference units. Therefore, the strong interference targets or the strong sea clutter near the target to be detected are screened and then removed, the amplitude probability density function of the sea clutter is enabled to be more consistent with Rayleigh distribution, the power of the background clutter is not affected by the interference targets, the background clutter is more consistent with the precondition used by the CFAR detection method, and meanwhile, the detection threshold is prevented from being accidentally raised by the interference factors of the first distance units of which the linear relation is linearly related, so that the detection performance of target detection is improved, and missing detection is reduced.

The phase feature-based constant false alarm probability detection device provided by the present invention is described below, and the phase feature-based constant false alarm probability detection device described below and the phase feature-based constant false alarm probability detection method described above may be referred to in correspondence with each other.

Referring to fig. 11, the present invention provides a constant false alarm probability detecting device based on phase characteristics, including:

the first obtaining module 201 is configured to obtain a pulse echo obtained by detecting a target to be detected in a sea surface environment by a coherent radar;

a second obtaining module 202, configured to obtain phase information of multiple consecutive pulse echoes of a first range bin of a reference bin in the pulse echoes;

a linear relationship determining module 203 for determining a linear relationship of the phase information of the first range bin based on the phase information of the plurality of continuous pulse echoes;

and a target detection module 204, configured to delete the pulse echo of the first range bin when the linear relationship meets a preset condition, and perform target detection on the target to be detected by using a constant false alarm probability detection method based on the remaining reference bins.

In one embodiment, the linear relationship determining module is configured to determine the linear relationship of the phase information of the first range bin by calculating the phase information of the plurality of continuous pulse echoes through a unary linear regression method.

In an embodiment, the calculating the phase information of the multiple continuous pulse echoes by using a unary linear regression method to determine the linear relationship of the phase information of the first range bin specifically includes: calculating the phase information of the multiple continuous pulse echoes through a regression function to obtain model statistics; and determining the linear relation of the phase information of the first distance unit according to the size of the model statistic and the size of the significance probability value.

In one embodiment, the determining a linear relationship of the phase information of the first distance unit according to the magnitude of the model statistic and the magnitude of the significance probability value comprises:

and when the model statistic is larger than or equal to a first set threshold value and the significance probability value is smaller than or equal to a second set threshold value, determining that the phase information of the first distance unit is linearly related.

In one embodiment, the first set threshold is in the range of [0.95,0.98].

In one embodiment, the object detection module includes:

the updated background clutter power estimation value calculation module is used for averaging pulse echoes based on the rest reference units to obtain an updated background clutter power estimation value;

an update threshold calculation module, configured to obtain an update threshold based on a product of the updated background clutter power estimation value and a normalization factor;

and the final target detection module is used for carrying out target detection based on the comparison result of the updated threshold value and the pulse echo of the unit to be detected.

The constant false alarm probability detection device based on the phase characteristics obtains the pulse echo by detecting the target to be detected in the sea surface environment through the acquired coherent radar; acquiring phase information of a plurality of continuous pulse echoes of a first distance unit of a reference unit in the pulse echoes; determining a linear relationship of the phase information of the first range bin based on the phase information of the plurality of successive pulse echoes; and under the condition that the linear relation is linearly related, deleting the pulse echo of the first distance unit, and carrying out target detection on the target to be detected through a constant false alarm probability detection method based on the rest reference units. When the coherent radar is adopted to continuously detect the target to be detected in the sea surface environment, the phase characteristics of land islands, strong interference targets or sea peaks are linear under the condition that certain confidence coefficient allows. Therefore, the embodiment of the present invention determines the linear relationship of the phase information of the first range bin based on the phase information of a plurality of consecutive pulse echoes of the first range bin of the reference bin in the pulse echoes; and under the condition that the linear relation is linearly related, deleting the pulse echo of the first distance unit, and carrying out target detection on the target to be detected through a constant false alarm probability detection method based on the rest reference units. Therefore, the strong interference targets or the strong sea clutter near the target to be detected are screened and then removed, the amplitude probability density function of the sea clutter is enabled to be more consistent with Rayleigh distribution, the power of the background clutter is not affected by the interference targets, the background clutter is more consistent with the precondition used by the CFAR detection method, and meanwhile, the detection threshold is prevented from being accidentally raised by the interference factors of the first distance units of which the linear relation is linearly related, so that the detection performance of target detection is improved, and missing detection is reduced.

Fig. 12 illustrates a physical structure diagram of an electronic device, which may include, as shown in fig. 12: a processor (processor) 1210, a communication Interface (Communications Interface) 1220, a memory (memory) 1230, and a communication bus 1240, wherein the processor 1210, the communication Interface 1220, and the memory 1230 communicate with each other via the communication bus 1240. Processor 1210 may invoke logic instructions in memory 1230 to perform a method of constant false alarm probability detection based on phase signatures, the method comprising: acquiring a pulse echo obtained by detecting a target to be detected in a sea surface environment by a coherent radar; acquiring phase information of a plurality of continuous pulse echoes of a first distance unit of a reference unit in the pulse echoes; determining a linear relationship of the phase information of the first range bin based on the phase information of the plurality of successive pulse echoes; and under the condition that the linear relation is linearly related, deleting the pulse echo of the first distance unit, and carrying out target detection on the target to be detected through a constant false alarm probability detection method based on the rest reference units.

In addition, the logic instructions in the memory 1230 may be implemented in software functional units and stored in a computer readable storage medium when the logic instructions are sold or used as a stand-alone product. Based on such understanding, the technical solution of the present invention or a part thereof which substantially contributes to the prior art may be embodied in the form of a software product, which is stored in a storage medium and includes several instructions for causing a computer device (which may be a personal computer, a server, or a network device) to execute all or part of the steps of the method according to the embodiments of the present invention. And the aforementioned storage medium includes: a U-disk, a removable hard disk, a Read-Only Memory (ROM), a Random Access Memory (RAM), a magnetic disk or an optical disk, and other various media capable of storing program codes.

In another aspect, the present invention also provides a computer program product, the computer program product comprising a computer program, the computer program being stored on a non-transitory computer-readable storage medium, wherein when the computer program is executed by a processor, the computer is capable of executing the phase feature-based constant false alarm probability detection method provided by the above methods, the method comprising: acquiring a pulse echo obtained by detecting a target to be detected in a sea surface environment by a coherent radar; acquiring phase information of a plurality of continuous pulse echoes of a first distance unit of a reference unit in the pulse echoes; determining a linear relationship of the phase information of the first range bin based on the phase information of the plurality of successive pulse echoes; and under the condition that the linear relation is linearly related, deleting the pulse echo of the first distance unit, and carrying out target detection on the target to be detected through a constant false alarm probability detection method based on the rest reference units.

In yet another aspect, the present invention also provides a non-transitory computer-readable storage medium having stored thereon a computer program, which when executed by a processor, implements a phase feature-based constant false alarm probability detection method provided by the above methods, the method comprising: acquiring a pulse echo obtained by detecting a target to be detected in a sea surface environment by a coherent radar; acquiring phase information of a plurality of continuous pulse echoes of a first distance unit of a reference unit in the pulse echoes; determining a linear relationship of the phase information of the first range bin based on the phase information of the plurality of successive pulse echoes; and under the condition that the linear relation is linearly related, deleting the pulse echo of the first distance unit, and carrying out target detection on the target to be detected through a constant false alarm probability detection method based on the rest reference units.

The above-described embodiments of the apparatus are merely illustrative, and the units described as separate parts may or may not be physically separate, and parts displayed as units may or may not be physical units, may be located in one place, or may be distributed on a plurality of network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution of this embodiment. One of ordinary skill in the art can understand and implement it without inventive effort.

Through the above description of the embodiments, those skilled in the art will clearly understand that each embodiment can be implemented by software plus a necessary general hardware platform, and certainly can also be implemented by hardware. Based on the understanding, the above technical solutions substantially or otherwise contributing to the prior art may be embodied in the form of a software product, which may be stored in a computer-readable storage medium, such as ROM/RAM, magnetic disk, optical disk, etc., and includes several instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) to execute the method according to the various embodiments or some parts of the embodiments.

Finally, it should be noted that: the above examples are only intended to illustrate the technical solution of the present invention, and not to limit it; although the present invention has been described in detail with reference to the foregoing embodiments, it should be understood by those of ordinary skill in the art that: the technical solutions described in the foregoing embodiments may still be modified, or some technical features may be equivalently replaced; and such modifications or substitutions do not depart from the spirit and scope of the corresponding technical solutions of the embodiments of the present invention.

Claims

1. A constant false alarm probability detection method based on phase characteristics is characterized by comprising the following steps:

2. The phase signature based constant false alarm probability detection method of claim 1, wherein the determining the linear relationship of the phase information of the first range bin based on the phase information of the plurality of successive pulse echoes comprises:

and calculating the phase information of the plurality of continuous pulse echoes by a unary linear regression method, and determining the linear relation of the phase information of the first distance unit.

3. The phase signature-based constant false alarm probability detection method of claim 2, wherein the determining the linear relationship of the phase information of the first range bin by calculating the phase information of the plurality of successive pulse echoes by a unary linear regression method comprises:

calculating the phase information of the multiple continuous pulse echoes through a regression function to obtain model statistics;

and determining the linear relation of the phase information of the first distance unit according to the size of the model statistic and the size of the significance probability value.

4. The method of claim 3, wherein the determining the linear relationship of the phase information of the first distance unit according to the magnitude of the model statistic and the magnitude of the significance probability value comprises

5. The phase signature-based constant false alarm probability detection method of claim 4, wherein the first set threshold is in the range of [0.95,0.98].

6. The phase signature-based constant false alarm probability detection method of claim 1, wherein the target detection of the target to be detected by the constant false alarm probability detection method based on the remaining reference units comprises:

averaging pulse echoes based on the rest reference units to obtain an updated background clutter power estimation value;

obtaining an updated threshold value based on the product of the updated background clutter power estimation value and a normalization factor;

and carrying out target detection based on the comparison result of the updated threshold value and the pulse echo of the unit to be detected.

7. A constant false alarm probability detection device based on phase characteristics is characterized by comprising:

and the target detection module is used for deleting the pulse echo of the first distance unit under the condition that the linear relation accords with a preset condition, and carrying out target detection on the target to be detected through a constant false alarm probability detection method based on the rest reference units.

8. An electronic device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, wherein the processor when executing the program implements the steps of the phase feature based constant false alarm probability detection method according to any of claims 1 to 6.

9. A non-transitory computer readable storage medium having stored thereon a computer program, which when executed by a processor implements the steps of the phase feature based constant false alarm probability detection method according to any one of claims 1 to 6.

10. A computer program product comprising a computer program, characterized in that the computer program, when being executed by a processor, realizes the steps of the phase feature based constant false alarm probability detection method of any one of claims 1 to 6.