CN105353644A - Radar target track derivative system on the basis of information mining of real-equipment data and method thereof - Google Patents
Radar target track derivative system on the basis of information mining of real-equipment data and method thereof Download PDFInfo
- Publication number
- CN105353644A CN105353644A CN201510627873.XA CN201510627873A CN105353644A CN 105353644 A CN105353644 A CN 105353644A CN 201510627873 A CN201510627873 A CN 201510627873A CN 105353644 A CN105353644 A CN 105353644A
- Authority
- CN
- China
- Prior art keywords
- data
- radar
- track
- agent
- flight path
- 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
Landscapes
- Radar Systems Or Details Thereof (AREA)
Abstract
The present invention belongs to the field of the radar simulation technology, and discloses an intelligent acquisition and intelligence automatic generation system on the basis of Agent radar real-equipment data and a method thereof. The system provided by the invention comprises a radar intelligence generation server and embedded acquisition devices. The first end of the radar intelligence generation server is connected with a plurality of embedded acquisition devices through a local area network, and each embedded acquisition device is connected with a real-equipment radar through the local area network or a RS485 netting twine and a RS232 netting twine; and the first end of the radar intelligence generation server is connected with a simulation application system of a radar data output device through the local area network or the RS485 netting twine and the RS232 netting twine, and the embedded acquisition devices are acquisition Agent hardware architectures. According to the invention, a plurality of real-equipment radar data may be acquired to realize radar real-equipment data mining and derivation of radar target track data, obtain optimal track truth values and noise models and synthesize the optimal track truth values and noise models to be a new radar simulation intelligence, therefore the application scope of the automatic generation technology of radar target tracks is further developed.
Description
Technical field
The invention belongs to technical field of radar simulation, relate to a kind of Radar Target Track flavor based on real-equipment data information excavating and method, namely based on Agent radar real-equipment data intelligent acquisition and information automatic creation system and method, be applicable to the exigent occasion of the fidelity of radar information simulation, can as the radar intelligence (RADINT) source of all-digital simulation system and semi-matter simulating system.
Background technology
Radar information simulation normally utilizes "black box" to carry out modeling and simulating, and method has two kinds: a kind of is that the nominal detection precision provided according to producer generates white noise, then is added to by white noise in emulation true value, and method is simple, but effect is poor; Another kind carries out analyzing according to historical data and statistics obtains systematic error and standard deviation, recycling Gaussian noise model carries out emulation synthesis, but the method can only reflect the statistical law that error is overall, can not real error characteristics be reflected, have impact on the fidelity of radar simulation.
Summary of the invention
In view of some simulation processes are to the demand of radar track high fidelity, the invention provides a kind of Radar Target Track flavor based on real-equipment data information excavating and method, can as the radar intelligence (RADINT) source of all-digital simulation system and semi-matter simulating system.
For achieving the above object, the present invention adopts following technical scheme:
A kind of Radar Target Track flavor based on real-equipment data information excavating, comprise: radar intelligence (RADINT) generation server, embedded harvester, described radar intelligence (RADINT) generation server first end is connected by the embedded harvester of LAN (Local Area Network) and several, and each embedded harvester is connected with actual load radar by LAN (Local Area Network), RS485 netting twine or RS232 netting twine; Radar intelligence (RADINT) generation server first end is connected with Simulation Application system by LAN (Local Area Network) or RS485 netting twine, RS232 netting twine, and described embedded harvester is for gathering Agent hardware architecture.
A kind of Radar Target Track flavor based on real-equipment data information excavating, described collection Agent hardware architecture, for realizing the harvester of actual load radar data, form a centerized fusion mode device by Agent management control module, data processing and memory module and port module;
Described Agent management control module is used for dynamic assignment and the scheduling of task of being responsible for, and coordinates the competition and cooperation between each Agent;
Described data processing and memory module are used for process and the storage of being responsible for data, carry out data preparation, data analysis, coordinate conversion and data and store;
Described port module is used for being responsible for digital received and sent, the reception of order and transmission and communication protocol is shaken hands.
A kind of Radar Target Track flavor based on real-equipment data information excavating, described radar intelligence (RADINT) generation server is realize the industrial computer that radar real-equipment data excavates and Radar Target Track data are derivative, comprise: the data mining processing module of first stage, the Radar Target Track automatic derivatization module of subordinate phase, described data mining processing module is off-line and online processing mode, model storage after having processed is in cluster data storehouse, loose coupling is formed with Radar Target Track automatic derivatization process, the model that as long as Query Database has been built when carrying out data and deriving,
Described industrial computer gathers Agent hardware architecture as management, for carrying out carrying out parallel optimization to gatherer process to collection Agent hardware architecture, achieves simultaneously to the collection that multiple actual load radar data carries out.
A kind of method of the Radar Target Track flavor based on real-equipment data information excavating, adopt the tactic pattern that radar intelligence (RADINT) generation server and embedded harvester networking are formed, by the unified management of data acquisition A gent, actual load radar data is gathered, and data are transported to radar intelligence (RADINT) generation server, namely based on actual load radar data acquisition with excavate on basis, by distributed system framework, Multi-Agent way to manage, with Ethernet, fieldbus for telecommunication media, carry out alternately with actual load radar data and Simulation Application system; The radar kind inputted according to demand, type of flight, radar accuracy parameter, realize the high confidence level radar target track simulation data automatically generating dynamic realtime, and can be transported to Simulation Application system; Its step is as follows:
1) the radar intelligence (RADINT) generation server adopted for realize actual load radar data and excavate and Radar Target Track data derivative, comprising: the data mining processing module of first stage, the Radar Target Track automatic derivatization module of subordinate phase;
The data mining processing module of first stage, in data processing, proposes and uses wavelet decomposition to be separated radar data with filtering, radar data is resolved into gross error, systematic error, stochastic error and flight path true value;
The data mining processing module of first stage, in model process of establishing, propose and use multinomial model to carry out segmentation modeling to flight path true value, employ the method to error enters gross error, systematic error, stochastic error carry out layering statistics, achieve being separated of flight path true value and error information, the accurate foundation of targetpath model, the flight path for data is sorted out and search coupling configuration unified interface;
The Radar Target Track automatic derivatization module of subordinate phase, the priority match principle that have employed characteristic of division carries out the pattern match of model, according to the priority search of radar signature class > signature of flight path class > environmental characteristic class > signal characteristic class, obtain flight path true value and noise model more accurately;
The Radar Target Track automatic derivatization module of subordinate phase, adopts flight path Secondary Match technology, when making existing model in database without coupling, can obtain track Simulation value, expanding the range of applicability that radar data is derivative by calculating.
The Radar Target Track automatic derivatization module of subordinate phase, during building database index, set up aspect indexing, condition index, spatial class index, time class index, shape class index, location class index simultaneously, and it is different on the impact of truth according to different index class, set up priority and sequencing, accelerate retrieval rate while improving accuracy rate, reduce magnetic disc access times, improve execution efficiency.
2) hardware architecture of the Multi-Agent adopted completes radar track information acquisition, is a kind of centerized fusion pattern, comprises: Agent management control module, data processing and memory module and port module.
Agent intelligent node is the embedded data collection terminal (node) based on WindowsCE system, and the unified management according to data acquisition A gent gathers actual load radar data, and data are transported to radar intelligence (RADINT) generation server;
The hardware architecture that have employed Multi-Agent in radar intelligence (RADINT) collection and hop completes radar track information acquisition, be a kind of centerized fusion pattern, be mainly divided into 3 parts: Agent management control module, data processing and memory module and port module;
Agent management control module is the major part of data collection architecture, and be responsible for dynamic assignment and the scheduling of task, coordinate the competition and cooperation between each Agent, its software is installed on radar intelligence (RADINT) generation server; The process of data processing and memory module primary responsibility data and storage, major function has data preparation, data analysis, coordinate conversion and data to store.Port module major function is digital received and sent, the reception of order and transmission and communication protocol are shaken hands, and its software is installed on Agent intelligent node.
A kind of method based on Agent radar real-equipment data intelligent acquisition and information automatic creation system, the data mining processing module of the first stage of described radar intelligence (RADINT) generation server work, be that the radar data received is carried out data mining process, concrete implementation step is as follows:
1) first utilize wavelet decomposition to be separated by radar data with the method for filtering, radar data is resolved into gross error, systematic error, stochastic error and flight path true value by frequency domain;
2) then the track data be separated is processed,
Flight path true value is carried out segmentation and modeling for utilizing Second-Order Discrete rate analytic approach by first content;
Second content is the statistics of gross error, the probability of happening of statistics gross error and peak ranges thereof;
3rd content is that the variation tendency of analytic system error carries out trend analysis;
4th content is the statistics of stochastic error, and stochastic error and noise signal, be generally normal distribution, obtains the high low signal distribution situation of the random noise of segmentation flight path, and calculates spectral characteristic and the distribution character of each section;
3) then the average flight speed per hour of the radar track data results of gained and target, track data sampling period and environmental factor are aggregated into database server, clustering method is utilized to set up data repository, realize Data classification, layer-management, be convenient to the data query of data derivatization process.
A kind of method based on Agent radar real-equipment data intelligent acquisition and information automatic creation system, the subordinate phase Radar Target Track automatic derivatization module of described radar intelligence (RADINT) generation server work, be the automatic derivatization utilizing cluster data storehouse to carry out Radar Target Track data according to demand, completed by radar intelligence (RADINT) generation server;
The first step is according to Simulation Application demand, obtain the priori of targetpath, and be quantified as and be input to data base querying and flight path and generate parameter: straight path, runway track, the 8 word tracks of the speed per hour of the detection accuracy of detection radar corresponding to radar parameter and flight path, scan period and data sampling period, scope of reconnaissance, radar detection probability, flight path, the weather of environmental factor, physical features, electromagnetic environment situation and story of a play or opera design, cluster data storehouse is inquired about track according to these parameters, and finds consistent or approximate data source;
Second step is combined by the flight path of data source, the track deception of synthesis Pass Test demand, realize the restructuring of noise and model according to the model of synthesis at the particular location of radar coverage and obtain required track data, last track data outputs in analogue system, completes the automatic generation of Radar Target Track.
A kind of method based on Agent radar real-equipment data intelligent acquisition and information automatic creation system, described radar intelligence (RADINT) generation server and the communication of Agent intelligent node are polling type, when radar intelligence (RADINT) generation server sends request of data to certain Agent intelligent node, the current radar data collected up is carried by Agent intelligent node; It is application pattern that Agent intelligent node receives actual load radar data, when Agent receiver module receives radar data, then produce application request, by Agent management control module according to application request processing data, send data by Agent sending module according to the non-polling case of higher level.
A kind of method based on Agent radar real-equipment data intelligent acquisition and information automatic creation system, that each Agent intelligent node is connected with corresponding actual load radar, and manage each Agent intelligent node by radar intelligence (RADINT) generation server, gather actual load radar data, by radar intelligence (RADINT) generation server process data, when coupled Simulation Application system sends the instruction needing service, then the radar intelligence (RADINT) of derivative synthesis is sent to Simulation Application system; Its concrete steps are as follows:
1), the work of radar intelligence (RADINT) generation server comprises, and the first stage is the step by " filtering separation-piecewise fitting-feature clustering ", sets up cluster data storehouse; Subordinate phase is the step of " feature association-flight path restructuring ", be according to Simulation Application system need generate radar track;
A large amount of radar intelligence (RADINT) data, for utilizing data mining technology, are carried out classification process, are divided out, set up empty feelings signature of flight path vector clusters data storehouse by the sample populations with similar features vector by the first stage of radar intelligence (RADINT) generation server work;
A. certain data processing method is first utilized (to be REA data by radar track data, i.e. inclined range, the angle of pitch and three groups, position angle data) be separated, utilizing the algorithm of wavelet decomposition and filtering, radar track data separating is by frequency domain noise signal and flight path true value two parts.
B. then the track data be separated is processed,
First content is the segmentation modeling of flight path true value, because flight path is divided at the uniform velocity, accelerates and variable accelerated motion in speed, track has and is divided into straight line, camber line and curve, therefore Second-Order Discrete rate analytic approach is utilized to carry out segmentation to flight path, and utilize fitting of a polynomial to carry out modeling to segmentation true value flight path, so that set up one comparatively close to the model of Live Flying flight path, improve the probability that when information generates, the match is successful;
Second content is the statistics of noise signal, analyzes the spectral characteristic of noise in each flight path section.
Finally the radar track data results of gained and Radar Objective Characteristics (comprising the information such as the detection accuracy of radar, investigative range and scan period) and environmental characteristics (comprising weather conditions etc.) are aggregated into database server, utilize clustering method that similar characteristic information is divided into respective colony, set up data repository, realize Data classification, layer-management, be convenient to the data query of data generating procedure.
Subordinate phase is the demand according to emulation, associating and model that law generation is new of the model and the data that obtain according to data mining.
The first step is according to emulation demand, obtain the priori of targetpath, and be quantified as and be input to parameter in data base querying and Track Software, as the speed per hour of radar parameter, flight path, environmental factor and the story of a play or opera (as straight path, runway track, 8 word tracks etc.) that designs, cluster data storehouse is inquired about track according to these parameters, and finds consistent or approximate data source;
Second step is combined by the flight path of data source, synthesis meets the track deception of emulation demand, realize the restructuring of noise and model according to the model of synthesis at the particular location of radar coverage and obtain required track data, last track data outputs in Simulation Application system, completes the automatic generation of Radar Target Track.
Based on a method for Agent radar real-equipment data intelligent acquisition and information automatic creation system, the flow process of described radar track data separating, concrete steps are as follows:
First utilize Algorithms of Wavelet Analysis to carry out noise reduction process to radar track data, flight path after noise reduction is carried out segmentation and carries out fitting of a polynomial, using the track after matching as true value, and subtract each other with former track data, obtain the error information of flight path track; Rule of judgment according to outlier extracts outlier, the peak ranges of statistics outlier and probability of happening; The error information of removing outlier is carried out second time filtering, and random noise is separated with systematic error the most at last, and analyzes respectively.
A kind of method based on Agent radar real-equipment data intelligent acquisition and information automatic creation system, the described workflow setting up radar track data clusters database, it is the feature clustering that similarity between object and the mutual relationship in expression of space are divided an object set, the proper vector setting up radar track data is the parameter vector in order to set up a cluster, clustering parameter vector is divided into four classes, i.e. radar signature class, signature of flight path class, environmental characteristic class and signal characteristic class, the order of the priority of four classes is radar signature class > signature of flight path class > environmental characteristic class > signal characteristic class, the foundation of cluster data storehouse index be according to preferential and order realize, thus realize hierarchical cluster, concrete steps are:
1) according to sample frequency or cycle, the detection accuracy of radar and the number of clusters of investigative range determination radar of radar data, set up the ground floor of multiclass cluster data group, a remarks column can be set, explain model and the type of radar;
2) then analyze according to the type of result to flight path of track segmentation and sort out, wherein average flight speed per hour needs setting cluster threshold value, the speed per hour of flight path is divided into multiple grade, sets up the cluster data group of signature of flight path class;
3) execution environment feature class record, what record content is the weather conditions of track data moment, can set up weather remark information;
4) cluster of signal characteristic class is the final step of data clusters, is an annotation to whole track data information, have recorded the characteristic quantity of signal, but not as the important information of flight path cluster association, only as the reference information of flight path signal level;
5) under the cluster data index set up, model combination parameter and flight path compressing original data are stored in database.
Owing to adopting technical scheme as above, the present invention has following superiority:
The present invention is a kind of Radar Target Track flavor based on real-equipment data information excavating and method, can as the radar intelligence (RADINT) source of all-digital simulation system and semi-matter simulating system.Invention have employed with the tactic pattern of industrial computer+built-in terminal, industrial computer is radar intelligence (RADINT) generation server, is Radar Information Processing center, realize radar real-equipment data excavate and Radar Target Track data derive; Built-in terminal is Agent intelligent acquisition node, gathers actual load radar data.Industrial computer carries out carrying out parallel optimization to gatherer process to Agent intelligent acquisition node as Management Agent simultaneously, can gather multiple actual load radar data.Radar intelligence (RADINT) generation server and Agent intelligent acquisition node are polling communication pattern, and Agent intelligent acquisition node and actual load radar are interrupt communication pattern.
In data mining part, invention design Wavelets Filtering Algorithm realizes the noise separation of radar real-equipment data, propose the algorithm by the data sectional modeling of flight path true value and noise data feature, then clustering parameter vector is utilized cluster feature to be divided into four classes such as radar signature class, signature of flight path class, environmental characteristic class and signal characteristic, and set priority, achieve the hierarchical cluster of radar real-equipment data.In data automatic derivatization part, invent and obtain optimal trajectory true value and noise model by the priority match method of cluster feature and synthesize new radar simulation information; Devise flight path Secondary Match technology, improve and be matched to power; During without Matching Model, utilize the radar mockup parameter in database, generate emulation flight path, expanded the applicable scope of Radar Target Track Auto further.
Accompanying drawing explanation
Fig. 1 is the structural representation based on Agent radar real-equipment data intelligent acquisition and information automatic creation system;
Fig. 2 is the schematic diagram of the radar intelligence (RADINT) sample and transform system architecture based on Multi-Agent;
Fig. 3 is the principle of work block scheme based on Agent radar real-equipment data intelligent acquisition and information automatic creation system;
Fig. 4 is Radar Target Track data mining processing flow chart;
Fig. 5 is Radar Target Track data derivation process process flow diagram;
Fig. 6 is that empty feelings prefer proper vector search pattern process flow diagram;
Fig. 7 is radar track data separating process flow diagram;
Fig. 8 is for setting up radar track data clusters database program process flow diagram.
Embodiment
As shown in Fig. 1 to 8, a kind of Radar Target Track flavor based on real-equipment data information excavating, comprise: radar intelligence (RADINT) generation server, embedded harvester, described radar intelligence (RADINT) generation server first end is connected by the embedded harvester of LAN (Local Area Network) and several, and each embedded harvester is connected with actual load radar by LAN (Local Area Network) or RS485 netting twine, RS232 netting twine; Radar intelligence (RADINT) generation server first end is connected with the Simulation Application system of radar data output unit by LAN (Local Area Network) or RS485 netting twine, RS232 netting twine, and described embedded harvester is for gathering Agent hardware architecture.
Native system have employed the tactic pattern of industrial computer+embedded collection, and structure as shown in Figure 1, is formed primarily of Agent intelligent node and the networking of radar intelligence (RADINT) generation server.The major function of radar intelligence (RADINT) generation server has the derivative etc. of the excavation of data capture management Agent, radar intelligence (RADINT) and cluster and radar intelligence (RADINT).In the data acquisition based on actual load radar with on excavation basis, by distributed system framework, Multi-Agent way to manage, with Ethernet, fieldbus etc. for telecommunication media and radar fashionable dress data and the demand of emulation equipment carry out alternately, the parameter such as radar kind, type of flight, data precision inputted according to demand, realize the high confidence level radar target track simulation data automatically generating dynamic realtime, and Simulation Application system can be transported to.
Agent intelligent node is the embedded data collection terminal (node) based on WindowsCE system, unified management according to data acquisition A gent gathers actual load radar data, and data are transported to radar intelligence (RADINT) generation server, its primary structure is as shown in Figure 2.The hardware architecture that have employed Multi-Agent in radar intelligence (RADINT) collection and hop completes radar track information acquisition, be a kind of centerized fusion pattern, be mainly divided into 3 parts: Agent management control module, data processing and memory module and port module.Agent management control module is the major part of data collection architecture, and be responsible for dynamic assignment and the scheduling of task, coordinate the competition and cooperation between each Agent, its software is installed on radar intelligence (RADINT) generation server.The process of data processing and memory module primary responsibility data and storage, major function has data preparation, data analysis, coordinate conversion and data to store.Port module major function is digital received and sent, the reception of order and transmission and communication protocol are shaken hands, and its software is installed on Agent intelligent node.
The course of work of radar intelligence (RADINT) generation server is divided into two stages, and as shown in Figure 3, first stage is that the radar data received is carried out data mining process, and its main process as shown in Figure 4.First utilize wavelet decomposition to be separated by radar data with the method for filtering, radar data is resolved into gross error, systematic error, stochastic error and flight path true value by frequency domain.Then process the track data be separated, this step has four contents, and flight path true value is carried out segmentation and modeling for utilizing Second-Order Discrete rate analytic approach by first content; Second content is the statistics of gross error, the statistics probability of happening of gross error and the scope of peak value thereof; 3rd content is the variation tendency of analytic system error; 4th content is the statistics of stochastic error, and stochastic error and noise signal, be generally normal distribution, obtains the size distribution situation of the random noise of segmentation flight path, and calculates spectral characteristic and the distribution character of each section.Then the average flight speed per hour of the radar track data results of gained and target, track data sampling period and environmental factor are aggregated into database server, clustering method is utilized to set up data repository, realize Data classification, layer-management, be convenient to the data query of data derivatization process.Subordinate phase is the automatic derivatization utilizing cluster data storehouse to carry out Radar Target Track data according to demand, is completed by radar intelligence (RADINT) generation server, and its main process as shown in Figure 5.The first step is according to Simulation Application demand, obtain the priori of targetpath, and be quantified as and be input to parameter in data base querying and Track Software, as radar parameter (the i.e. detection accuracy of detection radar corresponding to flight path, scan period (data sampling period), scope of reconnaissance, radar detection probability etc.), the speed per hour of flight path, environmental factor is (as weather, physical features, electromagnetic environment situation etc.) and the story of a play or opera that designs (as straight path, runway track, 8 word tracks etc.), cluster data storehouse is inquired about track according to these parameters, and find consistent or approximate data source, second step is combined by the flight path of data source, the track deception of synthesis Pass Test demand, realize the restructuring of noise and model according to the model of synthesis at the particular location of radar coverage and obtain required track data, last track data outputs in analogue system, completes the automatic generation of Radar Target Track.
The communication of radar intelligence (RADINT) generation server and Agent intelligent node is polling type, and when radar intelligence (RADINT) generation server sends request of data to certain Agent intelligent node, the current radar data collected up is carried by Agent intelligent node.It is application pattern that Agent intelligent node receives actual load radar data, when Agent receiver module receives radar data, then produce application request, by Agent management control module according to application request processing data, send data by Agent sending module according to the non-polling case of higher level.
Native system realizes comprising software and hardware, wherein hardware configuration is industrial computer (2.0GHz or more processor, 1G or more internal memory, 512M or more video memory, 1T or more hard disk, the color monitor of 800*600 or more resolution, DVD-ROM), Agent data collection station device polylith, Ethernet card one piece, RS485 Bus PC I plug-in card one piece, RS232 serial ports PCI plug-in card one piece and router one, can match hardware is printer one.System card software, radar intelligence (RADINT) automatic generating software a set of (comprising the data mining mode software based on Multi-Agent acquisition node and radar intelligence (RADINT) data spin-off model software) when software section comprises 485 bus driver, high precision.
1, each part description
Described industrial computer is radar intelligence (RADINT) generation server, has installed 2 softwares, namely based on data mining mode software and the radar intelligence (RADINT) data spin-off model software of Multi-Agent acquisition node.Data mining mode software based on Multi-Agent acquisition node can carry out management and use data digging method to each Agent intelligent acquisition node and generate cluster data storehouse, and function comprises peer distribution, node starts, gather beginning, data record, data processing, data clusters.
The physical link that described Ethernet card connects for the LAN (Local Area Network) setting up whole system.
Described router is for building link and the IP address assignment of LAN (Local Area Network).
Described RS485 Bus PC I plug-in card, RS485 address card, for realizing native system and outside RS485 bus communication, can carry out information interaction.
Described RS232 serial ports PCI plug-in card, RS232 serial ports, for realizing native system and outside RS232 serial communication, can carry out information interaction.
Described Agent data collection station is Agent intelligent node, for a kind of based on the embedded data acquisition device of WindowsCE operating system, be integrated with and receive data Agent, data processing and storage Agent, sends the modules such as data Agent.Containing RS485, RS232 and Ethernet interface on acquisition terminal, and function button and touch 7 cun of chromatic liquid crystal screens.
2, guardian technique and principle explanation
Each Agent intelligent node is connected with corresponding actual load radar, and manage each Agent intelligent node by radar intelligence (RADINT) generation server, gather radar real-equipment data, by radar intelligence (RADINT) generation server process data, when coupled Simulation Application system sends the instruction needing service, then the radar intelligence (RADINT) of derivative synthesis is sent to Simulation Application system.The work of radar intelligence (RADINT) generation server is divided into two stages, and as shown in Figure 6, the first stage sets up cluster data storehouse by " filtering separation-piecewise fitting-feature clustering " several step; Another stage is " feature association-flight path restructuring " two steps, be according to Simulation Application system need generate radar track.
A large amount of radar intelligence (RADINT) data, for utilizing data mining technology, are carried out classification process, are divided out, set up empty feelings signature of flight path vector clusters data storehouse by the sample populations with similar features vector by the first stage of radar intelligence (RADINT) generation server work.Key step is: first utilize certain data processing method (to be REA data by radar track data, i.e. inclined range, the angle of pitch and three groups, position angle data) be separated, utilizing the algorithm of wavelet decomposition and filtering, radar track data separating is by frequency domain noise signal and flight path true value two parts.Then the track data be separated is processed, this step has two contents, first content is the segmentation modeling of flight path true value, because flight path can be divided at the uniform velocity, accelerate and variable accelerated motion in speed, track has and can be divided into straight line, camber line and curve, therefore Second-Order Discrete rate analytic approach is utilized to carry out segmentation to flight path, and utilize fitting of a polynomial to carry out modeling to segmentation true value flight path, so that set up one comparatively close to the model of Live Flying flight path, improve the probability that when information generates, the match is successful; Second content is the statistics of noise signal, analyzes the spectral characteristic of noise in each flight path section.Finally the radar track data results of gained and Radar Objective Characteristics (comprising the information such as the detection accuracy of radar, investigative range and scan period) and environmental characteristics (comprising weather conditions etc.) are aggregated into database server, utilize clustering method that similar characteristic information is divided into respective colony, set up data repository, realize Data classification, layer-management, be convenient to the data query of data generating procedure.Subordinate phase is the demand according to emulation, associating and model that law generation is new of the model and the data that obtain according to data mining.There are two steps: the first step is according to emulation demand, obtain the priori of targetpath, and be quantified as and be input to parameter in data base querying and Track Software, as the speed per hour of radar parameter, flight path, environmental factor and the story of a play or opera (as straight path, runway track, 8 word tracks etc.) that designs, cluster data storehouse is inquired about track according to these parameters, and finds consistent or approximate data source; Second step is combined by the flight path of data source, synthesis meets the track deception of emulation demand, realize the restructuring of noise and model according to the model of synthesis at the particular location of radar coverage and obtain required track data, last track data outputs in Simulation Application system, completes the automatic generation of Radar Target Track.
The flow process of radar track data separating as shown in Figure 7, whole process is mainly divided into 3 steps: first utilize Algorithms of Wavelet Analysis to carry out noise reduction process to radar track data, flight path after noise reduction is carried out segmentation and carries out fitting of a polynomial, using the track after matching as true value, and subtract each other with former track data, obtain the error information of flight path track; Rule of judgment according to outlier extracts outlier, the peak ranges of statistics outlier and probability of happening; The error information of removing outlier is carried out second time filtering, and random noise is separated with systematic error the most at last, and analyzes respectively.
Track data segmentation carries out segmentation by filtered track data, and its object has two: one to be to set up more accurate targetpath model; Another is that the flight path conveniently realizing data is sorted out and search coupling.Polynomial modeling utilizes polynomial expression to carry out matching to flight path, and it is higher to obtain precision, and with true value flight path mathematical model closely, to obtain the back propagation net of more closing to reality.The segmentation of Radar Target Track utilizes single order dispersion ratio and Second-Order Discrete rate to analyze the curvature of track to determine the unique point of flight path curve, to locate track segmentation point.The matching of Radar Target Track utilizes the Polynomial modeling method on 5 rank to carry out matching.
Feature clustering is divided an object set similarity between object and the mutual relationship in expression of space, and the proper vector setting up radar track data is the parameter vector in order to set up a cluster, to construct a cluster.Clustering parameter is the data structure or the array that describe radar track information.The clustering parameter vector comprised as shown in the figure.Although track deception combination does not belong to clustering parameter vector with raw data, cluster data storehouse must be put into, for the coupling of flight path and generation provide foundation and condition simultaneously.In addition, course system error belongs to optional element, and the real needs for flight path design judge whether to need as a Consideration.Clustering parameter vector is divided into four classes, i.e. radar signature class, signature of flight path class, environmental characteristic class and signal characteristic class, the order of the priority of four classes is radar signature class > signature of flight path class > environmental characteristic class > signal characteristic class.The foundation of cluster data storehouse index be according to preferential and order realize, thus realize hierarchical cluster, concrete steps are: 1) according to sample frequency or cycle, the detection accuracy of radar and the number of clusters of investigative range determination radar of radar data, set up the ground floor of multiclass cluster data group, a remarks column can be set, explain model and the type of radar; 2) then analyze according to the type of result to flight path of track segmentation and sort out, wherein average flight speed per hour needs setting cluster threshold value, the speed per hour of flight path is divided into multiple grade, sets up the cluster data group of signature of flight path class; 3) execution environment feature class record, what record content is the weather conditions of track data moment, can set up weather remark information; 4) cluster of signal characteristic class is the final step of data clusters, is an annotation to whole track data information, have recorded the characteristic quantity of signal, but not as the important information of flight path cluster association, only as the reference information of flight path signal level; 5) under the cluster data index set up, model combination parameter and flight path compressing original data are stored in database.Set up radar track data clusters database program flow process as shown in Figure 8.
Data derive be from set up cluster data search for and find out the signature of flight path that satisfied empty feelings prefer design requirement, and recall corresponding data model or historical data, and reassembling into new Radar Target Track, its key step is divided into the flux matched and new track data of data characteristics to combine two steps.Feature class coupling picks out with the history radar track data that space length is minimum, similarity is maximum or numerical expression is consistent of proper vector that the empty feelings of design prefer, mainly contain situation three kinds: 1) definite value characteristic quantity coupling, namely characteristic quantity is definite value, is priority match item; 2) threshold value coupling, what namely meet threshold condition reaches coupling; 3) track matching, namely utilizes theorem in Euclid space distance to carry out the match search of flight path.Characteristic quantity match search is according to the priority search of radar signature class > signature of flight path class > environmental characteristic class > signal characteristic class, as shown in the figure, radar signature class and signature of flight path class are the emphasis in characteristic quantity, if these two it fails to match, then data spin-off model failure, selects simulation model to set up new flight path.Feature class coupling has following several modes: 1) radar signature class is without match objects: the database not setting up this class radar in this situation database of descriptions, this situation can only generate Gaussian noise according to given radar detection precision, then the empty feelings set that are added to prefer, what this method generated is a kind of air scenario ideally, is distributed with different with the radar noise of reality; 2) signature of flight path class Secondary Match: the coupling of flight path is the coupling between the polynomial expression track deception of matching in the sectional curve and database preferred.Coupling may there will be two kinds of situations for flight path class, be none match objects of signature of flight path after segmentation, another kind of situation only has part flight path section without match objects, therefore for problems, need flight path section to be divided into straight line and camber line to process respectively, carry out Secondary Match.3) flight path exists without match objects.Namely database has radar signature class to mate but exists without signature of flight path class coupling, therefore the radar signature category information of existing statistics must be utilized to carry out emulating the new flight path of synthesis, main proper vector is radar noise distribution envelope, get a certain amount of flight path noise data, according to the detection range of radar, set up radar noise distribution envelope diagram, reflection be noise profile situation in this radar coverage, be the process that an emulation flight path generates.
Building database index is one of important step of similar inquiry.The flight path of each time series should have individual identifier, can set up match query network as soon as possible.Index mainly contains following a few class: 1) tagsort index: mainly contain 01-radar signature class, 02-signature of flight path class, 03 environmental characteristic class, 04-signal characteristic class.2) condition class index: what show is certain situation or condition, as state of flight, ambient conditions etc.3) spatial class index: reflection be interval or the scope of certain section of spatial data; 4) time class index: reflection be the time domain of certain segment data; 5) shape class index: reflection be the shape of track data, as 01-straight line, 02-camber line; 6) class index is located: as slope and the home position of straight line.Index is a hierarchical structure, is according to the preferential of feature class and sets up, and can not only accelerate retrieval rate, and can reduce magnetic disc access times, improve execution efficiency.
Based on Multi-Agent data acquisition intelligent node structure as shown in the figure, be made up of reception, sending module and process and memory module, all Agent intelligent nodes have the Management Agent in radar intelligence (RADINT) generation server to carry out unified management, the working method of Management Agent has: the 1) startup of Management Agent: analyze network topology, set up good network state, the duty of each acquisition node of poll, judges whether each radar works, and waits collection and the transmission of pending data; 2) work of Management Agent: the state monitoring each Agent, manages the information gathering Agent, collects the data that Agent gathers, proceeds to mining data process software and carry out data processing.Data Collection be according to the size of the IP sequence number of Agent as priority to realize the transmission successively of data.3) commander of Management Agent: Management Agent has command & control function, can realize commanding each Agent node and controlling according to the rule base set up, make the co-ordination of multi-Agent energy, the task of data acquisition.And gather according to the information commander that each Agent returns or stop data collection.Other Agent is in intelligent frame grabber, be one be the embedded equipment of core by dsp processor, wherein data processing and storage Agent are subordinates of this acquisition system, main working method is as follows: the 1) process of data: mainly radar data is carried out coordinate conversion according to the order of Management Agent, as converted geographic coordinate or geocentric coordinate to.Data scrubbing is carried out to radar intelligence (RADINT), the track data that track data and outlier probability of happening as deleted negligible amounts are larger; 2) data store: back up data in storer, and data are stamped markers and annotation; 3) information interaction: carry out information interaction with Management Agent, performs the instruction of Management Agent and returns work state information.Receive data Agent groundwork mode as follows: 1) receiving radar information data: receive the empty feelings information from radar, and be stored in host memory or buffer zone with the form of queuing up, wait for data output instruction; 2) surveillance radar information input: whether whether surveillance radar in work, and have data to input; 3) information interaction: the instruction of receiving management Agent and other Agent, return state information.Transmitting and receiving data Agent groundwork mode is as follows: 1) carry text writing: set up the XML text of the empty feelings information of radar and be compressed into zip format file: 2) communication protocol is shaken hands: shake hands with Management Agent communication protocol, ensures the network utilisation efficiency of data transmission; 3) information interaction: the instruction of receiving management Agent, return state information.
3, system software
Software mainly contains two blocks, and one is the data mining of radar intelligence (RADINT) generation server and derivative software, and another is the Agent intelligent acquisition node embedded software based on WindowsCE.The data mining of radar intelligence (RADINT) generation server and derivative software comprise two sub-softwares, and namely data mining and data acquisition routines and radar intelligence (RADINT) data derive subroutine.
Data mining and data acquisition routines have two kinds of mode of operations, i.e. Multi-Agent drainage pattern and text entry mode.Multi-Agent drainage pattern is the acquisition node information online acquisition radar intelligence (RADINT) data according to arranging; Text entry mode be data from record text by data importing in database, be a kind of data acquisition modes of off-line.
It is according to the prior radar parameter, the flight path track that design that radar intelligence (RADINT) derives software, generates the radar detection flight path meeting and impose a condition.
Agent intelligent acquisition node software is work and the state display of three Agent, can configure No. IP and the communication port of each Agent, building topology network structure.
Claims (10)
1. one kind based on Agent radar real-equipment data intelligent acquisition and information automatic creation system, it is characterized in that: comprising: radar intelligence (RADINT) generation server, embedded harvester, described radar intelligence (RADINT) generation server first end is connected by the embedded harvester of LAN (Local Area Network) and several, and each embedded harvester is connected with actual load radar by LAN (Local Area Network) or RS485 netting twine, RS232 netting twine; Radar intelligence (RADINT) generation server first end is connected with the Simulation Application system of radar data output unit by LAN (Local Area Network) or RS485 netting twine, RS232 netting twine, and described embedded harvester is for gathering Agent hardware architecture.
2. one according to claim 1 is based on Agent radar real-equipment data intelligent acquisition and information automatic creation system, it is characterized in that: described collection Agent hardware architecture, for realizing the harvester of actual load radar data, form a centerized fusion mode device by Agent management control module, data processing and memory module and port module;
Described Agent management control module is used for dynamic assignment and the scheduling of task of being responsible for, and coordinates the competition and cooperation between each Agent;
Described data processing and memory module are used for process and the storage of being responsible for data, carry out data preparation, data analysis, coordinate conversion and data and store;
Described port module is used for being responsible for digital received and sent, the reception of order and transmission and communication protocol is shaken hands.
3. one according to claim 1 is based on Agent radar real-equipment data intelligent acquisition and information automatic creation system, it is characterized in that: described radar intelligence (RADINT) generation server is realize the industrial computer that radar real-equipment data excavates and Radar Target Track data are derivative, comprise: the data mining processing module of first stage, the Radar Target Track automatic derivatization module of subordinate phase
Described data mining processing module is off-line and online processing mode, model storage after having processed is in cluster data storehouse, loose coupling is formed, as long as the model that Query Database has been built when carrying out data and deriving with Radar Target Track automatic derivatization process;
Described industrial computer gathers Agent hardware architecture as management, for carrying out carrying out parallel optimization to gatherer process to collection Agent hardware architecture, achieves simultaneously to the collection that multiple actual load radar data carries out.
4. a kind of method based on Agent radar real-equipment data intelligent acquisition and information automatic creation system as claimed in claim 1, it is characterized in that: adopt the tactic pattern that radar intelligence (RADINT) generation server and embedded harvester networking are formed, by the unified management of data acquisition A gent, actual load radar data is gathered, and data are transported to radar intelligence (RADINT) generation server, namely based on actual load radar data acquisition with excavate on basis, by distributed system framework, Multi-Agent way to manage, with Ethernet, fieldbus is telecommunication media, with fashionable dress radar data with emulate demand equipment and carry out alternately, the radar kind inputted according to demand, type of flight, data precision parameter, realize the high confidence level radar target track simulation data automatically generating dynamic realtime, and can be transported to Simulation Application system, its step is as follows:
1) the radar intelligence (RADINT) generation server adopted for realize actual load radar data and excavate and Radar Target Track data derivative, comprising: the data mining processing module of first stage, the Radar Target Track automatic derivatization module of subordinate phase;
The data mining processing module of first stage, in data processing, proposes and uses wavelet decomposition to be separated radar data with filtering, radar data is resolved into gross error, systematic error, stochastic error and flight path true value;
The data mining processing module of first stage, in model process of establishing, propose and use multinomial model to carry out segmentation modeling to flight path true value, employ the method to error enters gross error, systematic error, stochastic error carry out layering statistics, achieve being separated of flight path true value and error information, the accurate foundation of targetpath model, the flight path for data is sorted out and search coupling configuration unified interface;
The Radar Target Track automatic derivatization module of subordinate phase, the priority match principle that have employed characteristic of division carries out the pattern match of model, according to the priority search of radar signature class > signature of flight path class > environmental characteristic class > signal characteristic class, obtain flight path true value and noise model more accurately;
The Radar Target Track automatic derivatization module of subordinate phase, adopts flight path Secondary Match technology, when making existing model in database without coupling, can obtain track Simulation value, expanding the range of applicability that radar data is derivative by calculating;
The Radar Target Track automatic derivatization module of subordinate phase, during building database index, set up aspect indexing, condition index, spatial class index, time class index, shape class index, location class index simultaneously, and it is different on the impact of truth according to different index class, set up priority and sequencing, accelerate retrieval rate while improving accuracy rate, reduce magnetic disc access times, improve execution efficiency;
2) hardware architecture of the Multi-Agent adopted completes radar track information acquisition, is a kind of centerized fusion pattern, comprises: Agent management control module, data processing and memory module and port module;
Agent intelligent node is the embedded data collection terminal (node) based on WindowsCE system, unified management according to data acquisition A gent gathers actual load radar data, and data are transported to radar intelligence (RADINT) generation server, its primary structure is as shown in Figure 2; The hardware architecture that have employed Multi-Agent in radar intelligence (RADINT) collection and hop completes radar track information acquisition, be a kind of centerized fusion pattern, be mainly divided into 3 parts: Agent management control module, data processing and memory module and port module;
Agent management control module is the major part of data collection architecture, and be responsible for dynamic assignment and the scheduling of task, coordinate the competition and cooperation between each Agent, its software is installed on radar intelligence (RADINT) generation server; The process of data processing and memory module primary responsibility data and storage, major function has data preparation, data analysis, coordinate conversion and data to store; Port module major function is digital received and sent, the reception of order and transmission and communication protocol are shaken hands, and its software is installed on Agent intelligent node.
5. a kind of method based on Agent radar real-equipment data intelligent acquisition and information automatic creation system according to claim 1, it is characterized in that: the data mining processing module of the first stage of described radar intelligence (RADINT) generation server work, be that the radar data received is carried out data mining process, concrete implementation step is as follows:
1) first utilize wavelet decomposition to be separated by radar data with the method for filtering, radar data is resolved into gross error, systematic error, stochastic error and flight path true value by frequency domain;
2) then the track data be separated is processed,
Flight path true value is carried out segmentation and modeling for utilizing Second-Order Discrete rate analytic approach by first content;
Second content is the statistics of gross error, the probability of happening of statistics gross error and peak ranges thereof;
3rd content is that the variation tendency of analytic system error carries out trend analysis;
4th content is the statistics of stochastic error, and stochastic error and noise signal, be generally normal distribution, obtains the high low signal distribution situation of the random noise of segmentation flight path, and calculates spectral characteristic and the distribution character of each section;
3) then the average flight speed per hour of the radar track data results of gained and target, track data sampling period and environmental factor are aggregated into database server, clustering method is utilized to set up data repository, realize Data classification, layer-management, be convenient to the data query of data derivatization process.
6. a kind of method based on Agent radar real-equipment data intelligent acquisition and information automatic creation system according to claim 1, it is characterized in that: the subordinate phase Radar Target Track automatic derivatization module of described radar intelligence (RADINT) generation server work, be the automatic derivatization utilizing cluster data storehouse to carry out Radar Target Track data according to demand, completed by radar intelligence (RADINT) generation server;
The first step is according to Simulation Application demand, obtain the priori of targetpath, and be quantified as and be input to data base querying and flight path and generate parameter: straight path, runway track, the 8 word tracks of the speed per hour of the detection accuracy of detection radar corresponding to radar parameter and flight path, scan period and data sampling period, scope of reconnaissance, radar detection probability, flight path, the weather of environmental factor, physical features, electromagnetic environment situation and story of a play or opera design, cluster data storehouse is inquired about track according to these parameters, and finds consistent or approximate data source;
Second step is combined by the flight path of data source, the track deception of synthesis Pass Test demand, realize the restructuring of noise and model according to the model of synthesis at the particular location of radar coverage and obtain required track data, last track data outputs in analogue system, completes the automatic generation of Radar Target Track.
7. a kind of method based on Agent radar real-equipment data intelligent acquisition and information automatic creation system according to claim 1, it is characterized in that: described radar intelligence (RADINT) generation server and the communication of Agent intelligent node are polling type, when radar intelligence (RADINT) generation server sends request of data to certain Agent intelligent node, the current radar data collected up is carried by Agent intelligent node; It is application pattern that Agent intelligent node receives actual load radar data, when Agent receiver module receives radar data, then produce application request, by Agent management control module according to application request processing data, send data by Agent sending module according to the non-polling case of higher level.
8. a kind of method based on Agent radar real-equipment data intelligent acquisition and information automatic creation system according to claim 1, it is characterized in that: be that each Agent intelligent node is connected with corresponding actual load radar, and manage each Agent intelligent node by radar intelligence (RADINT) generation server, gather actual load radar data, by radar intelligence (RADINT) generation server process data, when coupled Simulation Application system sends the instruction needing service, then the radar intelligence (RADINT) of derivative synthesis is sent to Simulation Application system; Its concrete steps are as follows:
1), the work of radar intelligence (RADINT) generation server comprises, and the first stage is the step by " filtering separation-piecewise fitting-feature clustering ", sets up cluster data storehouse; Subordinate phase is the step of " feature association-flight path restructuring ", be according to Simulation Application system need generate radar track;
A large amount of radar intelligence (RADINT) data, for utilizing data mining technology, are carried out classification process, are divided out, set up empty feelings signature of flight path vector clusters data storehouse by the sample populations with similar features vector by the first stage of radar intelligence (RADINT) generation server work;
A. certain data processing method is first utilized (to be REA data by radar track data, i.e. inclined range, the angle of pitch and three groups, position angle data) be separated, utilizing the algorithm of wavelet decomposition and filtering, radar track data separating is by frequency domain noise signal and flight path true value two parts;
B. then the track data be separated is processed,
First content is the segmentation modeling of flight path true value, because flight path is divided at the uniform velocity, accelerates and variable accelerated motion in speed, track has and is divided into straight line, camber line and curve, therefore Second-Order Discrete rate analytic approach is utilized to carry out segmentation to flight path, and utilize fitting of a polynomial to carry out modeling to segmentation true value flight path, so that set up one comparatively close to the model of Live Flying flight path, improve the probability that when information generates, the match is successful;
Second content is the statistics of noise signal, analyzes the spectral characteristic of noise in each flight path section;
Finally the radar track data results of gained and Radar Objective Characteristics (comprising the information such as the detection accuracy of radar, investigative range and scan period) and environmental characteristics (comprising weather conditions etc.) are aggregated into database server, utilize clustering method that similar characteristic information is divided into respective colony, set up data repository, realize Data classification, layer-management, be convenient to the data query of data generating procedure;
Subordinate phase is the demand according to emulation, associating and model that law generation is new of the model and the data that obtain according to data mining;
The first step is according to emulation demand, obtain the priori of targetpath, and be quantified as and be input to parameter in data base querying and Track Software, as the speed per hour of radar parameter, flight path, environmental factor and the story of a play or opera (as straight path, runway track, 8 word tracks etc.) that designs, cluster data storehouse is inquired about track according to these parameters, and finds consistent or approximate data source;
Second step is combined by the flight path of data source, synthesis meets the track deception of emulation demand, realize the restructuring of noise and model according to the model of synthesis at the particular location of radar coverage and obtain required track data, last track data outputs in Simulation Application system, completes the automatic generation of Radar Target Track.
9. a kind of method based on Agent radar real-equipment data intelligent acquisition and information automatic creation system according to claim 1, it is characterized in that: the flow process of described radar track data separating, concrete steps are as follows:
First utilize Algorithms of Wavelet Analysis to carry out noise reduction process to radar track data, flight path after noise reduction is carried out segmentation and carries out fitting of a polynomial, using the track after matching as true value, and subtract each other with former track data, obtain the error information of flight path track; Rule of judgment according to outlier extracts outlier, the peak ranges of statistics outlier and probability of happening; The error information of removing outlier is carried out second time filtering, and random noise is separated with systematic error the most at last, and analyzes respectively.
10. a kind of method based on Agent radar real-equipment data intelligent acquisition and information automatic creation system according to claim 1, it is characterized in that: the described workflow setting up radar track data clusters database, it is the feature clustering that similarity between object and the mutual relationship in expression of space are divided an object set, the proper vector setting up radar track data is the parameter vector in order to set up a cluster, clustering parameter vector is divided into four classes, i.e. radar signature class, signature of flight path class, environmental characteristic class and signal characteristic class, the order of the priority of four classes is radar signature class > signature of flight path class > environmental characteristic class > signal characteristic class, the foundation of cluster data storehouse index be according to preferential and order realize, thus realize hierarchical cluster, concrete steps are:
1) according to sample frequency or cycle, the detection accuracy of radar and the number of clusters of investigative range determination radar of radar data, set up the ground floor of multiclass cluster data group, a remarks column can be set, explain model and the type of radar;
2) then analyze according to the type of result to flight path of track segmentation and sort out, wherein average flight speed per hour needs setting cluster threshold value, the speed per hour of flight path is divided into multiple grade, sets up the cluster data group of signature of flight path class;
3) execution environment feature class record, what record content is the weather conditions of track data moment, can set up weather remark information;
4) cluster of signal characteristic class is the final step of data clusters, is an annotation to whole track data information, have recorded the characteristic quantity of signal, but not as the important information of flight path cluster association, only as the reference information of flight path signal level;
5) under the cluster data index set up, model combination parameter and flight path compressing original data are stored in database.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510627873.XA CN105353644B (en) | 2015-09-29 | 2015-09-29 | Radar Target Track flavor and method based on real-equipment data information excavating |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510627873.XA CN105353644B (en) | 2015-09-29 | 2015-09-29 | Radar Target Track flavor and method based on real-equipment data information excavating |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN105353644A true CN105353644A (en) | 2016-02-24 |
| CN105353644B CN105353644B (en) | 2018-06-15 |
Family
ID=55329633
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201510627873.XA Active CN105353644B (en) | 2015-09-29 | 2015-09-29 | Radar Target Track flavor and method based on real-equipment data information excavating |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN105353644B (en) |
Cited By (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107346107A (en) * | 2016-05-04 | 2017-11-14 | 深圳光启合众科技有限公司 | Diversified motion control method and system and the robot with the system |
| CN108225276A (en) * | 2018-01-11 | 2018-06-29 | 中国人民解放军战略支援部队航天工程大学 | A kind of list star imageable target kinetic characteristic inversion method and system |
| CN109493644A (en) * | 2018-11-16 | 2019-03-19 | 四川大学 | A kind of four-dimensional flying track conjecture method excavated based on history track data |
| CN109739921A (en) * | 2019-01-07 | 2019-05-10 | 北京云基数技术有限公司 | A kind of big data acquisition system and method |
| CN109783515A (en) * | 2019-01-25 | 2019-05-21 | 上海创景信息科技有限公司 | More relation data tracks retroactive method and system based on database |
| CN109992569A (en) * | 2019-02-19 | 2019-07-09 | 平安科技（深圳）有限公司 | Cluster log feature extracting method, device and storage medium |
| CN111145599A (en) * | 2019-12-17 | 2020-05-12 | 中国民用航空总局第二研究所 | Curve flight segment and error distribution establishing method |
| CN113484841A (en) * | 2021-07-26 | 2021-10-08 | 中国船舶重工集团公司第七二三研究所 | Track head processing method under search radar multitask parallel |
| CN114063653A (en) * | 2021-12-03 | 2022-02-18 | 中国人民解放军63892部队 | Space and time consistency constraint small unmanned aerial vehicle high-precision track control method |
| CN116415161A (en) * | 2023-04-17 | 2023-07-11 | 中交第四航务工程局有限公司 | Fitting complementation method for geological drilling detection and different physical wave detection data of string-shaped karst cave |
| CN117098255A (en) * | 2023-10-19 | 2023-11-21 | 南京波达电子科技有限公司 | Edge calculation-based decentralization radar ad hoc network method |
| CN118034605A (en) * | 2024-03-15 | 2024-05-14 | 上海正泰自动化软件系统有限公司 | Data processing method and device, electronic equipment and storage medium |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101483545A (en) * | 2008-12-31 | 2009-07-15 | 中国建设银行股份有限公司 | Financial service monitoring method and system |
| CN203630359U (en) * | 2013-11-28 | 2014-06-04 | 西安工程大学 | Radar-based on-line monitoring system for protecting power transmission line from breakage by external factors |
| US8935319B2 (en) * | 2012-04-16 | 2015-01-13 | International Business Machines Corporation | Data capture system |
-
2015
- 2015-09-29 CN CN201510627873.XA patent/CN105353644B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101483545A (en) * | 2008-12-31 | 2009-07-15 | 中国建设银行股份有限公司 | Financial service monitoring method and system |
| US8935319B2 (en) * | 2012-04-16 | 2015-01-13 | International Business Machines Corporation | Data capture system |
| CN203630359U (en) * | 2013-11-28 | 2014-06-04 | 西安工程大学 | Radar-based on-line monitoring system for protecting power transmission line from breakage by external factors |
Non-Patent Citations (1)
| Title |
|---|
| 吴兵: "《多Agent推荐系统及应用》", 31 July 2014 * |
Cited By (17)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107346107A (en) * | 2016-05-04 | 2017-11-14 | 深圳光启合众科技有限公司 | Diversified motion control method and system and the robot with the system |
| CN108225276B (en) * | 2018-01-11 | 2020-06-23 | 中国人民解放军战略支援部队航天工程大学 | Single-star imaging target motion characteristic inversion method and system |
| CN108225276A (en) * | 2018-01-11 | 2018-06-29 | 中国人民解放军战略支援部队航天工程大学 | A kind of list star imageable target kinetic characteristic inversion method and system |
| CN109493644A (en) * | 2018-11-16 | 2019-03-19 | 四川大学 | A kind of four-dimensional flying track conjecture method excavated based on history track data |
| CN109493644B (en) * | 2018-11-16 | 2020-08-14 | 四川大学 | Four-dimensional track conjecture method based on historical track data mining |
| CN109739921A (en) * | 2019-01-07 | 2019-05-10 | 北京云基数技术有限公司 | A kind of big data acquisition system and method |
| CN109783515A (en) * | 2019-01-25 | 2019-05-21 | 上海创景信息科技有限公司 | More relation data tracks retroactive method and system based on database |
| CN109992569A (en) * | 2019-02-19 | 2019-07-09 | 平安科技（深圳）有限公司 | Cluster log feature extracting method, device and storage medium |
| CN111145599A (en) * | 2019-12-17 | 2020-05-12 | 中国民用航空总局第二研究所 | Curve flight segment and error distribution establishing method |
| CN113484841A (en) * | 2021-07-26 | 2021-10-08 | 中国船舶重工集团公司第七二三研究所 | Track head processing method under search radar multitask parallel |
| CN114063653A (en) * | 2021-12-03 | 2022-02-18 | 中国人民解放军63892部队 | Space and time consistency constraint small unmanned aerial vehicle high-precision track control method |
| CN114063653B (en) * | 2021-12-03 | 2023-12-29 | 中国人民解放军63892部队 | Space and time consistency constraint unmanned aerial vehicle high-precision track control method |
| CN116415161A (en) * | 2023-04-17 | 2023-07-11 | 中交第四航务工程局有限公司 | Fitting complementation method for geological drilling detection and different physical wave detection data of string-shaped karst cave |
| CN116415161B (en) * | 2023-04-17 | 2023-09-08 | 中交第四航务工程局有限公司 | Fitting complementation method for geological drilling detection and different physical wave detection data of string-shaped karst cave |
| CN117098255A (en) * | 2023-10-19 | 2023-11-21 | 南京波达电子科技有限公司 | Edge calculation-based decentralization radar ad hoc network method |
| CN117098255B (en) * | 2023-10-19 | 2023-12-15 | 南京波达电子科技有限公司 | Edge calculation-based decentralization radar ad hoc network method |
| CN118034605A (en) * | 2024-03-15 | 2024-05-14 | 上海正泰自动化软件系统有限公司 | Data processing method and device, electronic equipment and storage medium |
Also Published As
| Publication number | Publication date |
|---|---|
| CN105353644B (en) | 2018-06-15 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN105353644A (en) | Radar target track derivative system on the basis of information mining of real-equipment data and method thereof | |
| CN113157800B (en) | Identification method for discovering dynamic target in air in real time | |
| CN104268428A (en) | Visual configuration method for index calculation | |
| CN110645988B (en) | Unmanned aerial vehicle path planning method based on optimal service life | |
| CN108133059A (en) | Couple in heterogeneous database system containing extensive low-power cluster it is effective under push away | |
| CN118017564B (en) | Energy storage method based on open source hong Meng system | |
| CN110427298A (en) | A kind of Automatic Feature Extraction method of distributed information log | |
| CN117172509B (en) | Construction project distribution system based on decoration construction progress analysis | |
| CN114338418A (en) | Virtual-real combined information network verification platform | |
| CN110827326A (en) | Method, device, equipment and storage medium for generating simulation human-vehicle conflict scene model | |
| US11836637B2 (en) | Construction method of human-object-space interaction model based on knowledge graph | |
| CN112766595A (en) | Command control device, method, system, computer equipment and medium | |
| CN111159152B (en) | Secondary operation and data fusion method based on big data processing technology | |
| CN115981804A (en) | Industrial big data calculation task scheduling management system | |
| CN114741185A (en) | Edge computing system for multi-target video monitoring and working method thereof | |
| CN112925964A (en) | Big data acquisition method based on cloud computing service and big data acquisition service system | |
| CN109889366B (en) | Network traffic increment counting and analyzing method and system | |
| CN115130333B (en) | Digital twin model merging method | |
| CN109238271B (en) | Line fitting method based on time | |
| CN116668640A (en) | Equipment grouping method, device, storage medium and electronic device | |
| CN114417633B (en) | Network shooting range scene construction method and system based on parallel simulation six-tuple | |
| CN114757448A (en) | Manufacturing inter-link optimal value chain construction method based on data space model | |
| CN109508335B (en) | Mass ground clutter data classification storage method | |
| CN113434869A (en) | Data processing method and AI system based on threat perception big data and artificial intelligence | |
| CN117459516B (en) | Automatic data file synchronization method and system based on intelligent home system |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant |