CN116359907A - Track processing method of radar data - Google Patents
Track processing method of radar data Download PDFInfo
- Publication number
- CN116359907A CN116359907A CN202310646164.0A CN202310646164A CN116359907A CN 116359907 A CN116359907 A CN 116359907A CN 202310646164 A CN202310646164 A CN 202310646164A CN 116359907 A CN116359907 A CN 116359907A
- Authority
- CN
- China
- Prior art keywords
- sets
- data
- trace
- point
- fitting
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
- 238000003672 processing method Methods 0.000 title abstract description 15
- 238000001514 detection method Methods 0.000 claims abstract description 71
- 238000012937 correction Methods 0.000 claims abstract description 57
- 238000012545 processing Methods 0.000 claims abstract description 56
- 238000004364 calculation method Methods 0.000 claims abstract description 34
- 238000012549 training Methods 0.000 claims abstract description 27
- 238000010276 construction Methods 0.000 claims description 93
- 238000000034 method Methods 0.000 claims description 28
- 238000001914 filtration Methods 0.000 claims description 27
- 238000012216 screening Methods 0.000 claims description 21
- 238000012360 testing method Methods 0.000 claims description 18
- 230000001133 acceleration Effects 0.000 claims description 12
- 230000000875 corresponding effect Effects 0.000 claims description 12
- 238000013528 artificial neural network Methods 0.000 claims description 8
- 230000002596 correlated effect Effects 0.000 claims description 4
- 238000006243 chemical reaction Methods 0.000 claims description 3
- 230000000694 effects Effects 0.000 abstract description 11
- 239000000523 sample Substances 0.000 description 35
- 230000006870 function Effects 0.000 description 7
- 238000007405 data analysis Methods 0.000 description 5
- 238000012986 modification Methods 0.000 description 3
- 230000004048 modification Effects 0.000 description 3
- 238000004458 analytical method Methods 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 238000013461 design Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01S—RADIO DIRECTION-FINDING; RADIO NAVIGATION; DETERMINING DISTANCE OR VELOCITY BY USE OF RADIO WAVES; LOCATING OR PRESENCE-DETECTING BY USE OF THE REFLECTION OR RERADIATION OF RADIO WAVES; ANALOGOUS ARRANGEMENTS USING OTHER WAVES
- G01S13/00—Systems using the reflection or reradiation of radio waves, e.g. radar systems; Analogous systems using reflection or reradiation of waves whose nature or wavelength is irrelevant or unspecified
- G01S13/66—Radar-tracking systems; Analogous systems
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01S—RADIO DIRECTION-FINDING; RADIO NAVIGATION; DETERMINING DISTANCE OR VELOCITY BY USE OF RADIO WAVES; LOCATING OR PRESENCE-DETECTING BY USE OF THE REFLECTION OR RERADIATION OF RADIO WAVES; ANALOGOUS ARRANGEMENTS USING OTHER WAVES
- G01S7/00—Details of systems according to groups G01S13/00, G01S15/00, G01S17/00
- G01S7/02—Details of systems according to groups G01S13/00, G01S15/00, G01S17/00 of systems according to group G01S13/00
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02A—TECHNOLOGIES FOR ADAPTATION TO CLIMATE CHANGE
- Y02A90/00—Technologies having an indirect contribution to adaptation to climate change
- Y02A90/10—Information and communication technologies [ICT] supporting adaptation to climate change, e.g. for weather forecasting or climate simulation
Landscapes
- Engineering & Computer Science (AREA)
- Radar, Positioning & Navigation (AREA)
- Remote Sensing (AREA)
- Computer Networks & Wireless Communication (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Radar Systems Or Details Thereof (AREA)
Abstract
The invention provides a track processing method of radar data, which relates to the technical field of radar data processing, and comprises the following steps: obtaining a plurality of detection data sets and a plurality of precision parameter sets; acquiring a plurality of trace point data sets; obtaining J qualified point trace sets; weighting calculation is carried out on the J qualified point trace sets, and a plurality of correction point traces are obtained; based on ensemble learning, constructing a point trace fitting model, wherein the point trace fitting model comprises a plurality of point trace fitting units; the plurality of correction tracks are input into the plurality of track fitting units to obtain a plurality of fitting tracks, the plurality of fitting tracks are weighted and adjusted according to the accuracy of the plurality of track fitting units to obtain track processing results, the technical problems that in the prior art, data precision and track processing accuracy are insufficient, operation amount is large, model training efficiency is insufficient are solved, and the technical effects of improving track processing precision and accuracy are achieved.
Description
Technical Field
The invention relates to the technical field of radar data processing, in particular to a track processing method of radar data.
Background
Multi-target tracking is one of important tasks faced by modern radars, and track processing is a key to realizing multi-target tracking technology. With the increasing complexity of radar detection environments, uncertainty of radar detection becomes more severe, and track processing is significant for radar data processing. In the prior art, the analysis flow of radar detection data is not detailed enough, so that the accuracy of the radar detection data is not enough, and meanwhile, the existing track processing method needs to analyze by using a large amount of detection data, so that the calculated amount is large.
In summary, the prior art has the technical problems of insufficient data precision and track processing accuracy, large operand and insufficient model training efficiency.
Disclosure of Invention
The invention provides a track processing method of radar data, which is used for solving the technical problems of insufficient data precision and track processing accuracy, larger operand and insufficient model training efficiency in the prior art.
According to a first aspect of the present invention, there is provided a track processing method of radar data, comprising: detecting a target to be tracked based on a plurality of detection radars to obtain a plurality of detection data sets and a plurality of precision parameter sets; acquiring a plurality of trace point data sets according to the plurality of detection data sets; screening the plurality of trace point data sets according to a plurality of preset movement characteristics to obtain J qualified trace point sets; according to the plurality of precision parameter sets, weighting calculation is carried out on the J qualified point trace sets, and a plurality of correction point traces are obtained; based on ensemble learning, constructing a point trace fitting model, wherein the point trace fitting model comprises a plurality of point trace fitting units; and inputting the plurality of correction tracks into the plurality of track fitting units to obtain a plurality of fitting tracks, and carrying out weighted adjustment on the plurality of fitting tracks according to the accuracy of the plurality of track fitting units to obtain a track processing result.
According to a second aspect of the present invention, there is provided a track processing system for radar data, comprising: the target detection module is used for detecting a target to be tracked based on a plurality of detection radars, obtaining a plurality of detection data sets and obtaining a plurality of precision parameter sets; the trace point data acquisition module is used for acquiring a plurality of trace point data sets according to the plurality of detection data sets; the spot screening module is used for screening the plurality of spot data sets according to a plurality of preset moving characteristics to obtain J qualified spot sets; the spot correction module is used for carrying out weighted calculation on the J qualified spot sets according to the plurality of precision parameter sets to obtain a plurality of correction spots; the point track fitting model construction module is used for constructing a point track fitting model based on ensemble learning, and the point track fitting model comprises a plurality of point track fitting units; the track processing module is used for inputting the plurality of correction tracks into the plurality of track fitting units to obtain a plurality of fitting tracks, and carrying out weighted adjustment on the plurality of fitting tracks according to the accuracy of the plurality of track fitting units to obtain a track processing result.
According to the track processing method of the radar data, the following beneficial effects can be achieved: 1. in the embodiment, a target to be tracked is detected based on a plurality of detection radars, a plurality of detection data sets are obtained, and a plurality of precision parameter sets are obtained; acquiring a plurality of trace point data sets according to the plurality of detection data sets; screening the plurality of trace point data sets according to a plurality of preset movement characteristics to obtain J qualified trace point sets; according to the plurality of precision parameter sets, weighting calculation is carried out on the J qualified point trace sets, and a plurality of correction point traces are obtained; based on ensemble learning, constructing a point trace fitting model, wherein the point trace fitting model comprises a plurality of point trace fitting units; and inputting the plurality of correction tracks into the plurality of track fitting units to obtain a plurality of fitting tracks, and carrying out weighted adjustment on the plurality of fitting tracks according to the accuracy of the plurality of track fitting units to obtain track processing results so as to achieve the technical effect of improving the track processing precision and accuracy.
2. The method comprises the steps of obtaining a plurality of detection data sets by carrying out noise filtering processing on reflected wave data received by a radar, obtaining a plurality of first weight distribution results and a plurality of second weight distribution results by analyzing the reflected wave data quantity and the noise filtering data quantity, and obtaining a plurality of precision parameter sets by calculation so as to achieve the technical effects of providing a data analysis basis for track processing and improving data analysis precision. And further acquiring a plurality of trace data sets according to the plurality of detection data sets, screening the plurality of trace data sets to obtain J qualified trace sets, and carrying out weighted calculation on trace data in the J qualified trace sets based on a plurality of precision parameter sets, wherein a weighted calculation result is used as the plurality of correction traces, so that the technical effects of correcting trace data and improving the accuracy of track processing are achieved.
3. The method comprises the steps of constructing a plurality of trace fitting units through integrated learning, and in the construction process, adjusting construction data of a next unit by testing the accuracy of a previous unit, wherein the higher the accuracy of the previous unit is, the smaller the construction data quantity of the next unit is, otherwise, the more the construction data quantity of the next unit is, so that the technical effects of reducing the operation quantity, guaranteeing the accuracy and improving the convergence efficiency of integrated training are achieved.
It should be understood that the description in this section is not intended to identify key or critical features of the embodiments of the invention or to delineate the scope of the invention. Other features of the present invention will become apparent from the description that follows.
Drawings
In order to more clearly illustrate the invention or the technical solutions of the prior art, the following brief description will be given of the drawings used in the description of the embodiments or the prior art, it being obvious that the drawings in the description below are only exemplary and that other drawings can be obtained from the drawings provided without the inventive effort for a person skilled in the art.
Fig. 1 is a schematic flow chart of a track processing method of radar data according to an embodiment of the present invention;
FIG. 2 is a flow chart of obtaining a plurality of probe data sets and a plurality of accuracy parameters according to an embodiment of the present invention;
FIG. 3 is a schematic flow chart of obtaining a point trace fitting model according to an embodiment of the present invention;
fig. 4 is a schematic structural diagram of a track processing system for radar data according to an embodiment of the present invention.
Reference numerals illustrate: the system comprises a target detection module 11, a trace data acquisition module 12, a trace screening module 13, a trace correction module 14, a trace fitting model construction module 15 and a trace processing module 16.
Detailed Description
Exemplary embodiments of the present invention will now be described with reference to the accompanying drawings, in which various details of the embodiments of the present invention are included to facilitate understanding, and are to be considered merely exemplary. Accordingly, those of ordinary skill in the art will recognize that various changes and modifications of the embodiments described herein can be made without departing from the scope and spirit of the invention. Also, descriptions of well-known functions and constructions are omitted in the following description for clarity and conciseness.
In order to solve the technical problems of insufficient data precision and track processing accuracy, large operand and insufficient model training efficiency in the prior art, the inventor of the invention obtains the track processing method of radar data through creative labor.
Examples
Fig. 1 is a diagram of a track processing method of radar data according to an embodiment of the present invention, where the method includes:
step S100: detecting a target to be tracked based on a plurality of detection radars to obtain a plurality of detection data sets and a plurality of precision parameter sets;
as shown in fig. 2, step S100 of the embodiment of the present invention includes:
step S110: detecting the target based on a plurality of detection radars, and obtaining a plurality of reflection wave data sets based on the received reflection waves;
step S120: acquiring a preset reflected wave data volume;
step S130: according to the ratio of the reflected wave data quantity in the plurality of reflected wave data sets to the preset reflected wave data quantity, carrying out weight distribution to obtain a plurality of first weight distribution results;
step S140: performing noise filtering processing on the plurality of reflected wave data sets to obtain a plurality of detection data sets, and obtaining a plurality of noise filtering data volume sets;
step S150: according to the size of the noise filtering data quantity in the noise filtering data quantity sets, weight distribution is carried out, and a plurality of second weight distribution results are obtained;
step S160: and calculating to obtain the plurality of precision parameter sets according to the plurality of first weight distribution results and the plurality of second weight distribution results.
Specifically, the detection radar is an electronic device for detecting a target by using electromagnetic waves, and can automatically select the type of radar existing in the market based on actual conditions, without limitation, detect targets (such as an airplane, a helicopter, etc.) to be tracked from different directions based on a plurality of detection radars, obtain a plurality of detection data sets, and obtain a plurality of precision parameter sets.
Specifically, the target is detected based on a plurality of detection radars, the plurality of detection radars emit electromagnetic waves to irradiate the target and receive reflected waves thereof, a plurality of reflected wave data sets are obtained based on the received reflected waves, the reflected wave data is the radio waves reflected by the target, and the data amount of the radio waves in the plurality of reflected wave data sets is further obtained, for example, the density or the signal energy of the radio waves can be obtained. The preset reflected wave data amount is acquired, and is set by the distances between the plurality of detection radars and the target and the size of the target, and can be set based on a person skilled in the art, for example, the preset reflected wave data amount without other electromagnetic interference in an ideal state.
And then calculating the ratio of the data quantity of the radio waves reflected by the targets corresponding to the plurality of reflected wave data sets to the preset reflected wave data quantity, wherein the ratio reflects the accuracy of the received signals, namely the greater the ratio is, the greater the data quantity of the radio waves reflected by the received targets is, the greater the detection accuracy is, the weight distribution is carried out based on the greater the ratio is, the greater the weight is, and a plurality of first weight distribution results corresponding to the plurality of reflected wave data sets are obtained.
The method further comprises the steps of performing noise filtering processing on the plurality of reflected wave data sets by adopting an existing radar signal noise filtering method (such as wavelet noise filtering, mean value filtering, median filtering and the like), obtaining a plurality of detection data sets, obtaining a plurality of noise filtering data volume sets, wherein the plurality of detection data sets are the plurality of reflected wave data sets subjected to the noise filtering processing, and the plurality of noise filtering data volume sets are noise data volume sets respectively filtered by the plurality of reflected wave data sets. And then, according to the size of the noise filtering data quantity in the noise filtering data quantity sets, carrying out weight distribution, wherein the larger the noise filtering data quantity is, the smaller the accuracy of the corresponding detection data set is, the smaller the weight is, based on the weight, a plurality of second weight distribution results corresponding to the reflection wave data sets are obtained, the reflection wave data sets are in one-to-one correspondence with the detection data sets, the products of the first weight and the second weight corresponding to each reflection wave data set are calculated according to the first weight distribution results and the second weight distribution results, a plurality of products are obtained, and then the ratio of the sum of each product and the plurality of products is calculated, so that a plurality of ratios can be obtained as precision parameters, and a plurality of precision parameter sets are obtained, so that basic data are provided for subsequent track processing.
Step S200: acquiring a plurality of trace point data sets according to the plurality of detection data sets;
specifically, the detection radar emits electromagnetic waves to irradiate the target and receives reflected waves, the information such as the distance, the azimuth and the height from the target to the electromagnetic wave emission point can be obtained by analyzing the reflected wavelengths, and therefore a map coordinate system is established, the real-time coordinate position of the target can be obtained as a point trace, the position of the target can be dynamically changed, therefore, a plurality of point traces which are changed in real time can be obtained by continuously emitting radio waves through a plurality of detection radars, the plurality of point traces form a point trace data set, the plurality of detection data sets correspond to a plurality of point trace data sets, and the plurality of point trace data sets are different due to different precision of the detection data sets.
Step S300: screening the plurality of trace point data sets according to a plurality of preset movement characteristics to obtain J qualified trace point sets;
the step S300 of the embodiment of the present invention includes:
step S310: acquiring acceleration movement characteristics and speed movement characteristics;
step S320: and judging and screening the plurality of trace data sets according to the acceleration movement characteristics and the speed movement characteristics to obtain the J qualified trace data sets.
Specifically, any one point trace data set is extracted from a plurality of point trace data sets, the acquisition time of a plurality of point traces in the point trace data set is acquired, the time difference between two continuous point traces is calculated according to the acquisition time, the acceleration movement characteristics and the speed movement characteristics corresponding to the plurality of point trace data sets are obtained by calculating the distance difference between the two point traces and the time difference, a plurality of preset movement characteristics are acquired, the plurality of preset movement characteristics comprise maximum acceleration, maximum speed, minimum speed and the like, and the method is self-set based on actual conditions. Judging and screening the plurality of trace data sets based on a plurality of preset movement features, acceleration movement features and speed movement features to obtain the J qualified trace data sets, wherein the specific process can be expressed by the following formula:
Step S400: according to the plurality of precision parameter sets, weighting calculation is carried out on the J qualified point trace sets, and a plurality of correction point traces are obtained;
the step S400 of the embodiment of the present invention includes:
step S410: acquiring J precision parameter sets corresponding to the J qualified point trace sets in the precision parameter sets;
step S420: and weighting calculation is carried out on the point trace data in the J qualified point trace sets by adopting the size of the precision parameter in each precision parameter set in the J precision parameter sets, so as to obtain the plurality of correction point traces.
Specifically, J precision parameter sets corresponding to the J qualified point trace sets in the precision parameter sets are obtained, the sizes of the precision parameters in the J precision parameter sets are adopted to carry out weight setting on the J qualified point trace sets, the larger the precision parameters are, the larger the corresponding weights are, the position of the target changes in real time in the process of detecting the target by the detection radar, so that each qualified point trace set comprises a plurality of point traces, point trace data in the J qualified point trace sets corresponds to each other one by one, for example, when an aircraft arrives at any position, J position coordinates (namely point trace data) are obtained, but the J position coordinates are obtained based on different detection radars, under ideal conditions, namely under the condition that the detection data of different detection radars are completely accurate, the J position coordinates are consistent, but because the precision of the detection data of the plurality of radars is different, the J position coordinates have difference, so based on the size of precision parameters in J precision parameter sets, J point trace data of the same position are extracted according to the J qualified point trace sets and weighted calculation is carried out, specifically, J first point trace data in the J qualified point trace sets are respectively extracted, weighted calculation is carried out on the J first point trace data to obtain a corrected point trace, J second point trace data in the J qualified point trace sets are continuously extracted, weighted calculation is carried out, and so on, weighted calculation is carried out on the point trace data in the J qualified point trace sets, so that the plurality of corrected point traces can be obtained, the correction of the point trace data is realized, and the accuracy of track processing is improved.
Step S500: based on ensemble learning, constructing a point trace fitting model, wherein the point trace fitting model comprises a plurality of point trace fitting units;
as shown in fig. 3, step S500 in the embodiment of the present invention includes:
step S510: acquiring a plurality of sample correction trace point sets according to radar detection data in the historical time;
step S520: respectively constructing tracks according to the plurality of sample correction point track sets to obtain a plurality of sample tracks;
step S530: randomly selecting K groups of construction data in the plurality of sample correction trace sets and the plurality of sample tracks, and acquiring K first coefficients, wherein each first coefficient is 1/K, and K is an integer which is more than 1 and less than the number of the plurality of sample correction trace sets;
step S540: constructing and training to obtain a first point trace fitting unit in the plurality of point trace fitting units based on a feedforward neural network by adopting the K group construction data;
step S550: testing the first track fitting unit by adopting the K-set configuration data to obtain a first accuracy rate;
step S560: according to the deviation between the first accuracy and the preset accuracy, adjusting the K-set construction data to obtain M-set construction data, and calculating to obtain M second coefficients, wherein M is an integer greater than or equal to K and smaller than the number of a plurality of sample correction trace sets or an integer smaller than K and larger than 0;
Step S570: constructing and training the construction data of the M groups according to the M second coefficients and based on the M second coefficients to obtain a second point track fitting unit, wherein the training computational power resources of the construction data of each group are positively correlated with the sizes of the second coefficients;
step S580: and testing the second point track fitting unit by adopting the M-group construction data to obtain a second accuracy rate, and continuously constructing and obtaining the plurality of point track fitting units to obtain the point track fitting model.
The step S550 of the embodiment of the present invention includes:
step S551: testing the first track fitting unit by adopting the K-set construction data to obtain K predicted tracks, and calculating the first accuracy according to K sample tracks in the K-set construction data, wherein the first accuracy is represented by the following formula:
wherein ,
for the first accuracy, Q is the statistic of the i-th set of construction data when tested.
The step S560 of the embodiment of the present invention includes:
step S561: calculating the ratio of the first accuracy to the preset accuracy, and calculating and adjusting K to obtain M;
step S562: according to the K first coefficients and the first accuracy, K primary second coefficients are calculated and obtained, wherein the K primary second coefficients are represented by the following formula:
wherein ,
to construct the data x divided by the second coefficient, +.>
To construct the first coefficient of data x +.>
Is the sum of K first coefficients, +.>
The ratio of the first accuracy to the sum of the accuracy of all the constructed point track fitting units is set;
step S563: when M is greater than or equal to K, M-K primary second coefficients of which the M-K group construction data are 1/M are obtained, the M secondary coefficients are obtained through calculation by combining the K primary second coefficients, when M is smaller than K, M primary second coefficients are obtained according to the K primary second coefficients, and the M secondary coefficients are obtained through calculation.
Specifically, the learning task is completed by constructing and combining a plurality of machine learners, and in this embodiment, the point trace fitting model is obtained by integrating a plurality of point trace fitting units.
Specifically, a plurality of sample correction point trace sets are obtained according to radar detection data in a historical time (such as the past month), and track construction is performed based on the prior art according to the plurality of sample correction point trace sets, so as to obtain a plurality of sample tracks. K groups of construction data are randomly selected from the plurality of sample correction point trace sets and the plurality of sample tracks, each group of construction data comprises one sample correction point trace set and a corresponding sample track, K first coefficients are obtained, each first coefficient is 1/K, and K is an integer which is more than 1 and less than the number of the plurality of sample correction point trace sets. The K-group construction data are adopted, first point trace fitting units in the point trace fitting units are constructed and trained based on a feedforward neural network, the feedforward neural network is an artificial neural network, a common feedforward neural network comprises a perceptron, a BP (back propagation) network and the like, the first point trace fitting units constructed based on the feedforward neural network can form network parameters such as weights, thresholds and the like connected between simple units in the supervision training process, and the first point trace fitting units after training can carry out complex nonlinear logic operation according to input data to output a predicted flight path.
Further testing the first track fitting unit by adopting the K-set construction data to obtain a first accuracy rate, wherein the specific process is as follows: the first track fitting unit is tested by adopting the K-set construction data, namely, the K sample correction track sets in the K-set construction data are respectively input into the first track fitting unit, the K predicted tracks are output, the K sample tracks and the K predicted tracks in the K-set construction data are compared, and the first accuracy rate is calculated by the following formula:
wherein ,
for the first accuracy, Q is the statistical value of the i-th set of construction data when testing, that is, when the predicted track is consistent with the sample track, Q takes 1, otherwise Q takes 0, the number of Q taking 1 is counted, and the ratio is made with K, and the obtained ratio result is the first accuracy.
Further, according to the deviation between the first accuracy and the preset accuracy, the K-set construction data is adjusted to obtain M-set construction data, and M second coefficients are obtained by calculation, where M is an integer greater than or equal to K and less than the number of the plurality of sample correction trace sets, or an integer less than K and greater than 0, and the specific process is as follows:
firstly, calculating the ratio of the first accuracy to the preset accuracy, calculating and adjusting K to obtain M, and colloquially, after the first track fitting unit is constructed, adjusting K group construction data, and selecting M group construction data from the plurality of sample correction track sets and the plurality of sample tracks, wherein the M group construction data is used for constructing a second track fitting unit. The preset accuracy is set according to the actual situation, for example, the preset accuracy is set to 80%, if the first accuracy is greater than the preset accuracy, M is an integer smaller than K and greater than 0, M sets of construction data are randomly selected from the K sets of construction data, and the test accuracy of the last unit, namely the first trace fitting unit, is higher, so that the number of construction data can be reduced when the second trace fitting unit is constructed, and the higher the first accuracy is, the smaller M is. For example, if the ratio of the first accuracy to the preset accuracy is 1.1, m=k/1.1 and rounding is performed.
If the first accuracy is smaller than or equal to the preset accuracy, M is an integer greater than or equal to K and smaller than the number of the plurality of sample correction trace sets, because the accuracy of the first trace fitting unit is poor at this time, the number of the construction data needs to be increased when the second trace fitting unit is constructed, and therefore the accuracy of the second trace fitting unit is improved. And randomly selecting M-K set of construction data in the plurality of sample correction track sets and the plurality of sample tracks, combining the K set of construction data to obtain M set of construction data, wherein the K set of construction data and the M set of construction data comprise the same data, if the same set of construction data are inconsistent with the sample tracks in the test of the previous unit (such as the first track fitting unit), the fact that the set of construction data fail in the prediction of the previous unit is explained, and giving greater weight to the set of construction data in the training of the next unit so as to improve the accuracy of the whole model on the set of construction data, otherwise, giving smaller weight to give out training resources, thereby achieving the technical effects of ensuring the accuracy and improving the convergence efficiency of the integrated training. The training resource may be, for example, the number of training sessions, etc.
Further, according to the K first coefficients and the first accuracy, K primary second coefficients are calculated and obtained, wherein the following formula is as follows:
wherein ,
to construct the data x divided by the second coefficient, +.>
To construct the first coefficient of data x +.>
Is the sum of K first coefficients, +.>
Is the ratio of the first accuracy to the sum of the accuracy of all the trace-fitting units constructed.
When M is greater than or equal to K, namely when the amount of construction data is increased, M-set construction data are composed of K-set construction data of a first track fitting unit and newly added M-K-set construction data, the M-K-set construction data are newly added construction data, so that a first coefficient cannot be obtained, namely an initial second coefficient cannot be calculated by using the formula, M-K initial second coefficients of the newly added M-K-set construction data are initialized to 1/M, meanwhile, the initial second coefficients of the K-set construction data are obtained by adopting the formula to calculate, the sum of the M-K initial second coefficients and the K initial second coefficients is further calculated, and then the ratio of each initial second coefficient to the sum is calculated to obtain M second coefficients; when M is smaller than K, namely, the construction data quantity is reduced, the M-set construction data at the moment is a subset of the K-set construction data of the first track fitting unit, M primary second coefficients are obtained directly according to the K primary second coefficients through calculation according to the formula, the sum of the M primary second coefficients is further calculated, and the ratio of the M primary second coefficients to the sum of the M primary second coefficients is calculated respectively, so that the M second coefficients can be obtained.
And constructing and training the construction data of each group of construction data based on the M second coefficients to obtain a second point track fitting unit, wherein the training calculation resources of each group of construction data are positively correlated with the size of the second coefficients, that is, the larger the second coefficients are, the more the calculation training resources are, and the calculation training resources can be understood as calculation capability, such as training times.
And testing the second point trace fitting unit by adopting the M group of construction data, calculating to obtain the second accuracy rate in the same mode as the first accuracy rate, comparing the second accuracy rate with the preset accuracy rate, determining the quantity of construction data of a third point trace fitting unit according to the deviation of the second accuracy rate and the preset accuracy rate, carrying out the third point trace fitting unit, and so on, continuing to construct the next point trace fitting unit, thereby constructing and obtaining the plurality of point trace fitting units, and integrating the plurality of point trace fitting units to obtain the point trace fitting model.
Step S600: and inputting the plurality of correction tracks into the plurality of track fitting units to obtain a plurality of fitting tracks, and carrying out weighted adjustment on the plurality of fitting tracks according to the accuracy of the plurality of track fitting units to obtain a track processing result.
The step S600 of the embodiment of the present invention includes:
step S610: performing polynomial conversion processing on the fitting tracks to obtain a plurality of polynomial sets;
step S620: according to the accuracy of the plurality of point trace fitting units, weight distribution is carried out, and a plurality of fitting weight values are obtained;
step S630: and weighting parameters in the polynomial sets by adopting the fitting weight values to obtain a fitting polynomial set, and converting to obtain the track processing result.
Specifically, the plurality of correction tracks are input into the plurality of track fitting units to obtain a plurality of fitting tracks, wherein the fitting tracks refer to tracks of a target (namely a flying object) flying according to the plurality of track prediction, and the plurality of fitting tracks are weighted and adjusted according to the accuracy of the plurality of track fitting units to obtain a track processing result.
And according to the accuracy of the plurality of point track fitting units, the process of weighting and adjusting the plurality of fitting tracks is as follows: the curve polynomial fitting is a common data analysis method, and can be used for converting curves and polynomial functions, such as curve polynomial fitting by using a least square method in the prior art, in this embodiment, the multiple fitting tracks can be regarded as multiple track curves, the curves can be converted into polynomial functions, such as polynomial of nth power function, thereby obtaining multiple polynomial sets, and conventionally, a polynomial function is fitted through a series of correction points on the track curves, so that the function can best describe the trend and rule of the tracks, and each polynomial function contains multiple polynomials, thereby obtaining multiple polynomial sets through combination. In the process of constructing a plurality of point track fitting units, testing the accuracy of each point track fitting unit, carrying out weight distribution according to the accuracy of the plurality of point track fitting units, wherein the larger the accuracy is, the larger the weight value is, so as to obtain a plurality of fitting weight values, weighting parameters in a plurality of polynomial sets by adopting the plurality of fitting weight values, so as to obtain a fitting polynomial set, namely, weighting a plurality of polynomials, and then converting the fitting polynomial set into a track curve as a track processing result, thereby realizing track processing and achieving the technical effect of improving the track processing precision and accuracy.
Based on the analysis, the invention provides a track processing method of radar data, which has the following beneficial effects:
1. in the embodiment, a target to be tracked is detected based on a plurality of detection radars, a plurality of detection data sets are obtained, and a plurality of precision parameter sets are obtained; acquiring a plurality of trace point data sets according to the plurality of detection data sets; screening the plurality of trace point data sets according to a plurality of preset movement characteristics to obtain J qualified trace point sets; according to the plurality of precision parameter sets, weighting calculation is carried out on the J qualified point trace sets, and a plurality of correction point traces are obtained; based on ensemble learning, constructing a point trace fitting model, wherein the point trace fitting model comprises a plurality of point trace fitting units; and inputting the plurality of correction tracks into the plurality of track fitting units to obtain a plurality of fitting tracks, and carrying out weighted adjustment on the plurality of fitting tracks according to the accuracy of the plurality of track fitting units to obtain track processing results so as to achieve the technical effect of improving the track processing precision and accuracy.
2. The method comprises the steps of obtaining a plurality of detection data sets by carrying out noise filtering processing on reflected wave data received by a radar, obtaining a plurality of first weight distribution results and a plurality of second weight distribution results by analyzing the reflected wave data quantity and the noise filtering data quantity, and obtaining a plurality of precision parameter sets by calculation so as to achieve the technical effects of providing a data analysis basis for track processing and improving data analysis precision. And further acquiring a plurality of point trace data sets according to the plurality of detection data sets, screening the plurality of point trace data sets to obtain J qualified point trace sets, and carrying out weighted calculation on the J qualified point trace sets based on a plurality of precision parameter sets, wherein a weighted calculation result is used as the plurality of correction point traces, so that the technical effects of correcting the point trace data and improving the accuracy of track processing are achieved.
3. The method comprises the steps of constructing a plurality of trace fitting units through integrated learning, and in the construction process, adjusting construction data of a next unit by testing the accuracy of a previous unit, wherein the higher the accuracy of the previous unit is, the smaller the construction data quantity of the next unit is, otherwise, the more the construction data quantity of the next unit is, so that the technical effects of reducing the operation quantity, guaranteeing the accuracy and improving the convergence efficiency of integrated training are achieved.
Examples
Based on the same inventive concept as the track processing method of radar data in the foregoing embodiment, as shown in fig. 4, the present invention further provides a track processing system of radar data, where the system includes:
the target detection module 11 is configured to detect a target to be tracked based on a plurality of detection radars, obtain a plurality of detection data sets, and obtain a plurality of precision parameter sets;
a trace data acquisition module 12, where the trace data acquisition module 12 is configured to acquire a plurality of trace data sets according to the plurality of detection data sets;
the spot screening module 13 is used for screening the plurality of spot data sets according to a plurality of preset movement characteristics to obtain J qualified spot sets;
The trace correction module 14 is configured to perform weighted calculation on the J qualified trace sets according to the multiple precision parameter sets, so as to obtain multiple correction traces;
the point trace fitting model construction module 15 is used for constructing a point trace fitting model based on ensemble learning, wherein the point trace fitting model construction module 15 comprises a plurality of point trace fitting units;
the track processing module 16 is configured to input the plurality of correction tracks into the plurality of track fitting units to obtain a plurality of fitted tracks, and weight-adjust the plurality of fitted tracks according to the accuracy of the plurality of track fitting units to obtain a track processing result.
Further, the object detection module 11 is further configured to:
detecting the target based on a plurality of detection radars, and obtaining a plurality of reflection wave data sets based on the received reflection waves;
acquiring a preset reflected wave data volume;
according to the ratio of the reflected wave data quantity in the plurality of reflected wave data sets to the preset reflected wave data quantity, carrying out weight distribution to obtain a plurality of first weight distribution results;
Performing noise filtering processing on the plurality of reflected wave data sets to obtain a plurality of detection data sets, and obtaining a plurality of noise filtering data volume sets;
according to the size of the noise filtering data quantity in the noise filtering data quantity sets, weight distribution is carried out, and a plurality of second weight distribution results are obtained;
and calculating to obtain the plurality of precision parameter sets according to the plurality of first weight distribution results and the plurality of second weight distribution results.
Further, the trace point screening module 13 is further configured to:
acquiring acceleration movement characteristics and speed movement characteristics;
and judging and screening the plurality of trace data sets according to the acceleration movement characteristics and the speed movement characteristics to obtain the J qualified trace data sets.
Further, the trace point correction module 14 is further configured to:
acquiring J precision parameter sets corresponding to the J qualified point trace sets in the precision parameter sets;
and weighting calculation is carried out on the point trace data in the J qualified point trace sets by adopting the size of the precision parameter in each precision parameter set in the J precision parameter sets, so as to obtain the plurality of correction point traces.
Further, the track fitting model building module 15 is further configured to:
Acquiring a plurality of sample correction trace point sets according to radar detection data in the historical time;
respectively constructing tracks according to the plurality of sample correction point track sets to obtain a plurality of sample tracks;
randomly selecting K groups of construction data in the plurality of sample correction trace sets and the plurality of sample tracks, and acquiring K first coefficients, wherein each first coefficient is 1/K, and K is an integer which is more than 1 and less than the number of the plurality of sample correction trace sets;
constructing and training to obtain a first point trace fitting unit in the plurality of point trace fitting units based on a feedforward neural network by adopting the K group construction data;
testing the first track fitting unit by adopting the K-set configuration data to obtain a first accuracy rate;
according to the deviation between the first accuracy and the preset accuracy, adjusting the K-set construction data to obtain M-set construction data, and calculating to obtain M second coefficients, wherein M is an integer greater than or equal to K and smaller than the number of a plurality of sample correction trace sets or an integer smaller than K and larger than 0;
constructing and training the construction data of the M groups according to the M second coefficients and based on the M second coefficients to obtain a second point track fitting unit, wherein the training computational power resources of the construction data of each group are positively correlated with the sizes of the second coefficients;
And testing the second point track fitting unit by adopting the M-group construction data to obtain a second accuracy rate, and continuously constructing and obtaining the plurality of point track fitting units to obtain the point track fitting model.
Further, the track fitting model building module 15 is further configured to:
calculating the ratio of the first accuracy to the preset accuracy, and calculating and adjusting K to obtain M;
according to the K first coefficients and the first accuracy, K primary second coefficients are calculated and obtained, wherein the K primary second coefficients are represented by the following formula:
wherein ,
to construct the data x divided by the second coefficient, +.>
To construct the first coefficient of data x +.>
Is the sum of K first coefficients, +.>
The ratio of the first accuracy to the sum of the accuracy of all the constructed point track fitting units is set;
when M is greater than or equal to K, M-K primary second coefficients of which the M-K group construction data are 1/M are obtained, the M secondary coefficients are obtained through calculation by combining the K primary second coefficients, when M is smaller than K, M primary second coefficients are obtained according to the K primary second coefficients, and the M secondary coefficients are obtained through calculation.
Further, the track fitting model building module 15 is further configured to:
testing the first track fitting unit by adopting the K-set construction data to obtain K predicted tracks, and calculating the first accuracy according to K sample tracks in the K-set construction data, wherein the first accuracy is represented by the following formula:
wherein ,
for the first accuracy, Q is the statistic of the i-th set of construction data when tested.
Further, the track processing module 16 is further configured to:
performing polynomial conversion processing on the fitting tracks to obtain a plurality of polynomial sets;
according to the accuracy of the plurality of point trace fitting units, weight distribution is carried out, and a plurality of fitting weight values are obtained;
and weighting parameters in the polynomial sets by adopting the fitting weight values to obtain a fitting polynomial set, and converting to obtain the track processing result.
The specific example of the track processing method for radar data in the first embodiment is also applicable to the track processing system for radar data in the present embodiment, and the detailed description of the track processing method for radar data in the present embodiment is clearly known to those skilled in the art, so the detailed description thereof is omitted herein for brevity.
It should be appreciated that various forms of the flows shown above may be used to reorder, add, or delete steps. For example, the steps described in the present invention may be performed in parallel, sequentially, or in a different order, as long as the desired results of the technical solution disclosed in the present invention can be achieved, and are not limited herein.
The above embodiments do not limit the scope of the present invention. It will be apparent to those skilled in the art that various modifications, combinations, sub-combinations and alternatives are possible, depending on design requirements and other factors. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the present invention should be included in the scope of the present invention.
Claims (9)
1. A method of track processing of radar data, the method comprising:
detecting a target to be tracked based on a plurality of detection radars to obtain a plurality of detection data sets and a plurality of precision parameter sets;
acquiring a plurality of trace point data sets according to the plurality of detection data sets;
screening the plurality of trace point data sets according to a plurality of preset movement characteristics to obtain J qualified trace point sets;
according to the plurality of precision parameter sets, weighting calculation is carried out on the J qualified point trace sets, and a plurality of correction point traces are obtained;
based on ensemble learning, constructing a point trace fitting model, wherein the point trace fitting model comprises a plurality of point trace fitting units;
and inputting the plurality of correction tracks into the plurality of track fitting units to obtain a plurality of fitting tracks, and carrying out weighted adjustment on the plurality of fitting tracks according to the accuracy of the plurality of track fitting units to obtain a track processing result.
2. The method of claim 1, wherein detecting an object to be tracked based on a plurality of detection radars, obtaining a plurality of detection data sets, and obtaining a plurality of precision parameter sets, comprises:
detecting the target based on a plurality of detection radars, and obtaining a plurality of reflection wave data sets based on the received reflection waves;
acquiring a preset reflected wave data volume;
according to the ratio of the reflected wave data quantity in the plurality of reflected wave data sets to the preset reflected wave data quantity, carrying out weight distribution to obtain a plurality of first weight distribution results;
performing noise filtering processing on the plurality of reflected wave data sets to obtain a plurality of detection data sets, and obtaining a plurality of noise filtering data volume sets;
according to the size of the noise filtering data quantity in the noise filtering data quantity sets, weight distribution is carried out, and a plurality of second weight distribution results are obtained;
and calculating to obtain the plurality of precision parameter sets according to the plurality of first weight distribution results and the plurality of second weight distribution results.
3. The method of claim 1, wherein screening the plurality of trace data sets according to a plurality of preset movement characteristics to obtain J qualified trace sets comprises:
Acquiring acceleration movement characteristics and speed movement characteristics;
and judging and screening the plurality of trace data sets according to the acceleration movement characteristics and the speed movement characteristics to obtain the J qualified trace data sets.
4. The method of claim 1, wherein weighting the J qualified trace sets based on the plurality of precision parameter sets to obtain a plurality of correction traces comprises:
acquiring J precision parameter sets corresponding to the J qualified point trace sets in the precision parameter sets;
and weighting calculation is carried out on the point trace data in the J qualified point trace sets by adopting the size of the precision parameter in each precision parameter set in the J precision parameter sets, so as to obtain the plurality of correction point traces.
5. The method of claim 1, wherein constructing a point trace fitting model based on ensemble learning, the point trace fitting model including a plurality of point trace fitting units therein, comprises:
acquiring a plurality of sample correction trace point sets according to radar detection data in the historical time;
respectively constructing tracks according to the plurality of sample correction point track sets to obtain a plurality of sample tracks;
Randomly selecting K groups of construction data in the plurality of sample correction trace sets and the plurality of sample tracks, and acquiring K first coefficients, wherein each first coefficient is 1/K, and K is an integer which is more than 1 and less than the number of the plurality of sample correction trace sets;
constructing and training to obtain a first point trace fitting unit in the plurality of point trace fitting units based on a feedforward neural network by adopting the K group construction data;
testing the first track fitting unit by adopting the K-set configuration data to obtain a first accuracy rate;
according to the deviation between the first accuracy and the preset accuracy, adjusting the K-set construction data to obtain M-set construction data, and calculating to obtain M second coefficients, wherein M is an integer greater than or equal to K and smaller than the number of a plurality of sample correction trace sets or an integer smaller than K and larger than 0;
constructing and training the construction data of the M groups according to the M second coefficients and based on the M second coefficients to obtain a second point track fitting unit, wherein the training computational power resources of the construction data of each group are positively correlated with the sizes of the second coefficients;
and testing the second point track fitting unit by adopting the M-group construction data to obtain a second accuracy rate, and continuously constructing and obtaining the plurality of point track fitting units to obtain the point track fitting model.
6. The method of claim 5, wherein adjusting the K-set construction data according to the deviation of the first accuracy rate from a preset accuracy rate, obtaining M-set construction data, and calculating to obtain M second coefficients, comprises:
calculating the ratio of the first accuracy to the preset accuracy, and calculating and adjusting K to obtain M;
according to the K first coefficients and the first accuracy, K primary second coefficients are calculated and obtained, wherein the K primary second coefficients are represented by the following formula:
wherein ,
to construct the data x divided by the second coefficient, +.>
To construct the first coefficient of data x +.>
Is the sum of K first coefficients, +.>
The ratio of the first accuracy to the sum of the accuracy of all the constructed point track fitting units is set;
when M is greater than or equal to K, M-K primary second coefficients of which the M-K group construction data are 1/M are obtained, the M secondary coefficients are obtained through calculation by combining the K primary second coefficients, when M is smaller than K, M primary second coefficients are obtained according to the K primary second coefficients, and the M secondary coefficients are obtained through calculation.
7. The method of claim 5, wherein testing the first trace-fitting unit with the K-set of construction data to obtain a first accuracy comprises:
Testing the first track fitting unit by adopting the K-set construction data to obtain K predicted tracks, and calculating the first accuracy according to K sample tracks in the K-set construction data, wherein the first accuracy is represented by the following formula:
wherein ,
for the first accuracy, Q is the statistic of the i-th set of construction data when tested.
8. The method of claim 1, wherein weighting the plurality of fitted tracks according to the accuracy of the plurality of spot-fit units comprises:
performing polynomial conversion processing on the fitting tracks to obtain a plurality of polynomial sets;
according to the accuracy of the plurality of point trace fitting units, weight distribution is carried out, and a plurality of fitting weight values are obtained;
and weighting parameters in the polynomial sets by adopting the fitting weight values to obtain a fitting polynomial set, and converting to obtain the track processing result.
9. A track processing system for radar data, the system comprising:
the target detection module is used for detecting a target to be tracked based on a plurality of detection radars, obtaining a plurality of detection data sets and obtaining a plurality of precision parameter sets;
The trace point data acquisition module is used for acquiring a plurality of trace point data sets according to the plurality of detection data sets;
the spot screening module is used for screening the plurality of spot data sets according to a plurality of preset moving characteristics to obtain J qualified spot sets;
the spot correction module is used for carrying out weighted calculation on the J qualified spot sets according to the plurality of precision parameter sets to obtain a plurality of correction spots;
the point track fitting model construction module is used for constructing a point track fitting model based on ensemble learning, and the point track fitting model comprises a plurality of point track fitting units;
the track processing module is used for inputting the plurality of correction tracks into the plurality of track fitting units to obtain a plurality of fitting tracks, and carrying out weighted adjustment on the plurality of fitting tracks according to the accuracy of the plurality of track fitting units to obtain a track processing result.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310646164.0A CN116359907B (en) | 2023-06-02 | 2023-06-02 | Track processing method of radar data |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310646164.0A CN116359907B (en) | 2023-06-02 | 2023-06-02 | Track processing method of radar data |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN116359907A true CN116359907A (en) | 2023-06-30 |
| CN116359907B CN116359907B (en) | 2023-08-15 |
Family
ID=86913382
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310646164.0A Active CN116359907B (en) | 2023-06-02 | 2023-06-02 | Track processing method of radar data |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116359907B (en) |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20200151610A1 (en) * | 2018-11-09 | 2020-05-14 | Industrial Technology Research Institute | Ensemble learning predicting method and system |
| CN111898749A (en) * | 2020-05-28 | 2020-11-06 | 南京理工雷鹰电子科技有限公司 | Unmanned aerial vehicle track prediction method based on neural network |
| CN113128342A (en) * | 2021-03-19 | 2021-07-16 | 中国人民解放军战略支援部队信息工程大学 | Track data preprocessing method and aerial target identification method |
| CN115563574A (en) * | 2022-09-20 | 2023-01-03 | 沈阳飞机设计研究所扬州协同创新研究院有限公司 | Multi-sensor air target point trace data fusion method based on comprehensive criterion |
| CN116106890A (en) * | 2023-02-08 | 2023-05-12 | 南京莱斯电子设备有限公司 | Radar target track starting method based on quantum particle swarm and LGBM |
-
2023
- 2023-06-02 CN CN202310646164.0A patent/CN116359907B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20200151610A1 (en) * | 2018-11-09 | 2020-05-14 | Industrial Technology Research Institute | Ensemble learning predicting method and system |
| CN111898749A (en) * | 2020-05-28 | 2020-11-06 | 南京理工雷鹰电子科技有限公司 | Unmanned aerial vehicle track prediction method based on neural network |
| CN113128342A (en) * | 2021-03-19 | 2021-07-16 | 中国人民解放军战略支援部队信息工程大学 | Track data preprocessing method and aerial target identification method |
| CN115563574A (en) * | 2022-09-20 | 2023-01-03 | 沈阳飞机设计研究所扬州协同创新研究院有限公司 | Multi-sensor air target point trace data fusion method based on comprehensive criterion |
| CN116106890A (en) * | 2023-02-08 | 2023-05-12 | 南京莱斯电子设备有限公司 | Radar target track starting method based on quantum particle swarm and LGBM |
Non-Patent Citations (5)
| Title |
|---|
| ZHANCHUN GAO: "Aviation Surveillance Information Fusion Based on Ensemble Learning", 2020 IEEE INTERNATIONAL CONFERENCE ON INFORMATION TECHNOLOGY, BIG DATA AND ARTIFICIAL INTELLIGENCE (ICIBA) * |
| 吴显亮;官慧峰;尹良泽;: "多雷达系统数据融合与航迹预测", 数学的实践与认识, no. 15 * |
| 寇英信: "基于改进核极限学习机和集成学习理论的目标机动轨迹预测", 国防科技大学学报 * |
| 甄绪: "面向非完备信息的机载多传感器航迹融合算法研究", 中国硕士学位论文全文数据库 信息科技专辑 * |
| 陈卫东;刘重阳;沈文亮;: "基于权值概率的多目标航迹关联算法研究", 无线电工程, no. 10 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN116359907B (en) | 2023-08-15 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111722199B (en) | Radar signal detection method based on convolutional neural network | |
| JP6699904B2 (en) | Radar device and radar signal processing method thereof | |
| CN111460597B (en) | Radar station distribution method based on improved multi-target particle swarm optimization algorithm | |
| CN111639747A (en) | GNSS-R sea surface wind speed inversion method and system based on BP neural network | |
| CN108919193A (en) | A kind of parameter agile radar signal sorting method excavated based on sequence fragment | |
| JP6773606B2 (en) | Radar device and its radar signal processing method | |
| CN110058222B (en) | Double-layer particle filter tracking-before-detection method based on sensor selection | |
| CN110007299B (en) | Weak target detection tracking method based on mixed coordinate pseudo-spectrum technology | |
| CN105137418A (en) | Multi-object tracking and data interconnection method based on whole neighborhood fuzzy clustering | |
| CN110378387B (en) | Cloud bottom height monitoring method based on wind-cloud-fourth satellite and ground-based millimeter wave radar | |
| CN111639746A (en) | GNSS-R sea surface wind speed inversion method and system based on CNN neural network | |
| CN108572353B (en) | Pulse time sequence planning method for low-interception radar | |
| Bocquel et al. | Multitarget particle filter addressing ambiguous radar data in TBD | |
| CN112162244A (en) | Event trigger target tracking method under correlated noise and random packet loss environment | |
| CN113341383A (en) | Radar anti-interference intelligent decision method based on DQN algorithm | |
| CN114117912A (en) | Sea clutter modeling and inhibiting method under data model dual drive | |
| CN115061126A (en) | Radar cluster target behavior identification method based on multi-dimensional parameter neural network | |
| Tian et al. | Performance evaluation of deception against synthetic aperture radar based on multifeature fusion | |
| CN116359907B (en) | Track processing method of radar data | |
| CN112785052B (en) | Wind speed and wind direction prediction method based on particle filter algorithm | |
| CN113608193A (en) | Radar multi-target distance and speed estimation method based on UNet | |
| CN106199552A (en) | A kind of packet generalized likelihood test method under local uniform sea clutter background | |
| CN110764084A (en) | Radar detection method for shielding target under complex terrain | |
| RU82045U1 (en) | DEVELOPMENT OF NEURAL NETWORK RECOGNITION OF TARGETS FOR THE TOTALITY OF SIGNS | |
| RU2471201C2 (en) | Method for radar scanning of space and radar set for realising said method (versions) |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant | ||
| CP02 | Change in the address of a patent holder |
Address after: Room 810, Building C, 8th Floor, Chuangye Building, No. 16 Gaoxin 1st Road, Xi'an City, Shaanxi Province, 710065 Patentee after: Xi'an Shengxin Technology Co.,Ltd. Address before: 710000 Productivity Building, No. 3, Dianzi West Street, Electronic Industrial Park, Yanta District, Xi'an, Shaanxi Patentee before: Xi'an Shengxin Technology Co.,Ltd. |
|
| CP02 | Change in the address of a patent holder |