CN105182311A - Omnidirectional radar data processing method and system - Google Patents

Abstract

The invention relates to the technical field of radar data processing and discloses an omnidirectional radar data processing method and system. The method comprises the following steps: determining a grid where a target locates; carrying out association with targets in the adjacent grids; carrying out polar coordinate Kalman filtering; carrying out rectangular coordinate Kalman filtering by utilizing the result of the previous stage of filtering; and carrying out target track management. According to the method and system, grid division is carried out on radar detection airspace, and neighborhood space association is carried out, so that calculation amount can be effectively reduced, and running speed of the system is improved; and a multi-step multistage Kalman filtering model is used, so that advantages of the two can be effectively utilized, and measurement accuracy of the whole radar system is improved.

Description

Omnidirectional's radar data disposal route and system

Technical field

The present invention relates to radar data processing technology field, more particularly, relate to a kind of omnidirectional radar data disposal route and system.

Background technology

Recently, along with the develop rapidly of wireless communication technology, radar has become the critical device of communication apparatus, also plays conclusive effect for business such as the quality communicated, broadcast and TVs.The process such as interconnected, the tracking that radar data process refers to that radar carries out after obtaining the measurement data such as the radial distance of target, radial velocity, orientation and the angle of pitch, filtering, level and smooth, prediction, these process effectively can suppress the stochastic error caused in measuring process, accurate estimating target position and kinematic parameter, the next position of target of prediction, forms stable targetpath.Along with the development of infotech, the research of radar data process has following developing direction: supercomputing and parallel processing; Multi-sensor information fusion is integrated with control; Search, tracking, guiding, identification, strike and commander are integrated.

Wherein, multiple target tracking, data interconnection under present stage complex environment, and the tracking of maneuvering target remains difficult point and the key of data handling system.The feature of the treatment scheme of radar data processing system and functional characteristics and radar has substantial connection, particularly multiobject tracking, correlated process and radar scanning system have close ties, and most of radar data processing system is for generally traditional mechanical scanning or Electronically Scanned Array radar.But, limit by beam angle, in a certain detect cycle, mechanical scanning radar or Electronically Scanned Array radar just detect the target of a direction, these radar data processing systems, when carrying out multiple target point mark and associating with track data, only need the target considering a direction.Omnidirectional's detection radar is from generally to scan radar different, simultaneously to the full spatial domain detection of a target in detect cycle, have the feature that it is intrinsic, need a large amount of Targets Dots that real-time process arrives simultaneously, the calculated amount that multiple target tracking is correlated with is considerably beyond general radar system.Due to the target of omnidirectional's detection radar within the scope of a certain detect cycle detection omnidirectional spatial domain, need to process a large amount of Targets Dots data simultaneously, multiple target tracking correlation computations process is complicated, calculated amount increases by geometric progression, traditional data processing technique can not meet online calculation requirement in real time, needs to adopt new technical method to improve the performance of data handling system.

Further, radar data processing system is observed object in the polar coordinate system centered by radar generally, and carry out being correlated with in rectangular coordinate system, the process such as filtering.Rectangular coordinate Kalman filter requires that the deflection measured value of radar is comparatively accurate, otherwise deflection measuring error can introduce coupled problem in X-direction and Y-direction, and the tracking performance of whole system has and significantly declines.Meanwhile, rectangular coordinate Kalman filtering is weak to maneuvering target tracking ability, needs to carry out measuring error conversion, increases the calculated amount of system.The basic skills of omnidirectional's detection radar angle measurement is measuring angle by comparing amplitude, and when engineering realizes, angle measurement accuracy is subject to the impact of the factor such as noise, landform, and single rectangular coordinate Filtering Model is not suitable with omnidirectional's radar data processing requirements.

Summary of the invention

For the above-mentioned defect of prior art, technical matters to be solved by this invention how to improve the multiple target tracking ability of omnidirectional's detection radar.

For solving the problems of the technologies described above, on the one hand, the invention provides a kind of omnidirectional radar data disposal route, comprise step:

Dissection process is carried out to the some mark data obtained, obtains the distance of a mark, position angle, and estimate a grid at mark place;

For every bit mark, judge that whether it is relevant to the target having set up flight path in contiguous grid;

Use the polar coordinates kalman filter state of some mark to target meeting target related request to upgrade, export the latest state information of target to next stage wave filter;

The rectangular coordinate kalman filter state of result to target using upper level wave filter to export upgrades, if target does not have lastest imformation, then pushes away in advance, exports the flight path information that target is up-to-date;

The life cycle of all targetpaths is managed, and exports the latest state information of all flight paths to terminal.

Preferably, described method also comprises step:

Radar detection spatial domain is divided into square or the square net region of equivalent, the length of side of described grid was determined by the maximum flying speed of the detectable target of radar and radar detection cycle.

Preferably, described relevant judgement comprises step further:

Whether the information that rough judging point mark comprises meets the movement tendency of target, gets rid of the some mark obviously not meeting target travel trend;

Nearest-neighbor association algorithm is adopted to carry out associating of a mark and targetpath.

Preferably, in the polar coordinates kalman filter state of described target, the equation of motion of described target is X (k+1)=Φ (k+1|k) X (k)+Γ (k) W (k), and it is Z (k)=H (k) X (k)+N (k) that radar system measures equation;

Wherein, $X\left(k\right)=\left(\begin{array}{c}R\left(k\right)\\ \stackrel{\·}{R}\left(k\right)\\ \mathrm{\θ}\left(k\right)\\ \stackrel{\·}{\mathrm{\θ}}\left(k\right)\end{array}\right)$ For k moment dbjective state vector, comprise radial distance components R (k), radial velocity component deflection component θ (k) and angular velocity component Φ (k+1|k) is state-transition matrix, and Γ (k) is interference matrix, and W (k) is target travel noise vector, $Z\left(k\right)=\left(\begin{array}{c}{Z}_R\left(k\right)\\ Z_{\mathrm{\θ}}\left(k\right)\end{array}\right)$ For radargrammetry vector, comprise radial distance component Z _rk () and deflection measure component Z _θk (), H (k) is calculation matrix, and N (k) is measurement noises vector.

Preferably, the described polar coordinates kalman filter state to target is carried out renewal and is comprised step:

On the one hand, in k moment estimating target state, the dbjective state of subsequent time predicted and carries out measurement prediction, newly ceasing according to measurement equation and measurement predictor calculation, to difference mark statistical interval, obtain statistical distance smallest point mark and newly cease;

On the other hand, at k moment estimate covariance, the covariance of subsequent time is predicted, newly cease covariance and calculate, calculated gains matrix subsequently;

Result according to above-mentioned two aspects upgrades dbjective state, upgrades covariance according to gain matrix simultaneously.

In another aspect of this invention, a kind of omnidirectional radar data processing system is also provided simultaneously, comprises:

Parsing module, for carrying out dissection process to the some mark data obtained, obtaining the distance of a mark, position angle, and estimating a grid at mark place;

Correlation module, for for every bit mark, judges that whether it is relevant to the target having set up flight path in contiguous grid;

Polar coordinates filtration module, for using the polar coordinates kalman filter state of some mark to target meeting target related request to upgrade, exports the latest state information of target to next stage wave filter;

Rectangular coordinate filtration module, the rectangular coordinate kalman filter state of the result exported for using upper level wave filter to target upgrades, if target does not have lastest imformation, then pushes away in advance, exports the flight path information that target is up-to-date;

Targetpath administration module, for managing the life cycle of all targetpaths, and exports the latest state information of all flight paths to terminal.

Preferably, described system also comprises:

Spatial domain divides module, and for radar detection spatial domain being divided into square or the square net region of equivalent, the length of side of described grid was determined by the maximum flying speed of the detectable target of radar and radar detection cycle.

Preferably, described correlation module comprises further:

Get rid of module, whether the information comprised for rough judging point mark meets the movement tendency of target, gets rid of the some mark obviously not meeting target travel trend;

Relating module, carries out associating of a mark and targetpath for adopting nearest-neighbor association algorithm.

Preferably, in described polar coordinates filtration module, the equation of motion of described target is X (k+1)=Φ (k+1|k) X (k)+Γ (k) W (k), and it is Z (k)=H (k) X (k)+N (k) that radar system measures equation;

Wherein, $X\left(k\right)=\left(\begin{array}{c}R\left(k\right)\\ \stackrel{\·}{R}\left(k\right)\\ \mathrm{\θ}\left(k\right)\\ \stackrel{\·}{\mathrm{\θ}}\left(k\right)\end{array}\right)$ For k moment dbjective state vector, comprise radial distance components R (k), radial velocity component deflection component θ (k) and angular velocity component Φ (k+1|k) is state-transition matrix, and Γ (k) is interference matrix, and W (k) is target travel noise vector, $Z\left(k\right)=\left(\begin{array}{c}{Z}_R\left(k\right)\\ Z_{\mathrm{\θ}}\left(k\right)\end{array}\right)$ For radargrammetry vector, comprise radial distance component Z _rk () and deflection measure component Z _θk (), H (k) is calculation matrix, and N (k) is measurement noises vector.

Preferably, described polar coordinates filtration module comprises further:

Statistical interval module, in k moment estimating target state, predicts the dbjective state of subsequent time and carries out measurement prediction, newly ceasing, to difference mark statistical interval, obtain statistical distance smallest point mark and newly cease according to measurement equation and measurement predictor calculation;

Gain module, at k moment estimate covariance, predicts the covariance of subsequent time, newly ceases covariance and calculates, calculated gains matrix subsequently;

Update module, upgrades dbjective state for the result according to above-mentioned two modules, upgrades covariance simultaneously according to gain matrix.

Compared with prior art, the present invention, by carrying out stress and strain model to radar detection spatial domain, carries out neighborhood association and effectively can reduce calculated amount, improve the travelling speed of system; Use multiple step format hierarchical Kalman Filtering Model, effectively can utilize the advantage of both sides, improve the measuring accuracy of whole radar system.

Accompanying drawing explanation

Tu1Shi omnidirectional detection radar data handling system structural drawing;

Fig. 2 is the process flow diagram of a kind of omnidirectional of the present invention radar data disposal route wherein embodiment;

Fig. 3 is the spatial domain stress and strain model areal map of a kind of omnidirectional of the present invention radar data disposal route wherein in an embodiment;

Fig. 4 is the polar coordinates Kalman filtering recurrence calculation process of a kind of omnidirectional of the present invention radar data disposal route wherein in an embodiment and field association algorithm recently.

Embodiment

Below in conjunction with the accompanying drawing in the embodiment of the present invention, the technical scheme in the embodiment of the present invention is clearly and completely described.Obviously, described embodiment is for implementing better embodiment of the present invention, and described description is to illustrate for the purpose of rule of the present invention, and is not used to limit scope of the present invention.Protection scope of the present invention should be as the criterion with the claim person of defining, and based on the embodiment in the present invention, the every other embodiment that those of ordinary skill in the art obtain under the prerequisite not making creative work, all belongs to the scope of protection of the invention.

The point mark that omnidirectional's detection radar data handling system needs real-time online process radar signal processor to send, realizes the management to multi-target traces, comprise flight path life cycle management, some mark and track association, track filtering smoothly, prediction etc.Omnidirectional's detection radar data handling system structure as shown in Figure 1, system may operate in one and is equipped with on the multiprocessor computer (or two computing machines) of real time operating system, by pci bus (or other high speed communication mode) from signal processor acquisition point mark data, track data is exported to display control terminal by network (or other communication modes), two stage filter process is deployed on two processors respectively, implementation multiple step format calculates, remote process call (RPC) communication mode can be used to carry out exchanges data between the process of two processor operations.

As shown in Figure 2, in one embodiment of the invention, omnidirectional's radar data disposal route is divided into five process steps:

(step 101) determines the grid at target place: carry out dissection process to the some mark data obtained from signal processor, obtain the distance of a mark, position angle, and estimate a grid at mark place.

(step 102) is relevant to the target in adjacent objects: for every bit mark, judges that whether it is relevant to the target having set up flight path in contiguous grid.In order to reduce calculated amount, this correlated process can be divided into two stages, whether the first stage judging point mark comprises roughly information (distance, orientation, radial velocity) can meet the movement tendency of target, gets rid of the some mark obviously not meeting target travel trend; Subordinate phase adopts conventional nearest-neighbor association algorithm, and this algorithm needs the part intermediate result (statistical interval) using polar coordinates Kalman filter.

(step 103) polar coordinates Kalman filtering: use the polar coordinates kalman filter state of some mark to target meeting target related request to upgrade, exports last state (distance value, the orientation values) information of target to next stage wave filter.

(step 104) rectangular coordinate Kalman filtering: the rectangular coordinate kalman filter state of result to target using upper level wave filter to export upgrades, if target does not have lastest imformation, then pushes away in advance, exports the flight path information that target is up-to-date.

(step 105) targetpath manages: managed by the life cycle of flight path administration module to all targetpaths, as created the flight path of fresh target, eliminating the flight path do not upgraded for a long time, and exporting the latest state information of all flight paths to terminal.

Wherein, in the method for the invention, for a large amount of somes mark data in process detect cycle, effective correlation computations amount reduced between some mark and targetpath, detection spatial domain can be divided into the square net region of equivalent, as shown in Figure 3, the length of side L of grid is by the maximum flying speed V of the detectable target of radar _maxdetermine with radar detection cycle T: L=V _max× T.

To the every bit mark that data handling system receives, carry out relevant to existing targetpath in accordance with the following methods:

Radar detection spatial domain is divided into m capable × a n row grid;

According to the distance R in a mark information and deflection θ, determine the grid (i, j) at a mark place;

To grid (i-1, j-1), (i-1, j), (i-1, j+1), (i, j-1), (i, j), (i, j+1), (i+1, j-1), (i+1, j) all targetpaths, in (i+1, j+1) and some mark carry out correlated judgment.

Further, for three-dimensional radar, square grid can be used to go to divide and three-dimensionally cover spatial domain, division methods and correlated process and two-dimensional space similar.

There is decoupling problem in X-direction and Y-direction in rectangular coordinate Kalman filter, weak to maneuvering target tracking ability, needs to carry out measuring error conversion, and calculated amount is large, but Kalman filtering can effectively eliminate target travel noise, and flight path converges faster, flight path is level and smooth.Consider that radargrammetry value R and θ is completely uncorrelated, therefore filtering can be carried out in the directionr with on θ direction respectively, namely uses polar coordinates Kalman filter.Polar coordinates wave filter can better Tracking Maneuvering Targets, and effectively eliminate the radargrammetry stochastic error in R direction and θ direction, do not need error change, time relevant, calculated amount is little, but flight path convergence slowly.The relative merits of comprehensive two kinds of wave filters, the present invention uses dual stage filter in a data processing system, and the first order is polar coordinates wave filter, and the second level is rectangular coordinate wave filter.The first order polar coordinates wave filter input be a mark, export targetpath to second level wave filter, the second level smoothly, in advance pushes away targetpath further.Because system uses nearest-neighbor association algorithm to carry out a mark and track association, so need polar coordinates wave filter to precalculate statistical interval, wherein have in polar coordinates Kalman filtering process:

Equation of motion X (k+1)=Φ (k+1|k) X (the k)+Γ (k) W (k) of target; And radar system measures equation Z (k)=H (k) X (k)+N (k).

Wherein, $X\left(k\right)=\left(\begin{array}{c}R\left(k\right)\\ \stackrel{\·}{R}\left(k\right)\\ \mathrm{\θ}\left(k\right)\\ \stackrel{\·}{\mathrm{\θ}}\left(k\right)\end{array}\right)$ For k moment dbjective state vector, comprise radial distance components R (k), radial velocity component deflection component θ (k) and angular velocity component Φ (k+1|k) is state-transition matrix, and Γ (k) is interference matrix, and W (k) is target travel noise vector, $Z\left(k\right)=\left(\begin{array}{c}{Z}_R\left(k\right)\\ Z_{\mathrm{\θ}}\left(k\right)\end{array}\right)$ For radargrammetry vector, comprise radial distance component Z _rk () and deflection measure component Z _θk (), H (k) is calculation matrix, and N (k) is measurement noises vector.

Polar coordinates Kalman filtering recurrence calculation process and recently field association algorithm as shown in Figure 4, specifically comprise:

On the one hand, in k moment estimating target state, the Target state estimator value in k moment is had the dbjective state of subsequent time (k+1 moment) is predicted there is dbjective state predicted value wherein function phi (k+1|k) is state-transition matrix; Carry out measurement prediction, have measurement predicted value wherein function H (k+1) is calculation matrix; Calculate new breath to difference mark statistical interval $d_i^2(k+1)=V_i^T(k+1)S(k+1)V_i(k+1),$ Wherein function d _ik () is statistical interval, S (k) is for newly to cease covariance matrix; Obtain statistical distance smallest point mark and newly cease V _i(k+1).

On the other hand, at k moment estimate covariance, there is the estimate covariance matrix P (k|k) in k moment; The covariance of subsequent time (k+1 moment) is predicted to have prediction covariance matrix P (k+1|k)=Φ (k+1|k) P (k|k) Φ ^t(k+1|k)+Q (k+1), wherein function Q (k+1) is motion artifacts random vector covariance matrix.

; Newly cease covariance to calculate, have new breath covariance matrix S (k+1)=H (k+1) P (k+1|k) H ^t(k+1)+C (K+1); C (k+1) is measurement noises random vector covariance matrix, calculated gains matrix K (k+1)=P (k+1|k) H ^t(k+1) S ^-1(k+1);

Dbjective state is upgraded according to the result of above-mentioned two aspects (above-mentioned two aspects can parallel processing as required)

One of ordinary skill in the art will appreciate that, the all or part of step realized in above-described embodiment method is that the hardware that can carry out instruction relevant by program has come, described program can be stored in a computer read/write memory medium, this program is when performing, comprise each step of above-described embodiment method, and described storage medium can be: ROM/RAM, magnetic disc, CD, storage card etc.Therefore, relevant technical staff in the field will be understood that corresponding with method of the present invention, and the present invention also comprises a kind of omnidirectional radar data processing system simultaneously, with said method step correspondingly, this system comprises:

Parsing module, for carrying out dissection process to the some mark data obtained, obtaining the distance of a mark, position angle, and estimating a grid at mark place;

Correlation module, for for every bit mark, judges that whether it is relevant to the target having set up flight path in contiguous grid;

Polar coordinates filtration module, for using the polar coordinates kalman filter state of some mark to target meeting target related request to upgrade, exports the latest state information of target to next stage wave filter;

Rectangular coordinate filtration module, the rectangular coordinate kalman filter state of the result exported for using upper level wave filter to target upgrades, if target does not have lastest imformation, then pushes away in advance, exports the flight path information that target is up-to-date;

Targetpath administration module, for managing the life cycle of all targetpaths, and exports the latest state information of all flight paths to terminal.

In the present invention, the multiple goal correlation computations based on nearest-neighbor association algorithm only occurs in first order wave filter, first order wave filter input point mark, and export targetpath to second level wave filter, the second level smoothly, in advance pushes away targetpath further.The transmission of dual stage filter data only occurs in flight path rank, and data volume is little, two stage filter process can be deployed on different processors and carry out multiple step format calculating, in order to improve the arithmetic speed of whole system.Multiple step format multiple-stage filtering model, relative to single Filtering Model, does not increase the computation burden of system, is to a certain degree decreasing calculated amount, reaches real-time online calculation requirement.

Compared with prior art, a kind of omnidirectional provided by the present invention radar data disposal route, owing to adopting omnidirectional's radar detection spatial domain stress and strain model method, for multiple goal correlated process, effectively can reduce calculated amount, the travelling speed of raising system, making in general real time operating system, realize the online omnidirectional of process in real time radar data becomes possibility.In addition, the overall tracking performance owing to adopting multiple step format multiple-stage filtering model can improve omnidirectional's radar data processing system.First order polar coordinates wave filter can effective Tracking Maneuvering Targets, eliminates the measurement stochastic error in R direction and θ direction, does not need error change simultaneously, put calculated amount when mark is associated with flight path little.The output flight path of second level rectangular coordinate wave filter to first order wave filter carries out further filtering, pushes away in advance, effectively eliminates the stochastic error of overall radar system, and comparatively rapid convergence, level and smooth flight path, improve the measuring accuracy of whole omnidirectional radar system.Two stage filter process is deployed on different processors and carries out multiple step format calculating, effectively can improve the arithmetic speed of system.Multiple step format multiple-stage filtering model, relative to single Filtering Model, can fully utilize the advantage of different wave filter, reduces calculated amount, improves the overall performance of omnidirectional's data handling system comprehensively, improves the measuring accuracy of omnidirectional's radar system.

Above-mentioned explanation illustrate and describes some preferred embodiments of the present invention, but as previously mentioned, be to be understood that the present invention is not limited to the form disclosed by this paper, should not regard the eliminating to other embodiments as, and can be used for other combinations various, amendment and environment, and can in invention contemplated scope described herein, changed by the technology of above-mentioned instruction or association area or knowledge.And the change that those skilled in the art carry out and change do not depart from the spirit and scope of the present invention, then all should in the protection domain of claims of the present invention.

Claims (10)

1. omnidirectional's radar data disposal route, is characterized in that, described method comprises step:

Dissection process is carried out to the some mark data obtained, obtains the distance of a mark, position angle, and estimate a grid at mark place;

For every bit mark, judge that whether it is relevant to the target having set up flight path in contiguous grid;

Use the polar coordinates kalman filter state of some mark to target meeting target related request to upgrade, export the latest state information of target to next stage wave filter;

The rectangular coordinate kalman filter state of result to target using upper level wave filter to export upgrades, if target does not have lastest imformation, then pushes away in advance, exports the flight path information that target is up-to-date;

The life cycle of all targetpaths is managed, and exports the latest state information of all flight paths to terminal.

2. the method for claim 1, is characterized in that, described method also comprises step:

Radar detection spatial domain is divided into square or the square net region of equivalent, the length of side of described grid was determined by the maximum flying speed of the detectable target of radar and radar detection cycle.

3. the method for claim 1, is characterized in that, described relevant judgement comprises step further:

Whether the information that rough judging point mark comprises meets the movement tendency of target, gets rid of the some mark obviously not meeting target travel trend;

Nearest-neighbor association algorithm is adopted to carry out associating of a mark and targetpath.

4. the method for claim 1, it is characterized in that, in the polar coordinates kalman filter state of described target, the equation of motion of described target is X (k+1)=Φ (k+1|k) X (k)+Γ (k) W (k), and it is Z (k)=H (k) X (k)+N (k) that radar system measures equation;

Wherein, $X\left(k\right)=\left(\begin{array}{c}R\left(k\right)\\ \stackrel{\·}{R}\left(k\right)\\ \mathrm{\θ}\left(k\right)\\ \stackrel{\·}{\mathrm{\θ}}\left(k\right)\end{array}\right)$ For k moment dbjective state vector, comprise radial distance components R (k), radial velocity component deflection component θ (k) and angular velocity component Φ (k+1|k) is state-transition matrix, and Γ (k) is interference matrix, and W (k) is target travel noise vector, $Z\left(k\right)=\left(\begin{array}{c}{Z}_R\left(k\right)\\ Z_{\mathrm{\θ}}\left(k\right)\end{array}\right)$ For radargrammetry vector, comprise radial distance component Z _rk () and deflection measure component Z _θk (), H (k) is calculation matrix, and N (k) is measurement noises vector.

5. the method for claim 1, is characterized in that, the described polar coordinates kalman filter state to target is carried out renewal and comprised step:

On the one hand, in k moment estimating target state, the dbjective state of subsequent time predicted and carries out measurement prediction, newly ceasing according to measurement equation and measurement predictor calculation, to difference mark statistical interval, obtain statistical distance smallest point mark and newly cease;

On the other hand, at k moment estimate covariance, the covariance of subsequent time is predicted, newly cease covariance and calculate, calculated gains matrix subsequently;

Result according to above-mentioned two aspects upgrades dbjective state, upgrades covariance according to gain matrix simultaneously.

6. omnidirectional's radar data processing system, is characterized in that, described system comprises:

Parsing module, for carrying out dissection process to the some mark data obtained, obtaining the distance of a mark, position angle, and estimating a grid at mark place;

Correlation module, for for every bit mark, judges that whether it is relevant to the target having set up flight path in contiguous grid;

Polar coordinates filtration module, for using the polar coordinates kalman filter state of some mark to target meeting target related request to upgrade, exports the latest state information of target to next stage wave filter;

Rectangular coordinate filtration module, the rectangular coordinate kalman filter state of the result exported for using upper level wave filter to target upgrades, if target does not have lastest imformation, then pushes away in advance, exports the flight path information that target is up-to-date;

Targetpath administration module, for managing the life cycle of all targetpaths, and exports the latest state information of all flight paths to terminal.

7. system as claimed in claim 6, it is characterized in that, described system also comprises:

Spatial domain divides module, and for radar detection spatial domain being divided into square or the square net region of equivalent, the length of side of described grid was determined by the maximum flying speed of the detectable target of radar and radar detection cycle.

8. system as claimed in claim 6, it is characterized in that, described correlation module comprises further:

Get rid of module, whether the information comprised for rough judging point mark meets the movement tendency of target, gets rid of the some mark obviously not meeting target travel trend;

Relating module, carries out associating of a mark and targetpath for adopting nearest-neighbor association algorithm.

9. system as claimed in claim 6, it is characterized in that, in described polar coordinates filtration module, the equation of motion of described target is X (k+1)=Φ (k+1|k) X (k)+Γ (k) W (k), and it is Z (k)=H (k) X (k)+N (k) that radar system measures equation;

Wherein, $X\left(k\right)=\left(\begin{array}{c}R\left(k\right)\\ \stackrel{\·}{R}\left(k\right)\\ \mathrm{\θ}\left(k\right)\\ \stackrel{\·}{\mathrm{\θ}}\left(k\right)\end{array}\right)$ For k moment dbjective state vector, comprise radial distance components R (k), radial velocity component deflection component θ (k) and angular velocity component Φ (k+1|k) is state-transition matrix, and Γ (k) is interference matrix, and W (k) is target travel noise vector, $Z\left(k\right)=\left(\begin{array}{c}{Z}_R\left(k\right)\\ Z_{\mathrm{\θ}}\left(k\right)\end{array}\right)$ For radargrammetry vector, comprise radial distance component Z _rk () and deflection measure component Z _θk (), H (k) is calculation matrix, and N (k) is measurement noises vector.

10. system as claimed in claim 6, it is characterized in that, described polar coordinates filtration module comprises further:

Statistical interval module, in k moment estimating target state, predicts the dbjective state of subsequent time and carries out measurement prediction, newly ceasing, to difference mark statistical interval, obtain statistical distance smallest point mark and newly cease according to measurement equation and measurement predictor calculation;

Gain module, at k moment estimate covariance, predicts the covariance of subsequent time, newly ceases covariance and calculates, calculated gains matrix subsequently;

Update module, upgrades dbjective state for the result according to above-mentioned two modules, upgrades covariance simultaneously according to gain matrix.