CN106707233A - Multi-side positioning method and multi-side positioning device based on outlier detection - Google Patents
Multi-side positioning method and multi-side positioning device based on outlier detection Download PDFInfo
- Publication number
- CN106707233A CN106707233A CN201710124438.4A CN201710124438A CN106707233A CN 106707233 A CN106707233 A CN 106707233A CN 201710124438 A CN201710124438 A CN 201710124438A CN 106707233 A CN106707233 A CN 106707233A
- Authority
- CN
- China
- Prior art keywords
- cluster
- point
- coordinate
- outlier
- candidate
- 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.)
- Pending
Links
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
- G01S5/00—Position-fixing by co-ordinating two or more direction or position line determinations; Position-fixing by co-ordinating two or more distance determinations
- G01S5/02—Position-fixing by co-ordinating two or more direction or position line determinations; Position-fixing by co-ordinating two or more distance determinations using radio waves
- G01S5/0278—Position-fixing by co-ordinating two or more direction or position line determinations; Position-fixing by co-ordinating two or more distance determinations using radio waves involving statistical or probabilistic considerations
-
- 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
- G01S5/00—Position-fixing by co-ordinating two or more direction or position line determinations; Position-fixing by co-ordinating two or more distance determinations
- G01S5/02—Position-fixing by co-ordinating two or more direction or position line determinations; Position-fixing by co-ordinating two or more distance determinations using radio waves
- G01S5/0273—Position-fixing by co-ordinating two or more direction or position line determinations; Position-fixing by co-ordinating two or more distance determinations using radio waves using multipath or indirect path propagation signals in position determination
-
- 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
- G01S5/00—Position-fixing by co-ordinating two or more direction or position line determinations; Position-fixing by co-ordinating two or more distance determinations
- G01S5/02—Position-fixing by co-ordinating two or more direction or position line determinations; Position-fixing by co-ordinating two or more distance determinations using radio waves
- G01S5/0284—Relative positioning
Landscapes
- Physics & Mathematics (AREA)
- Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Radar, Positioning & Navigation (AREA)
- Remote Sensing (AREA)
- Probability & Statistics with Applications (AREA)
- Position Fixing By Use Of Radio Waves (AREA)
Abstract
The embodiment of the invention provides a multi-side positioning method based on outlier detection. According to the multi-side positioning method, a preset number of clusters are obtained by virtue of cluster analysis on a positioning result, and outliers in the clusters are separated so as to remove measurement values with relatively errors from positioning data, so that the negative effects of the measurement values with relatively large errors to the positioning result are eliminated; and finally, the clusters are subjected to weighted averaging, so as to obtain a positioning result of an unknown node, so that the technical problem that the negative effects of the measurement values with relatively large errors to the positioning result cannot be completely eliminated by virtue of a multi-side positioning method of an existing positioning wireless sensor is solved, the influence caused by errors in distance information to a position calculation result can be effectively reduced, and the positioning precision of the unknown node is effectively improved. The embodiment of the invention further provides a multi-side positioning device based on the outlier detection.
Description
Technical field
The present invention relates to wireless sensor network field, more particularly to a kind of polygon localization method based on outlier detection
And device.
Background technology
Wireless sensor network (WSN) is a kind of collection sensor technology, wireless communication technology, modern microelectronic technology, letter
The new generation network technology that number treatment technology and embedded technology etc. are integrated, the technology has information gathering, data processing, nothing
The functions such as line transmission.Sensor node of the wireless sensor network by random distribution in a network is individual into these sensor nodes
Collect, analysis, analysis result is simultaneously accurately transferred back to server by the information around treatment node.In monitoring activity, sensor
The clear and definite position of oneself is had to, if without accurate positional information, then the information that sensor is obtained is just nonsensical.Wirelessly
The self-localization algorithm of sensor network directly affects the availability of sensor network, so, the location technology of wireless senser
It is most important.
Location algorithm can be divided into two types, location algorithm based on distance and apart from unrelated location algorithm.It is based on
The location algorithm of distance needs to increase extra hardware device measurement distance and angle, relatively costly；Calculated apart from unrelated positioning
Method exchanges information by topological network without extra hardware device, and cost is relatively low.Location algorithm based on distance is based on
The positioning of TOA, the positioning based on TDOA, the positioning based on AOA, the positioning based on RSSI；Have apart from unrelated location algorithm
RSSI algorithms, DV-Hop algorithms, APIT algorithms.Multilateration is widely used universal side during WSN position locations calculate
Method, whether based on TOA, RSSI method found range or DV-Hop, DV-distance, Amorphous method without range finding
Position calculating can be carried out using polygon positioning mode, therefore multilateration has important in wireless sensor network positioning
Meaning.
Fig. 1 show the schematic diagram of multilateration, if the coordinate of unknown node is U (X, Y), beaconing nodes Bj (j=
1,2 ... coordinate n) is (Xj, Yj), and (j=1,2 ... n) then for dj for distance between unknown node and beaconing nodes
Above formula can be expressed as linear equation AX=B solutions.
The coordinate of unknown node is finally obtained using nonlinear IEM model method, computing formula is as follows：
Some scholars are had in recent years to be improved multilateration, are such as proposed a kind of based on minimal error square
The multilateration of sum.The algorithm calculates error sum of squares of the different beaconing nodes as the meta object generation that disappears respectively, leads to
Selection error sum of squares minimum value is crossed to determine position optimal estimation value.For the shortcoming of conventional wireless location algorithm, You Renti
Go out with received signal strength parameter, the wireless sensor network that compressed sensing technology and polygon measuring method are combined has been determined
Position method.The target orientation problem based on grid is converted into compressed sensing problem first, judges whether target is located in grid
The heart.When the target not heart within a grid, then the fine positioning of target is carried out with polygon measuring method, and employed based on connecing
Receive influence of the base station selected policies environmental factor of signal intensity to location algorithm.Also it has been proposed that Maximum-likelihood estimation is calculated
Method is used for the self-positioning principle of wireless sensor network node, and steepest descent algorithm seeks the principle of nonlinear equation optimal solution.
The document proposes, in the case where range measurement error is larger, maximum likelihood estimation algorithm gained to be optimized using steepest descent algorithm
Node locating value.
However, all in all, the above method is all to be studied for range error and Position-Solving method respectively, main
Investigate the influence for how reducing range error to positioning result, although the position error that range error causes is improved to some extent,
But for the measured value with larger error, above method can not completely remove its influence to positioning result.
Therefore, the polygon localization method of existing positioning wireless senser can not completely remove the measurement with larger error
Negative effect of the value to positioning result is those skilled in the art's technical issues that need to address.
The content of the invention
A kind of polygon localization method and device based on outlier detection are the embodiment of the invention provides, it is existing for solving
The polygon localization method of positioning wireless senser can not completely remove the measured value with larger error positioning result is born
The technical problem that face rings.
A kind of polygon localization method based on outlier detection provided in an embodiment of the present invention, including：
Unknown node and N number of range information of beaconing nodes that acquisition need to be positioned；
Three range informations are often chosen from the range information and obtains one calmly by multilateration calculating
Position result, obtains altogetherIndividual positioning result；
From describedK point is randomly selected in individual positioning result as initialization central point Ui, the k initialization center
The corresponding cluster of point is k UiCluster；
Will be described by cluster algorithmThe initialization central point U is removed in individual positioning resultiRemaining dot-dash
Divide to the k UiCluster, calculates the k UiThe cluster centre point C of clusteriCoordinate and cluster radius Ri；
Calculate UiPoint in cluster is to UiCluster centre point CiEuclidean distance dij, by the Euclidean distance dijMore than UiIt is poly-
Class radius RiCorresponding UiPoint in cluster is labeled as candidate's outlier；
The LOF values of each candidate's outlier are calculated, from UiCandidate's outlier of the LOF values maximum of m, weight are deleted in cluster
Newly calculate k UiThe coordinate of the cluster centre point of cluster is (Cix,Ciy) wherein i=1,2,3 ... ..k；
According to the coordinate (C of cluster centre pointix,Ciy) and UiThe number at remainder strong point seeks weighted average in cluster, obtains
The coordinate of unknown node.
Preferably, it is described to incite somebody to action described by cluster algorithmThe initialization central point U is removed in individual positioning resulti
Left point be divided to the k UiCluster, calculates the k UiThe cluster centre point C of clusteriCoordinate and cluster radius Ri
Specifically include：
Calculate describedThe initialization central point U is removed in individual positioning resultiLeft point to UiEach gathers in cluster
The distance of class central point, the minimum point of selected distance is added to corresponding U from the left pointiIn cluster, recalculate every
Individual UiThe cluster centre point of cluster；
Previous step is repeated, until k UiThe coordinate of the cluster centre point of cluster is constant, and cluster centre point is calculated respectively
CiCoordinate and cluster radius Ri, calculate cluster centre point CiCoordinate and cluster radius RiFormula be：
Wherein, XiIt is i-th UiData point in cluster, niIt is i-th UiThe number of data point in cluster.
Preferably, the calculating UiPoint in cluster is to UiCluster centre point CiEuclidean distance dij, by the Euclidean distance
dijMore than UiCluster radius RiCorresponding UiPoint in cluster is specially labeled as candidate's outlier：
According to UiPoint in cluster is to UiCluster centre point CiEuclidean distance dij=d (hij,Ci) calculate candidate and peel off point set
hij, wherein the Euclidean distance dijMore than UiCluster radius RiCorresponding UiPoint in cluster is labeled as candidate's outlier, meter
Candidate is calculated to peel off point set hijFormula be：
Wherein, RiIt is UiThe cluster radius of cluster, hijIt is UiPoint in cluster, niIt is i-th UiData point in cluster
Number, m is presetting outlier number；
Preferably, the coordinate (C according to cluster centre pointix,Ciy) and UiThe number at remainder strong point is asked and added in cluster
Weight average, the coordinate for obtaining unknown node is specially：
According to the coordinate (C of cluster centre pointix,Ciy) and UiThe seat that the number at remainder strong point passes through unknown node in cluster
Mark computing formula seeks weighted average, and the coordinate computing formula of unknown node is as follows：
Wherein, niIt is i-th UiThe number of the data point in cluster, m is presetting outlier number, and k is UiCluster
Number.
Preferably, also include after the unknown node that the acquisition need to be positioned and N number of range information of beaconing nodes：
Whether the number N of the range information is detected less than 3, if so, then terminating positioning.
A kind of polygon positioner based on outlier detection provided in an embodiment of the present invention, including：
Range information acquisition module, the N number of range information for obtaining the unknown node that need to be positioned and beaconing nodes；
Primary Location module, is calculated for often choosing three range informations from the range information by polygon positioning
Method is calculated and obtains a positioning result, is obtained altogetherIndividual positioning result；
Initialization cluster module, for from describedK point is randomly selected in individual positioning result as initialization central point
Ui, k corresponding cluster of the initialization central point is k UiCluster；
Cluster Analysis module, for inciting somebody to action described by cluster algorithmIn removing the initialization in individual positioning result
Heart point UiLeft point be divided to the k UiCluster, calculates the k UiThe cluster centre point C of clusteriCoordinate and cluster
Radius Ri；
Candidate's outlier determining module, for calculating UiPoint in cluster is to UiCluster centre point CiEuclidean distance dij, will
The Euclidean distance dijMore than UiCluster radius RiCorresponding UiPoint in cluster is labeled as candidate's outlier；
Outlier removing module, the LOF values for calculating each candidate's outlier, from UiThe LOF values of m are deleted in cluster
Maximum candidate's outlier, recalculates k UiThe coordinate of the cluster centre point of cluster is (Cix,Ciy) wherein i=1,2,3 ...
..k；
Unknown node computing module, for the coordinate (C according to cluster centre pointix,Ciy) and UiRemainder strong point in cluster
Number seek weighted average, obtain the coordinate of unknown node.
Preferably, the Cluster Analysis module is specifically included：
Cluster aggregation units, it is described for calculatingThe initialization central point U is removed in individual positioning resultiLeft point
To UiThe distance of each cluster centre point in cluster, the minimum point of selected distance is added to corresponding U from the left pointiIt is poly-
In class, each U is recalculatediThe cluster centre point of cluster；
Cluster cycling element, for repeating previous step, until k UiThe coordinate of the cluster centre point of cluster is constant,
Then cluster centre point C is calculated respectivelyiCoordinate and cluster radius Ri, calculate cluster centre point CiCoordinate and cluster radius Ri
Formula be：
Wherein, XiIt is i-th UiData point in cluster, niIt is i-th UiThe number of data point in cluster.
Preferably, candidate's outlier determining module specifically for：
According to UiPoint in cluster is to UiCluster centre point CiEuclidean distance dij=d (hij,Ci) calculate candidate and peel off point set
hij, wherein the Euclidean distance dijMore than UiCluster radius RiCorresponding UiPoint in cluster is labeled as candidate's outlier, meter
Candidate is calculated to peel off point set hijFormula be：
Wherein, RiIt is UiThe cluster radius of cluster, hijIt is UiPoint in cluster, niIt is i-th UiData point in cluster
Number, m is presetting outlier number；
Preferably, the unknown node computing module specifically for：
According to the coordinate (C of cluster centre pointix,Ciy) and UiThe seat that the number at remainder strong point passes through unknown node in cluster
Mark computing formula seeks weighted average, and the coordinate computing formula of unknown node is as follows：
Wherein, niIt is i-th UiThe number of the data point in cluster, m is presetting outlier number, and k is UiCluster
Number.
Preferably, the embodiment of the present invention also includes：
Whether range information number limits module, for detecting the number N of the range information less than 3, if so, then terminating
Positioning.
As can be seen from the above technical solutions, the embodiment of the present invention has advantages below：
A kind of polygon localization method based on outlier detection provided in an embodiment of the present invention is carried out by positioning result
Cluster analysis obtains the cluster of predetermined number, then the outlier in cluster is separated so that the survey with larger error
Value is removed from location data, eliminates negative effect of the measured value with larger error to positioning result, finally to poly-
Class is weighted the average positioning result for obtaining unknown node, solves the polygon localization method of existing positioning wireless senser
Technical problem of the measured value with larger error to the negative effect of positioning result can not completely be removed, can effectively reduce away from
Influence from error contained by information to position result of calculation, effectively improves the positioning precision of unknown node.
Brief description of the drawings
In order to illustrate more clearly about the embodiment of the present invention or technical scheme of the prior art, below will be to embodiment or existing
The accompanying drawing to be used needed for having technology description is briefly described, it should be apparent that, drawings in the following description are only this
Some embodiments of invention, for those of ordinary skill in the art, without having to pay creative labor, may be used also
Other accompanying drawings are obtained with according to these accompanying drawings.
Fig. 1 is the schematic diagram that the embodiment of the present invention is used to illustrate the existing polygon localization method for positioning wireless senser；
Fig. 2 is a kind of one embodiment of polygon localization method based on outlier detection provided in an embodiment of the present invention
Schematic diagram；
Fig. 3 is a kind of another implementation of polygon localization method based on outlier detection provided in an embodiment of the present invention
Example；
Fig. 4 is the analog node distribution map that the embodiment of the present invention is used to illustrate；
Fig. 5 is the instance graph that the embodiment of the present invention is used to illustrate cluster optimization；
Fig. 6 is the schematic diagram that the embodiment of the present invention is used to illustrate influence of the network node sum to position error；
Fig. 7 is the schematic diagram that the embodiment of the present invention is used to illustrate influence of the beaconing nodes proportion to position error.
Specific embodiment
A kind of polygon localization method and device based on outlier detection are the embodiment of the invention provides, it is existing for solving
The polygon localization method of positioning wireless senser can not completely remove the measured value with larger error positioning result is born
The technical problem that face rings.
To enable that goal of the invention of the invention, feature, advantage are more obvious and understandable, below in conjunction with the present invention
Accompanying drawing in embodiment, is clearly and completely described, it is clear that disclosed below to the technical scheme in the embodiment of the present invention
Embodiment be only a part of embodiment of the invention, and not all embodiment.Based on the embodiment in the present invention, this area
All other embodiment that those of ordinary skill is obtained under the premise of creative work is not made, belongs to protection of the present invention
Scope.
Refer to Fig. 2, an a kind of reality of the polygon localization method based on outlier detection provided in an embodiment of the present invention
Example is applied, including：
101st, N number of range information of the unknown node and beaconing nodes that need to be positioned is obtained；
102nd, three range informations are often chosen from the range information and obtains one by multilateration calculating
Individual positioning result, obtains altogetherIndividual positioning result；
103rd, from describedK point is randomly selected in individual positioning result as initialization central point Ui, the k initialization
The corresponding cluster of central point is k UiCluster；
104th, will be described by cluster algorithmThe initialization central point U is removed in individual positioning resultiResidue
Point is divided to the k UiCluster, calculates the k UiThe cluster centre point C of clusteriCoordinate and cluster radius Ri；
105th, U is calculatediPoint in cluster is to UiCluster centre point CiEuclidean distance dij, by the Euclidean distance dijIt is more than
UiCluster radius RiCorresponding UiPoint in cluster is labeled as candidate's outlier；
106th, the LOF values of each candidate's outlier are calculated, from UiThe candidate that the LOF values maximum of m is deleted in cluster peels off
Point, recalculates k UiThe coordinate of the cluster centre point of cluster is (Cix,Ciy) wherein i=1,2,3 ... ..k；
107th, according to the coordinate (C of cluster centre pointix,Ciy) and UiThe number at remainder strong point seeks weighted average in cluster,
Obtain the coordinate of unknown node.
A kind of polygon localization method based on outlier detection provided in an embodiment of the present invention is carried out by positioning result
Cluster analysis obtains the cluster of predetermined number, then the outlier in cluster is separated so that the survey with larger error
Value is removed from location data, eliminates negative effect of the measured value with larger error to positioning result, finally to poly-
Class is weighted the average positioning result for obtaining unknown node, solves the polygon localization method of existing positioning wireless senser
Technical problem of the measured value with larger error to the negative effect of positioning result can not completely be removed, can effectively reduce away from
Influence from error contained by information to position result of calculation, effectively improves the positioning precision of unknown node.
The embodiment of the present invention carries out three or three and merges using three sides by the range information received to unknown node first
Positioning mode obtains a series of sample point, sample point is pre-processed using K-means clustering methodologies secondly and is classified, and uses again
Outlier detection algorithm LOF by each cluster in influence the larger point of position error to pick out and reject and recalculate k it is poly-
The center point coordinate of class, finally with the number of sample point in each cluster and the business of sample point sum and the central point of each cluster
Coordinate is multiplied and is calculated the coordinate of unknown node.The positioning precision of multilateration can be effectively improved by the present invention.
LOF values in the embodiment of the present invention are consistent with the definition in outlier detection algorithm LOF.LOF values will below be entered
Row detailed description：
In outlier detection algorithm LOF, it is any point o in data set and its k-th nearest neighbors to define k distances
The distance between be referred to as the k distances of point o, be designated as k-distance (o).
K is apart from neighborhood：If o is the arbitrfary point in data set, then all distances to point o are less than the k distances of o in data set
The neighborhood referred to as k that is formed of point apart from neighborhood.
Nk(o)：What the k of the arbitrfary point o in data set was included in neighborhood number a little.
Nk(o)={ o'| o' ∈ D, distance (o, o')≤distance (o) }
Distance (p, q)：Euclidean distance between object p and q.
Reach distance (reachdistk(q←p))：If p, q are any object in data set, point p to point q it is reachable away from
From being that the maximum of Euclidean distance between p, q points and q point k distances is max { k-distance (q), distance (p, q) }；
Local reachability density lrdk(o):The local reachability density of data point o is maximum preceding k in o points to its neighborhood
The inverse of the average value of distance.lrdkO () has weighed sparse degree of the data point o in its nearest k recently point set.
Locally peel off factor LOFk(q):Locally peel off factor representation data point q and a kind of density variation of entirety.If office
Portion peels off factor values much larger than 1, then the local density around q points is larger with the density variation of entirety, it is believed that q is outlier.If
Close to 1, then the local density around q points is smaller with the difference of global density, it is believed that q is normal point for the value of the factor that locally peels off：
Above is an implementation to a kind of polygon localization method based on outlier detection provided in an embodiment of the present invention
Example is described in detail, below by a kind of polygon localization method based on outlier detection provided in an embodiment of the present invention
Another embodiment is described in detail.
Refer to Fig. 2, a kind of polygon localization method based on outlier detection provided in an embodiment of the present invention another
Embodiment, including：
201st, N number of range information of the unknown node and beaconing nodes that need to be positioned is obtained；
If the number of range information between beaconing nodes that unknown node can be received is N, the range information is { d1,
d2,d3.....dN}；
202nd, detect whether the number N of the range information is less than 3, if so, then terminating positioning.
The number N of information of adjusting the distance is judged, if N<3 cannot use multilateration, and positioning terminates；If N
>=3, then perform next step.
203rd, three range informations are often chosen from the range information and obtains one by multilateration calculating
Individual positioning result, obtains altogetherIndividual positioning result；
From N number of range information optional 3 as one, then obtainIndividual positioning result, is designated as
The range information for being received to unknown node carries out three or three merging and obtains a series of using three side positioning modes
Sample point；
204th, from describedK point is randomly selected in individual positioning result as initialization central point Ui, the k initialization
The corresponding cluster of central point is k UiCluster；
K point is randomly selected in set U and is designated as U as initialization centeri={ (Uix,Uiy) | i=1,2,3...k }, respectively
Cluster corresponding to individual central point is UiCluster；
205th, calculate describedThe initialization central point U is removed in individual positioning resultiLeft point to UiIt is each in cluster
The distance of individual cluster centre point, the minimum point of selected distance is added to corresponding U from the left pointiIn cluster, count again
Calculate each UiThe cluster centre point of cluster；
Remaining point assigns each data point to the distance of each central point in cluster according to distance minimum in set of computations U
To in closest cluster, the average of each cluster is recalculated as cluster centre point；
206th, step 205 is repeated, until k UiThe coordinate of the cluster centre point of cluster is constant, and cluster is calculated respectively
Central point CiCoordinate and cluster radius Ri, calculate cluster centre point CiCoordinate and cluster radius RiFormula be：
Wherein, XiIt is i-th UiData point in cluster, niIt is i-th UiThe number of data point in cluster.
Can also be described as：Repeat step 205 no longer changes up to the coordinate of the k central point of cluster, and k is calculated respectively
The center point coordinate of cluster, cluster radius.
207th, according to UiPoint in cluster is to UiCluster centre point CiEuclidean distance dij=d (hij,Ci) calculate candidate peel off
Point set hij, wherein the Euclidean distance dijMore than UiCluster radius RiCorresponding UiPoint in cluster peels off labeled as candidate
Point, calculates candidate and peels off point set hijFormula be：
Wherein, RiIt is UiThe cluster radius of cluster, hijIt is UiPoint in cluster, niIt is i-th UiData point in cluster
Number, m is presetting outlier number；
Also can be described as：Calculate each UiPoint in cluster is to UiCluster centre point CiEuclidean distance dij, compare dijWith
The cluster radius R of correspondence clusteriSize, if dijMore than RiThe point is then labeled as the point in the Candidate Set of outlier.
208th, the LOF values of each candidate's outlier are calculated, from UiThe candidate that the LOF values maximum of m is deleted in cluster peels off
Point, recalculates k UiThe coordinate of the cluster centre point of cluster is (Cix,Ciy) wherein i=1,2,3 ... ..k；
Also can be described as：Calculate hijIn each data point LOF values, then resulting LOF values are sorted, select
M outlier is removed the coordinate for recalculating the k central point of cluster as outlier for LOF maximum preceding m point；
209th, according to the coordinate (C of cluster centre pointix,Ciy) and UiThe number at remainder strong point passes through unknown node in cluster
Coordinate computing formula seek weighted average, the coordinate computing formula of unknown node is as follows：
Wherein, niIt is i-th UiThe number of the data point in cluster, m is presetting outlier number, and k is UiCluster
Number.
The number of contained data point is different in k cluster, as ith cluster UiIn contained point number it is bigger, unknown section
Point is in UiThe probability occurred in cluster is bigger, therefore gives the cluster bigger weight.
The beneficial effect of embodiment to illustrate the invention, below will be by verifying DV-Hop algorithms using of the invention polygon
Whether positioning precision improves after location algorithm, and analysis is of the invention to carry out feasibility.
(1) such as Fig. 4,100 sensors of random distribution are set in the square area of 100m*100m by Matlab
Node, wherein beaconing nodes number are set to 10, and unknown node is set to 90.Red represents beaconing nodes, and black represents unknown section
Point.
Beaconing nodes broadcast message in wireless sensor network, unknown node records the information of beaconing nodes；
Beaconing nodes try to achieve the average of its own beacon node according to the information safeguarded in self information table by weighting scheme
Single-hop distance；
Unknown node is multiplied with the average single-hop distance of nearest beaconing nodes and the hop count between beaconing nodes and obtains unknown
Distance between node and each beaconing nodes.Then, the coordinate data collection of unknown node is calculated with multilateration.Finally,
Data analysis is carried out to the data set with the research of the multilateration based on outlier detection and finally determines unknown node
Coordinate position.
(2) using the average localization error of all nodes in network as criterion.The calculating of average localization error is public
Formula is as follows：
In formula：EavIt is average localization error；(xrel,yrel) and (xest,yest) be respectively unknown node estimated coordinates and
True coordinate；N is total node number amount；N is the quantity of beaconing nodes；R is the communication radius of node.
Fig. 5, it is shown that certain unknown node positioning result, is to unknown node coordinate analysis by outlier algorithm LOF
An instance graph."×" represents the coordinate of unknown node, and the point of black represents the outlier in cluster, red, blue, green
Color, the different cluster of purple difference density；As can be seen from Fig., LOF algorithms can be good at recognizing outlier, the unknown section of raising
The positioning precision of point.
Fig. 6 illustrates the relation between node total number and position error in sensor network.This experiment sets beacon
Node proportion is 10% constant, and communication radius are 20m.Tradition DV-Hop algorithms as can be seen from Figure 4, this paper algorithms all with
The increase of node total number, position error declines, because increasing in the same area internal segment dot density and then improve positioning
Precision.The positioning of the more traditional DV-Hop of this paper algorithms and other two kinds of algorithms is more accurate as can be seen from Figure 4.
Fig. 7 reflects the relation between beaconing nodes proportion and position error.In this experiment, node total number is set
100 are set to, node communication radius R is 20m, beaconing nodes proportion scope changes between 5%~35%.Can be with from figure
Find out, the average localization error of three kinds of location algorithms all tapers off trend, is all as the increase of beaconing nodes averagely positions mistake
Subtractive is small.Because with the increase of beaconing nodes number, network-in-dialing degree enhancing, positioning precision is improved.
The present invention opens the algorithm that topic proposes has a wide applicability, i.e., whether to the RSSI based on range finding or right
It is just as in the DV-Hop algorithms without range finding applicable.This paper algorithms have appearance higher than traditional multilateration
Mistake, and algorithm performance is relatively stable, computation complexity is relatively low.The invention has the characteristics that：(1) scope of application is wider, can
To be applied in various location algorithms.(2) expense cost is relatively low, under the conditions of without extras are increased, can improve algorithm
Accuracy.(3) energy ezpenditure is the integral multiple of traditional multilateration, so the algorithm is effectively improving positioning precision
Meanwhile, energy consumption is relatively low.
Above is to a kind of another reality of the polygon localization method based on outlier detection provided in an embodiment of the present invention
Apply example to be described in detail, below by a kind of polygon positioner based on outlier detection provided in an embodiment of the present invention
One embodiment be described in detail.
A kind of one embodiment of polygon positioner based on outlier detection provided in an embodiment of the present invention, including：
Range information acquisition module, the N number of range information for obtaining the unknown node that need to be positioned and beaconing nodes；
Primary Location module, is calculated for often choosing three range informations from the range information by polygon positioning
Method is calculated and obtains a positioning result, is obtained altogetherIndividual positioning result；
Initialization cluster module, for from describedK point is randomly selected in individual positioning result as initialization central point
Ui, k corresponding cluster of the initialization central point is k UiCluster；
Cluster Analysis module, for inciting somebody to action described by cluster algorithmThe initialization is removed in individual positioning result
Central point UiLeft point be divided to the k UiCluster, calculates the k UiThe cluster centre point C of clusteriCoordinate and poly-
Class radius Ri；
Candidate's outlier determining module, for calculating UiPoint in cluster is to UiCluster centre point CiEuclidean distance dij, will
The Euclidean distance dijMore than UiCluster radius RiCorresponding UiPoint in cluster is labeled as candidate's outlier；
Outlier removing module, the LOF values for calculating each candidate's outlier, from UiThe LOF values of m are deleted in cluster
Maximum candidate's outlier, recalculates k UiThe coordinate of the cluster centre point of cluster is (Cix,Ciy) wherein i=1,2,3 ...
..k；
Unknown node computing module, for the coordinate (C according to cluster centre pointix,Ciy) and UiRemainder strong point in cluster
Number seek weighted average, obtain the coordinate of unknown node.
Cluster Analysis module is specifically included：
Cluster aggregation units, it is described for calculatingThe initialization central point U is removed in individual positioning resultiLeft point
To UiThe distance of each cluster centre point in cluster, the minimum point of selected distance is added to corresponding U from the left pointiIt is poly-
In class, each U is recalculatediThe cluster centre point of cluster；
Cluster cycling element, for repeating previous step, until k UiThe coordinate of the cluster centre point of cluster is constant,
Then cluster centre point C is calculated respectivelyiCoordinate and cluster radius Ri, calculate cluster centre point CiCoordinate and cluster radius Ri
Formula be：
Wherein, XiIt is i-th UiData point in cluster, niIt is i-th UiThe number of data point in cluster.
Candidate's outlier determining module specifically for：
According to UiPoint in cluster is to UiCluster centre point CiEuclidean distance dij=d (hij,Ci) calculate candidate and peel off point set
hij, wherein the Euclidean distance dijMore than UiCluster radius RiCorresponding UiPoint in cluster is labeled as candidate's outlier, meter
Candidate is calculated to peel off point set hijFormula be：
Wherein, RiIt is UiThe cluster radius of cluster, hijIt is UiPoint in cluster, niIt is i-th UiData point in cluster
Number, m is presetting outlier number；
Unknown node computing module specifically for：
According to the coordinate (C of cluster centre pointix,Ciy) and UiThe seat that the number at remainder strong point passes through unknown node in cluster
Mark computing formula seeks weighted average, and the coordinate computing formula of unknown node is as follows：
Wherein, niIt is i-th UiThe number of the data point in cluster, m is presetting outlier number, and k is UiCluster
Number.
The embodiment of the present invention also includes：
Whether range information number limits module, for detecting the number N of the range information less than 3, if so, then terminating
Positioning.
It is apparent to those skilled in the art that, for convenience and simplicity of description, the system of foregoing description,
The specific work process of device and unit, may be referred to the corresponding process in preceding method embodiment, will not be repeated here.
The above, the above embodiments are merely illustrative of the technical solutions of the present invention, rather than its limitations；Although with reference to preceding
Embodiment is stated to be described in detail the present invention, it will be understood by those within the art that：It still can be to preceding
State the technical scheme described in each embodiment to modify, or equivalent is carried out to which part technical characteristic；And these
Modification is replaced, and does not make the spirit and scope of the essence disengaging various embodiments of the present invention technical scheme of appropriate technical solution.
Claims (10)
1. a kind of polygon localization method based on outlier detection, it is characterised in that including：
Unknown node and N number of range information of beaconing nodes that acquisition need to be positioned；
Three range informations are often chosen from the range information one positioning knot is obtained by multilateration calculating
Really, obtain altogetherIndividual positioning result；
From describedK point is randomly selected in individual positioning result as initialization central point Ui, the k initialization central point pair
The cluster answered is k UiCluster；
Will be described by cluster algorithmThe initialization central point U is removed in individual positioning resultiLeft point be divided to
The k UiCluster, calculates the k UiThe cluster centre point C of clusteriCoordinate and cluster radius Ri；
Calculate UiPoint in cluster is to UiCluster centre point CiEuclidean distance dij, by the Euclidean distance dijMore than UiCluster half
Footpath RiCorresponding UiPoint in cluster is labeled as candidate's outlier；
The LOF values of each candidate's outlier are calculated, from UiCandidate's outlier of the LOF values maximum of m is deleted in cluster, is counted again
Calculate k UiThe coordinate of the cluster centre point of cluster is (Cix,Ciy) wherein i=1,2,3 ... ..k；
According to the coordinate (C of cluster centre pointix,Ciy) and UiThe number at remainder strong point seeks weighted average in cluster, obtains unknown
The coordinate of node.
2. a kind of polygon localization method based on outlier detection according to claim 1, it is characterised in that described to pass through
Cluster algorithm will be describedThe initialization central point U is removed in individual positioning resultiLeft point be divided to the k Ui
Cluster, calculates the k UiThe cluster centre point C of clusteriCoordinate and cluster radius RiSpecifically include：
Calculate describedThe initialization central point U is removed in individual positioning resultiLeft point to UiEach cluster centre in cluster
The distance of point, the minimum point of selected distance is added to corresponding U from the left pointiIn cluster, each U is recalculatediCluster
Cluster centre point；
Previous step is repeated, until k UiThe coordinate of the cluster centre point of cluster is constant, and cluster centre point C is calculated respectivelyi's
Coordinate and cluster radius Ri, calculate cluster centre point CiCoordinate and cluster radius RiFormula be：
Wherein, XiIt is i-th UiData point in cluster, niIt is i-th UiThe number of data point in cluster.
3. a kind of polygon localization method based on outlier detection according to claim 1, it is characterised in that the calculating
UiPoint in cluster is to UiCluster centre point CiEuclidean distance dij, by the Euclidean distance dijMore than UiCluster radius RiIt is right
The U for answeringiPoint in cluster is specially labeled as candidate's outlier：
According to UiPoint in cluster is to UiCluster centre point CiEuclidean distance dij=d (hij,Ci) calculate candidate and peel off point set hij,
Wherein described Euclidean distance dijMore than UiCluster radius RiCorresponding UiPoint in cluster is labeled as candidate's outlier, calculates and waits
Select the point set h that peels offijFormula be：
Wherein, RiIt is UiThe cluster radius of cluster, hijIt is UiPoint in cluster, niIt is i-th UiThe number of the data point in cluster,
M is presetting outlier number.
4. a kind of polygon localization method based on outlier detection according to claim 1, it is characterised in that the basis
Coordinate (the C of cluster centre pointix,Ciy) and UiThe number at remainder strong point seeks weighted average in cluster, obtains the seat of unknown node
Mark is specially：
According to the coordinate (C of cluster centre pointix,Ciy) and UiThe coordinate meter that the number at remainder strong point passes through unknown node in cluster
Calculate formula and seek weighted average, the coordinate computing formula of unknown node is as follows：
Wherein, niIt is i-th UiThe number of the data point in cluster, m is presetting outlier number, and k is UiWhat is clustered is individual
Number.
5. a kind of polygon localization method based on outlier detection according to claim 1, it is characterised in that the acquisition
Also include after the unknown node that need to be positioned and N number of range information of beaconing nodes：
Whether the number N of the range information is detected less than 3, if so, then terminating positioning.
6. a kind of polygon positioner based on outlier detection, it is characterised in that including：
Range information acquisition module, the N number of range information for obtaining the unknown node that need to be positioned and beaconing nodes；
Primary Location module, by often choosing three range informations from the range information by based on multilateration
Calculate and obtain a positioning result, obtain altogetherIndividual positioning result；
Initialization cluster module, for from describedK point is randomly selected in individual positioning result as initialization central point Ui, k
Individual corresponding cluster of the initialization central point is k UiCluster；
Cluster Analysis module, for inciting somebody to action described by cluster algorithmThe initialization central point is removed in individual positioning result
UiLeft point be divided to the k UiCluster, calculates the k UiThe cluster centre point C of clusteriCoordinate and cluster radius
Ri；
Candidate's outlier determining module, for calculating UiPoint in cluster is to UiCluster centre point CiEuclidean distance dij, will be described
Euclidean distance dijMore than UiCluster radius RiCorresponding UiPoint in cluster is labeled as candidate's outlier；
Outlier removing module, the LOF values for calculating each candidate's outlier, from UiThe LOF values that m is deleted in cluster are maximum
Candidate's outlier, recalculate k UiThe coordinate of the cluster centre point of cluster is (Cix,Ciy) wherein i=1,2,3 ... ..k；
Unknown node computing module, for the coordinate (C according to cluster centre pointix,Ciy) and UiRemainder strong point is individual in cluster
Number seeks weighted average, obtains the coordinate of unknown node.
7. a kind of polygon positioner based on outlier detection according to claim 6, it is characterised in that the cluster
Analysis module is specifically included：
Cluster aggregation units, it is described for calculatingThe initialization central point U is removed in individual positioning resultiLeft point to Ui
The distance of each cluster centre point in cluster, the minimum point of selected distance is added to corresponding U from the left pointiCluster
In, recalculate each UiThe cluster centre point of cluster；
Cluster cycling element, for repeating previous step, until k UiThe coordinate of the cluster centre point of cluster is constant, Ran Houfen
Ji Suan not cluster centre point CiCoordinate and cluster radius Ri, calculate cluster centre point CiCoordinate and cluster radius RiFormula
For：
Wherein, XiIt is i-th UiData point in cluster, niIt is i-th UiThe number of data point in cluster.
8. a kind of polygon positioner based on outlier detection according to claim 6, it is characterised in that the candidate
Outlier determining module specifically for：
According to UiPoint in cluster is to UiCluster centre point CiEuclidean distance dij=d (hij,Ci) calculate candidate and peel off point set hij,
Wherein described Euclidean distance dijMore than UiCluster radius RiCorresponding UiPoint in cluster is labeled as candidate's outlier, calculates and waits
Select the point set h that peels offijFormula be：
Wherein, RiIt is UiThe cluster radius of cluster, hijIt is UiPoint in cluster, niIt is i-th UiThe number of the data point in cluster,
M is presetting outlier number.
9. a kind of polygon positioner based on outlier detection according to claim 6, it is characterised in that described unknown
Node computing module specifically for：
According to the coordinate (C of cluster centre pointix,Ciy) and UiThe coordinate meter that the number at remainder strong point passes through unknown node in cluster
Calculate formula and seek weighted average, the coordinate computing formula of unknown node is as follows：
Wherein, niIt is i-th UiThe number of the data point in cluster, m is presetting outlier number, and k is UiWhat is clustered is individual
Number.
10. a kind of polygon positioner based on outlier detection according to claim 6, it is characterised in that also include：
Whether range information number limits module, for detecting the number N of the range information less than 3, if so, then terminating fixed
Position.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710124438.4A CN106707233A (en) | 2017-03-03 | 2017-03-03 | Multi-side positioning method and multi-side positioning device based on outlier detection |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710124438.4A CN106707233A (en) | 2017-03-03 | 2017-03-03 | Multi-side positioning method and multi-side positioning device based on outlier detection |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN106707233A true CN106707233A (en) | 2017-05-24 |
Family
ID=58917493
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710124438.4A Pending CN106707233A (en) | 2017-03-03 | 2017-03-03 | Multi-side positioning method and multi-side positioning device based on outlier detection |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN106707233A (en) |
Cited By (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108882225A (en) * | 2018-05-07 | 2018-11-23 | 中山大学 | Safe positioning method based on ranging in a kind of wireless sensor network |
| CN109740694A (en) * | 2019-01-24 | 2019-05-10 | 燕山大学 | A kind of smart grid inartful loss detection method based on unsupervised learning |
| CN110008976A (en) * | 2018-12-05 | 2019-07-12 | 阿里巴巴集团控股有限公司 | A kind of network behavior classification method and device |
| CN110094300A (en) * | 2019-01-24 | 2019-08-06 | 上海电力学院 | Wind turbines to windage losses antidote, device, equipment and medium |
| CN110346757A (en) * | 2019-07-18 | 2019-10-18 | 中电科仪器仪表有限公司 | Anti-multipath time difference positioning method, apparatus and system based on mobile measuring station |
| CN112130154A (en) * | 2020-08-21 | 2020-12-25 | 哈尔滨工程大学 | Self-adaptive K-means outlier de-constraint optimization method for fusion grid LOF |
| CN112230056A (en) * | 2020-09-07 | 2021-01-15 | 国网河南省电力公司电力科学研究院 | Multi-harmonic source contribution calculation method based on OFMMK-Means clustering and composite quantile regression |
| CN112699226A (en) * | 2020-12-29 | 2021-04-23 | 江苏苏宁云计算有限公司 | Method and system for semantic confusion detection |
| CN113068121A (en) * | 2021-03-31 | 2021-07-02 | 建信金融科技有限责任公司 | Positioning method, positioning device, electronic equipment and medium |
| CN113490140A (en) * | 2021-07-05 | 2021-10-08 | 南京领创信息科技有限公司 | Method for detecting abnormal position of group member |
| CN114079858A (en) * | 2021-10-25 | 2022-02-22 | 摩拜(北京)信息技术有限公司 | Positioning method and device of electronic equipment and storage medium |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US8750161B1 (en) * | 2010-12-20 | 2014-06-10 | At&T Intellectual Property I, L.P. | Metropolitan IP aggregation network design tool |
| CN104159297A (en) * | 2014-08-19 | 2014-11-19 | 吉林大学 | Multilateration algorithm of wireless sensor networks based on cluster analysis |
| CN105635964A (en) * | 2015-12-25 | 2016-06-01 | 河海大学 | Wireless sensor network node localization method based on K-medoids clustering |
-
2017
- 2017-03-03 CN CN201710124438.4A patent/CN106707233A/en active Pending
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US8750161B1 (en) * | 2010-12-20 | 2014-06-10 | At&T Intellectual Property I, L.P. | Metropolitan IP aggregation network design tool |
| CN104159297A (en) * | 2014-08-19 | 2014-11-19 | 吉林大学 | Multilateration algorithm of wireless sensor networks based on cluster analysis |
| CN105635964A (en) * | 2015-12-25 | 2016-06-01 | 河海大学 | Wireless sensor network node localization method based on K-medoids clustering |
Non-Patent Citations (2)
| Title |
|---|
| 张乙竹: "基于K-means 聚类点密度的WSNs 加权质心定位算法", 《传感器与微***》 * |
| 陶晶: "基于聚类和密度的离群点检测方法", 《中国优秀硕士学位论文全文数据库 信息科技辑》 * |
Cited By (15)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108882225B (en) * | 2018-05-07 | 2020-09-18 | 中山大学 | Safe positioning method based on distance measurement in wireless sensor network |
| CN108882225A (en) * | 2018-05-07 | 2018-11-23 | 中山大学 | Safe positioning method based on ranging in a kind of wireless sensor network |
| CN110008976A (en) * | 2018-12-05 | 2019-07-12 | 阿里巴巴集团控股有限公司 | A kind of network behavior classification method and device |
| CN109740694A (en) * | 2019-01-24 | 2019-05-10 | 燕山大学 | A kind of smart grid inartful loss detection method based on unsupervised learning |
| CN110094300A (en) * | 2019-01-24 | 2019-08-06 | 上海电力学院 | Wind turbines to windage losses antidote, device, equipment and medium |
| CN110346757B (en) * | 2019-07-18 | 2022-03-01 | 中电科思仪科技股份有限公司 | Anti-multipath time difference positioning method, device and system based on mobile measuring station |
| CN110346757A (en) * | 2019-07-18 | 2019-10-18 | 中电科仪器仪表有限公司 | Anti-multipath time difference positioning method, apparatus and system based on mobile measuring station |
| CN112130154A (en) * | 2020-08-21 | 2020-12-25 | 哈尔滨工程大学 | Self-adaptive K-means outlier de-constraint optimization method for fusion grid LOF |
| CN112230056A (en) * | 2020-09-07 | 2021-01-15 | 国网河南省电力公司电力科学研究院 | Multi-harmonic source contribution calculation method based on OFMMK-Means clustering and composite quantile regression |
| CN112230056B (en) * | 2020-09-07 | 2022-04-26 | 国网河南省电力公司电力科学研究院 | Multi-harmonic-source contribution calculation method based on OFMMK-Means clustering and composite quantile regression |
| CN112699226A (en) * | 2020-12-29 | 2021-04-23 | 江苏苏宁云计算有限公司 | Method and system for semantic confusion detection |
| CN113068121A (en) * | 2021-03-31 | 2021-07-02 | 建信金融科技有限责任公司 | Positioning method, positioning device, electronic equipment and medium |
| CN113490140A (en) * | 2021-07-05 | 2021-10-08 | 南京领创信息科技有限公司 | Method for detecting abnormal position of group member |
| CN113490140B (en) * | 2021-07-05 | 2023-07-07 | 南京领创信息科技有限公司 | Method for detecting abnormal position of group member |
| CN114079858A (en) * | 2021-10-25 | 2022-02-22 | 摩拜(北京)信息技术有限公司 | Positioning method and device of electronic equipment and storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN106707233A (en) | Multi-side positioning method and multi-side positioning device based on outlier detection | |
| CN108650628B (en) | Indoor positioning method combining distance measurement and fingerprint based on Wi-Fi network | |
| KR102116824B1 (en) | Positioning system based on deep learnin and construction method thereof | |
| Kumar et al. | Range-free 3D node localization in anisotropic wireless sensor networks | |
| CN102665277B (en) | A kind of method that wireless sensor network interior joint is positioned | |
| CN107027148B (en) | Radio Map classification positioning method based on UE speed | |
| CN104581943B (en) | Node positioning method for Distributed Wireless Sensor Networks | |
| Yang et al. | Localization algorithm in wireless sensor networks based on semi-supervised manifold learning and its application | |
| CN109640262B (en) | Positioning method, system, equipment and storage medium based on mixed fingerprints | |
| CN103533647A (en) | Radio frequency map self-adaption positioning method based on clustering mechanism and robust regression | |
| CN108650626A (en) | A kind of fingerprinting localization algorithm based on Thiessen polygon | |
| CN113596989B (en) | Indoor positioning method and system for intelligent workshop | |
| CN103491506A (en) | Method and system for cooperatively locating heterogeneous network based on WLAN and WSN | |
| Wang et al. | Research on APIT and Monte Carlo method of localization algorithm for wireless sensor networks | |
| CN108737952A (en) | Based on the improved polygon weighted mass center localization method of RSSI rangings | |
| CN103929717A (en) | Wireless sensor network positioning method based on weight Voronoi diagrams | |
| CN107290714B (en) | Positioning method based on multi-identification fingerprint positioning | |
| CN110933604A (en) | KNN indoor positioning method based on position fingerprint time sequence characteristics | |
| Kanjo et al. | CrowdTracing: overcrowding clustering and detection system for social distancing | |
| CN109541537B (en) | Universal indoor positioning method based on ranging | |
| Svertoka et al. | Evaluation of real-life LoRaWAN localization: Accuracy dependencies analysis based on outdoor measurement datasets | |
| CN107872873A (en) | Internet-of-things terminal localization method and device | |
| Zheng et al. | RSS-based indoor passive localization using clustering and filtering in a LTE network | |
| KR20120048375A (en) | Knn/pcm hybrid mehod using gath-geva method for indoor location determination in waln | |
| CN113219405B (en) | Indoor dynamic multi-target passive positioning and quantity estimation method |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20170524 |