Track-level cooperative anti-jamming method based on spatio-temporal consistency judgment of multi-source information

Abstract

The application provides a point track level cooperative anti-interference method based on multi-source information space-time consistency judgment, which is applied to a multi-source information fusion system, and implementation steps include: receiving n radar point track data, and unifying point track coordinates; judging a common view area for each point track; judging a time window for the point track; judging the point tracks of multiple different radars as being homologous; sending the point track after point track arrangement to a composite tracking module, and forming a target tracking track through multi-radar data correlation filtering; judging whether the current point track is a target source of an existing track, if the current point track is a target track target, point track arrangement is performed, otherwise, the current point track is regarded as an interference point and is removed. Through the method, the cooperative anti-interference problem in a multi-radar information source fusion system is solved without sacrificing radar detection performance, a more stable anti-interference effect is provided, and the starting of radar target tracking is more stable and reliable.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of radar technology, in particular to the data processing technology in the field of radar technology. BACKGROUND

[0002] With the advent of radar technology, radar jamming and anti-jamming technology has been developing in the game with each other, and the endless jamming methods promote the improvement of radar anti-jamming technology, and the implementation of various anti-jamming measures stimulates the innovation of radar jamming technology. In the face of increasingly complex electromagnetic environment, the strength of modern radar anti-jamming capability even directly affects the trend and final outcome of the war. The traditional single radar has single frequency band data, so the target information obtained is relatively limited. The existing fusion system anti-jamming measures are mostly based on single radar signal level anti-jamming, and are very sensitive to the waveform, spectral characteristics and other information of the jamming. In the case of changing jamming pattern, the anti-jamming effect is usually unstable, and at the same time, the radar usually needs to sacrifice more resources to achieve a relatively ideal effect. However, the modern fusion system can obtain rich multi-source sensor space, time and frequency information, which provides a new way for the fusion system to resist active jamming. By analyzing the fusion of target related data under different radars, the differences between true and false targets are found out, which can effectively identify or suppress the jamming, greatly improving the survival ability of the multi-source information fusion system under active jamming.

[0003] The existing radar fusion system anti-jamming measures face the following problems: 1) a large amount of radar resources and certain detection performance need to be sacrificed; 2) sensitive to jamming pattern, anti-jamming effect is unstable. SUMMARY

[0004] In view of the deficiencies of the existing radar fusion system anti-jamming method, the present application provides a point track level cooperative anti-jamming method based on multi-source information space-time consistency judgment, which solves the cooperative anti-jamming problem in the multi-radar information source fusion system without sacrificing radar detection resources, and provides more stable anti-jamming effect.

[0005] To achieve the above technical purpose, the present application provides a point track level cooperative anti-jamming method based on multi-source information space-time consistency judgment, the process includes:

[0006] Step 1: receiving n radar point track data, and converting the data of each radar to a unified coordinate system according to the position of each sensor;

[0007] Step 2: judging the common view area for each point track: judging whether the current point track is in the scanning area of multiple radars according to the position of the current point track, if the current point track is in the scanning range of two or more radars, then the next step is performed, otherwise the point track is regarded as a non-jamming point for point track arrangement;

[0008] Step 3: Time window judgment on the point trace: according to the time stamp of the current point trace and the sector data update time of each radar, the time-related region window is calculated, and then it is judged whether there is a point trace in the time-related region window of each radar, if there is a point trace in other radars, then the next step is performed, otherwise it jumps to step 6;

[0009] Step 4: Same source judgment on the point traces of multiple different radars: the time stamp of each point trace is taken out from the point traces of different radars, and the motion model in the sense of minimum root mean square error is calculated by using the position and time stamp of these point traces. The position deviation of the current point trace from the motion model is calculated, and it is judged whether the position deviation is greater than a threshold, if the deviation is greater than the threshold, the point trace is a non-same source point trace, and the next step is performed, otherwise the point trace is a same source point trace, and the point trace is regarded as a non-interference point for point trace arrangement;

[0010] Step 5: The point trace after point trace arrangement is sent to the composite tracking module, and the composite tracking module is filtered by multi-radar data association to form a target tracking track;

[0011] Step 6: Judgment on whether the current point trace is the target source of the existing track: the real-time extrapolation module extrapolates the target track to the current time according to the motion parameters of the target tracking track and the time stamp of the current point trace, and judges whether the current point trace is within the existing target track gate according to the extrapolated position, if the current point trace is within the target track gate, it is regarded as a non-interference point for point trace arrangement, otherwise it is regarded as an interference point and is removed.

[0012] Preferably, the threshold in step 4 is selected as 3 times the measurement error of the sensor corresponding to the radar.

[0013] Preferably, the calculation method of the motion model in the sense of minimum root mean square error in step 4 is:

[0014] Step 4.1: Two point traces are taken from the point traces of multiple different radars to initialize the parameters of the uniform straight line motion model;

[0015] Step 4.2: The time stamps of the point traces of each radar are taken out and brought into the motion model to calculate the positions corresponding to the time stamps;

[0016] Step 4.3: The Euclidean distance between the position of each radar point trace and the position calculated in the last step is calculated as the distance error at each time stamp;

[0017] Step 4.4: The root mean square error is calculated by using the distance error at each time stamp;

[0018] Step 4.5: It is judged whether the root mean square error is less than a threshold, if it is less than the threshold, the step is ended, otherwise the parameters of the motion model are increased by an offset and steps 4.2 to 4.5 are repeated.

[0019] Preferably, the threshold in step 4.5 is selected as the maximum value of the measurement error of each radar sensor.

[0020] The present application eliminates the false point tracks of the interfered radars by utilizing the different interference conditions of sensors from different sources, without losing the detection performance of the radars, provides the user with a clearer radar picture and an easily found target, and makes the starting of the radar tracking of the target more stable and reliable and the target situation more easily mastered by the operator. BRIEF DESCRIPTION OF DRAWINGS

[0021] Figure 1 The preferred flowchart of the present application is shown. DETAILED DESCRIPTION

[0022] The preferred implementation process of the present application is described as follows:

[0023] Figure 1 The preferred implementation process flowchart of the present application is shown, and the specific steps include:

[0024] Step 1: receiving n radar point track data, and converting the data of each radar to a unified coordinate system according to the position of the respective sensor;

[0025] Step 2: making a common view area judgment for each point track: judging whether the current point track is in the scanning area of multiple radars according to the position of the current point track, if the current point track is in the scanning range of two or more radars, then proceeding to the next step, otherwise regarding the point track as a non-interfered point to perform point track arrangement;

[0026] Step 3: making a time window judgment for the point track: calculating the time correlation area window v m |t i -t b | according to the time stamp t m of the current i b th point track and the sector data update time t b of the b b th radar, where v b is the maximum motion speed of the target of interest, and then judging whether there is a point track in the time correlation area window of each radar, and the result is shown as follows:

[0027]

[0028] If v , then proceeding to the next step, otherwise jumping to step 6;

[0029] Step 4: making a same-source judgment for the point tracks of multiple different radars: taking out the time stamp of each point track from the point tracks of different radars, and utilizing the position and time stamp Z(t b )f b|b = 1 : n, the motion model under the minimum root mean square error is calculated as The specific calculation steps are as follows:

[0030] Step 4.1: Take two point tracks in the point tracks of multiple different radars to initialize the parameters of the uniform straight line motion model;

[0031] Step 4.2: Take the time stamps of the point tracks of each radar and bring them into the motion model to calculate the positions corresponding to the time stamps;

[0032] Step 4.3: Calculate the Euclidean distances between the positions of each radar point track and the positions calculated in the previous step, respectively, as the distance error at each time stamp;

[0033] Step 4.4: Use the distance error at each time stamp to calculate the entire root mean square error;

[0034] Step 4.5: Determine whether the root mean square error is less than a threshold, if it is less than the threshold, end the step, otherwise increase the parameters of the motion model by an offset and repeat steps 4.2 to 4.5; wherein the threshold is selected as σ b represents the measurement error of the sensor corresponding to radar b.

[0035] Calculate the position deviation |Z(t b )-X(t b )| of the current point track and the motion model, and make the following judgment:

[0036] If |Z(t b )-X(t b )|>3σ b , the point track is a non-homogeneous point track, proceed to the next step, otherwise the point track is a homogeneous point track, and the point track is regarded as a non-interference point for point track arrangement, wherein σ b represents the measurement error of the sensor corresponding to radar b.

[0037] Step 5: Send the point track after point track arrangement to the composite tracking module, and the composite tracking module forms a target tracking track after multi-radar data association filtering;

[0038] Step 6: Determine whether the current point track is the target source of the existing track: the real-time extrapolation module extrapolates the target track to the current time according to the motion parameters of the target tracking track and the time stamp of the current point track, and determines whether the current point track is within the existing target track gate according to the extrapolated position, if the current point track is within the target track gate, it is regarded as a non-interference point for point track arrangement, otherwise it is regarded as an interference point and is rejected.

[0039] The above merely describes the preferred embodiments of the present application, and it should be pointed out that, for those skilled in the art, several improvements and refinements can be made without departing from the principles of the present application, and these improvements and refinements should also be considered as falling within the protection scope of the present application.

Claims

1. A point track level cooperative anti-jamming method based on multi-source information spatio-temporal consistency decision, characterized in that: Step 1: receiving point track data of n radars, and converting the point track data of each radar from the coordinate system of the respective sensor to a unified coordinate system; Step 2: judging the common view area for each point track: judging whether the current point track exists in the scanning area of multiple radars according to the position of the current point track, if the current point track exists in the scanning area of two or more radars, then the next step is performed, otherwise the point track is regarded as a non-jamming point for point track arrangement; Step 3: judging the time window for the point track: calculating the time-related area window according to the time stamp of the current point track and the sector data update time of each radar, and then judging whether the point track exists in the time-related area window of each radar, if the point track exists in the time-related area window of other radars, then the next step is performed, otherwise jump to step 6; Step 4: making a same-source decision for the point tracks of multiple different radars: taking the time stamp of each point track from the point tracks of different radars, and calculating the motion model in the sense of minimum root mean square error using the position and time stamp of the point tracks, calculating the position deviation of the current point track from the motion model, and judging whether the position deviation is greater than a threshold, if the deviation is greater than the threshold, then the point track is a non-same-source point track, and the next step is performed, otherwise the point track is a same-source point track, and the point track is regarded as a non-jamming point for point track arrangement; Step 5: sending the point tracks after point track arrangement to a composite tracking module, and the composite tracking module forms a target tracking track after multi-radar data association filtering; Step 6: judging whether the current point track is a target source of an existing track: a real-time extrapolation module extrapolates the target track to the current time according to the motion parameters of the target tracking track and the time stamp of the current point track, and judges whether the current point track is within the existing target track gate according to the extrapolated position, if the current point track is within the target track gate, then it is regarded as a non-jamming point for point track arrangement, otherwise it is regarded as a jamming point and is removed.

2. The point track level cooperative anti-jamming method based on multi-source information spatio-temporal consistency judgment according to claim 1, characterized in that: The threshold in step 4 is selected to be 3 times the measurement error of the sensor corresponding to the radar.

3. The point track level cooperative anti-jamming method based on multi-source information spatio-temporal consistency judgment according to claim 1, characterized in that: The calculation method of the motion model in the sense of minimum root mean square error in step 4 is as follows: Step 4.1: initializing the parameters of the uniform straight line motion model in multiple different radars point tracks taking two point tracks; Step 4.2: taking the time stamp of each radar point track, and bringing it into the motion model to calculate the position corresponding to the time stamp; Step 4.3: respectively calculating the Euclidean distance between the position of each radar point track and the position calculated in the last step as the distance error at each time stamp; Step 4.4: using the distance error at each time stamp to calculate the entire root mean square error; Step 4.5: judging whether the root mean square error is less than a threshold, if less than the threshold, then ending the step, otherwise increasing the parameters of the motion model by an offset and repeating steps 4.2 to 4.

5.

4. The point track level cooperative anti-jamming method based on multi-source information spatio-temporal consistency judgment according to claim 3, characterized in that: The threshold in step 4.5 is selected to be the maximum value of the measurement errors of the sensors of the radars.

Images