CN105717505B - The data correlation method of multiple target tracking is carried out using Sensor Network - Google Patents
The data correlation method of multiple target tracking is carried out using Sensor Network Download PDFInfo
- Publication number
- CN105717505B CN105717505B CN201610088231.1A CN201610088231A CN105717505B CN 105717505 B CN105717505 B CN 105717505B CN 201610088231 A CN201610088231 A CN 201610088231A CN 105717505 B CN105717505 B CN 105717505B
- Authority
- CN
- China
- Prior art keywords
- mrow
- msub
- target
- msubsup
- mover
- 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.)
- Active
Links
- 238000000034 method Methods 0.000 title claims abstract description 30
- 238000005259 measurement Methods 0.000 claims abstract description 28
- 238000004364 calculation method Methods 0.000 claims abstract description 9
- 239000013598 vector Substances 0.000 claims description 17
- 239000011159 matrix material Substances 0.000 claims description 8
- 238000005070 sampling Methods 0.000 claims description 7
- 238000012544 monitoring process Methods 0.000 claims description 4
- 238000010606 normalization Methods 0.000 claims description 4
- 230000002708 enhancing effect Effects 0.000 claims description 3
- 241000854291 Dianthus carthusianorum Species 0.000 description 6
- 238000001514 detection method Methods 0.000 description 5
- 230000033001 locomotion Effects 0.000 description 3
- 238000004891 communication Methods 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 230000001965 increasing effect Effects 0.000 description 2
- 238000011160 research Methods 0.000 description 2
- 206010027476 Metastases Diseases 0.000 description 1
- 230000000739 chaotic effect Effects 0.000 description 1
- 238000000205 computational method Methods 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 230000004807 localization Effects 0.000 description 1
- 230000009401 metastasis Effects 0.000 description 1
- 238000012545 processing Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01S—RADIO DIRECTION-FINDING; RADIO NAVIGATION; DETERMINING DISTANCE OR VELOCITY BY USE OF RADIO WAVES; LOCATING OR PRESENCE-DETECTING BY USE OF THE REFLECTION OR RERADIATION OF RADIO WAVES; ANALOGOUS ARRANGEMENTS USING OTHER WAVES
- G01S13/00—Systems using the reflection or reradiation of radio waves, e.g. radar systems; Analogous systems using reflection or reradiation of waves whose nature or wavelength is irrelevant or unspecified
- G01S13/66—Radar-tracking systems; Analogous systems
Landscapes
- Engineering & Computer Science (AREA)
- Radar, Positioning & Navigation (AREA)
- Remote Sensing (AREA)
- Computer Networks & Wireless Communication (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Radar Systems Or Details Thereof (AREA)
- Navigation (AREA)
Abstract
The present invention provides a kind of data correlation method that multiple target tracking is carried out using Sensor Network, including by sensor node sub-clustering；Node gathers the measurement information of target；Calculate the state discrete time equation of target；Calculate the status predication result of target；Calculate posterior probability of the target k in node i；Target following is carried out using Bayesian frame；Calculate the Posterior probability distribution of target；Repeat the tracking that above step completes target.The present invention is in order to accurately tracking multiple target, a kind of new method is proposed to solve the problems, such as the data correlation in wireless sensor network multi-target tracking, this method is based on Bayesian frame, not using only target location status information, while also operating speed status information.Compared with the method for some other modeling, method proposed by the invention is that calculation amount is small, precision is high, particularly realize intersect maneuvering target tracking in terms of can be in the case of low overhead accurately to multiple target into line trace.
Description
Technical field
The invention belongs to measurement and control areas, and in particular to a kind of data association that multiple target tracking is carried out using Sensor Network
Method.
Background technology
In wireless sensor network, the ability for tracking target is most important in numerous applications.But due in energy, sense
Know the limitation in ability, communication, storage and computing capability, they in application of higher wireless sensor network realize conventional target with
Track solution brings challenge.Since the collection of any data, processing and the operation of propagation information cause the increasing of resource consumption
Add, wireless sensor network target track algorithm should be able to effectively utilize resource, also reduce computation complexity.
At present, monotrack has obtained good research, but multiple target tracking (MTT) problem still has some to need to solve
Key issue certainly.Wherein data correlation is the problem of MTT is most difficult.As shown in Figure 1 under simple scenario two targets with
Track, including three phases.The distance between the first stage, target is big, and the target can individually be tracked.Second
In the stage, target is closer to each other, this eventually results in measurement distortion.The distortion measurement may cause mixed between measurement and target
Confuse, this distortion at this moment can be handled using data correlation.The distance between the phase III, target becomes larger, monotrack
Model is applicable once again.Chaotic or lost target can be caused by being happened at the mistake of second stage, then again first in the 3rd stage
Beginningization or to re-recognize be necessary.
Now with some researchs on wireless sensor network MTT, for example a kind of nothing is proposed in the frame of IDSQ
The algorithm of line sensor network MTT.Graphic model is used to calculate the association between dbjective state and sensor measurement.Target category
Property information be used for handling the measurement of distortion target.The algorithm realizes Distributed Calculation by the cooperation between sensor node
And localization process.However, introducing additional attribute information adds requirement to hardware complexity, also increase cost, calculating,
Communication and energy expense.
The content of the invention
It is an object of the invention to provide a kind of overhead is smaller, data correlation is accurate, so as to multiple target into
The data correlation method that multiple target tracking is carried out using Sensor Network that row accurately tracks.
This data correlation method that multiple target tracking is carried out using Sensor Network provided by the invention, is included the following steps：
S1. the sensor node monitored in region is subjected to sub-clustering；
S2. when target enters monitoring region, the measurement information of target is gathered by sensor node；
S3. the state discrete time equation of target is calculated；
S4. the result of calculation obtained according to step S3 calculates the status predication result of target using following formula：
WhereinIt is distributed for target k in the state at t-1 moment；
And the prediction calculating for measuring distribution is as follows：
Wherein,It is likelihood function；
S5. according to the result of calculation of step S4, posterior probability of the target k in node i is calculated：
S6. data correlation is carried out according to step S5, and target following is carried out using Bayesian frame；
S7. the Posterior probability distribution of target is calculated using following formula：
Wherein, α is a normalization factor, and the purpose is to willValue be transformed between 0~1；K=1 ..., N is target k vectorial in the enhancing of t moment, wherein xk(t) it is position vector,It is velocity vector,For target k the state estimation of t moment expectation；
S8. step S2~S7 is repeated, completes the tracking of target.
Sub-clustering described in step S1 carries out sub-clustering for node is pressed distance.
Sensor described in step S1 is receiving intensity sensor.
The measurement information of acquisition target described in step S2, for the signal received using following formula calculate node i in t moment
Intensity：
Wherein, zi(t) it is signal strength that node i is received in t moment；vi(t) it is ambient noise, N is the number of target
Amount, ak(t) it is signal strengths of the target k in t moment, xk(t) it is position vectors of the target k in t moment, ξiIt is sensor node i
Position vector.
State discrete time equation described in step S3, to calculate state discrete time equation using following formula：
yk(t)=Ayk(t-1)+ωk
Wherein,It is state-transition matrix, ωkIt is the Gaussian noise sequence that average is 0；In same sampling
Period, the variation of speed can be expressed asAnd using it as the selection criteria of the value of q, wherein Q is covariance
Matrix
Target following is carried out using Bayesian frame described in step S6, to calculate the estimation of target location using following formula
Value：
Wherein p (xk| z1:k) represent according to measurement z1:k={ z1,z2,…zkEstimation xkProbability density function；Measure zk=
hk(xk,wk) it is state model xk=fk(xk-1,vk-1) measured value.
The present invention is in order to accurately tracking multiple target, it is proposed that a kind of new method solves wireless sensing
Data correlation problem in the tracking of device network multi-target, this method are based on Bayesian frame, do not believe using only target location state
Breath, while also operating speed status information.Compared with the method for some other modeling, method calculation amount proposed by the invention
It is small, precision is high, particularly realize intersect maneuvering target tracking in terms of can be in the case of low overhead exactly to multiple target
Into line trace.
Description of the drawings
Fig. 1 is the example schematic of background technology.
Fig. 2 is flow chart of the method for the present invention.
Fig. 3 is specific embodiments of the present invention schematic diagram.
Specific embodiment
Be illustrated in figure 2 flow chart of the method for the present invention, it is provided by the invention it is this using Sensor Network carry out multiple target with
The data correlation method of track, includes the following steps：
S1. by taking receiving intensity sensor as an example, the sensor node in Sensor Network is pressed into distance and carries out sub-clustering；
S2. when target enters the monitoring region of Sensor Network, the measurement information of target is gathered by sensor node：Node
I is in the signal strength that t moment receives：
Wherein, vi(t) it is ambient noise, N is the quantity of target, ak(t) it is signal strengths of the target k in t moment, xk(t)
It is position vectors of the target k in t moment, ξiIt is the position vector of sensor node i.
S3. the state discrete time equation of target is calculated：
yk(t)=Ayk(t-1)+ωk
Wherein,It is state-transition matrix, ωkIt is the Gaussian noise sequence that average is 0；In same sampling
Period, the variation of speed can be expressed asAnd using it as the selection criteria of the value of q, wherein Q is covariance
Matrix
S4. the result of calculation obtained according to step S3 calculates the status predication result of target using following formula：
WhereinIt is distributed for target k in the state at t-1 moment；
And the prediction calculating for measuring distribution is as follows：
Wherein,It is likelihood function；
S5. according to the result of calculation of step S4, posterior probability of the target k in node i is calculated：
S6. data correlation is carried out according to step S5, and the estimate of target location is calculated using following formula：
Wherein p (xk| z1:k) represent according to measurement z1:k={ z1,z2,…zkEstimation xkProbability density function；Measure zk=
hk(xk,wk) it is state model xk=fk(xk-1,vk-1) measured value；
S7. the Posterior probability distribution of target is calculated using following formula：
Wherein, α is a normalization factor, and the purpose is to willValue be transformed between 0~1；K=1 ..., N is target k vectorial in the enhancing of t moment, wherein xk(t) it is position vector,It is velocity vector,For target k the state estimation of t moment expectation；
S8. step S2~S7 is repeated, completes the tracking of target.
The method of the present invention is further described below in conjunction with a specific embodiment and Fig. 2：
As shown in Fig. 2, substantial amounts of inexpensive sensor node is deployed on highway.These nodes are pressed distance point by we
Cluster often has some sensor nodes in cluster, and cluster head is elected in turn by sensor node (target following node), can also use dynamic
State cluster head selection algorithm is chosen.The each cluster of original state has a small amount of sensor node (target detection node) to be in moving type
State, other sensors node (target following node) are completely in sleep state；When detecting that target occurs, target detection section
The other sensors node that point will be waken up in cluster carries out target following, and target detection node then enters sleep state at this time.
The measurement information into during line trace, collecting the measurement information on target, and is being sent multiple target by sensor node
Data correlation is carried out to cluster head, and carries out Bayesian Estimation to carry out target following.At the same time, target is also considered
Location status information and speed state information are to calculate the Posterior probability distribution of target, it will be used in next estimation into
Row data correlation.
As can be seen from FIG. 2, for target A in the effective scope of detection of node 1,2 and 3, target B is effective node 2,3 and 4
In investigative range, target C is in the effective scope of detection of node 6.The measurement information of this 6 nodes is as shown in the figure, wherein x is represented
Dbjective state, z represent the measurement information of node.According to this figure, sensor 2 and 3 is influenced simultaneously by target A and B；Therefore it is single
The measurement of sensor be subject to multiple targets it is possible that can be influenced.On the other hand, single target can be simultaneously by multiple sensors
Node perceived arrives, for example other node 1,2 and 3 perceives target A simultaneously, and target A is in the state of t momentWithPhase
Association, therefore the information that can be measured by multiple sensors merges to obtain accurate target information.
Multiple target tracking data correlation method is described in detail below, flow is as shown in figure 3, be as follows：
S1. initialize：The suitable deployed position of node is selected, member node is deployed in the both sides of road；Assuming that
Detect corresponding target, and the number of target is K, and used multiple target tracking algorithm.
S2. target following：Information gathering is carried out using inexpensive sensor, when multiple targets are close to each other, carries out data
Association calculates, accurately to carry out multiple target tracking.
This step specifically includes following steps：
1) measurement information is gathered by bunch member node, and the status information is sent to cluster head, cluster head selection is to use
Certain mechanism allows each member node to be elected in turn, to realize balancing energy.
Assuming that k is the quantity of signal source, node i is following formula (1) in the signal strength that t moment receives：
zi(t)=si(t)+υi(t)
Wherein, vi(t) it is ambient noise, is normally provided as that average is 0, variance isGaussian noise, si(t) it is next
From the signal strength of signal source (monitoring objective) k, such as following formula (2) is calculated：
Wherein, ak(t) it is signal strengths of the signal source k in t moment, xk(t) it is position vectors of the signal source k in t moment,
ξiIt is the position vector of sensor node.
According to formula (1) and (2), node i is in the measurement z of t momenti(t) can calculate such as following formula (3)：
2) after cluster head receives the measurement information that step 1) is calculated, data correlation calculating is carried out.
Assuming that tsIt it is a sampling interval, the discrete time equation of state can be represented such as following formula (6)：
yk(t)=Ayk(t-1)+ωk
Wherein,It is state-transition matrix, ωkIt is the Gaussian noise sequence that average is 0.In same sampling
Period, the variation of speed can be expressed asAnd using it as the selection criteria of the value of q, wherein Q is association side
Poor matrix
It is understood that states of the signal source k at the t-1 moment is distributed asState can be obtained according to formula (6)
Metastasis model, then status predicationIt can calculate such as following formula (7)：
So, measuring the prediction of distribution can calculate such as following formula (8)：
Wherein,It is likelihood function.
According to formula (7), the measurement of node i includes N number of target, then posterior probability of the target k in node i can be counted
It calculates as follows：
3) calculate a Bayesian Estimation and carry out target following.Target following can be modeled as a dynamic state estimator
Problem, and the frame based on bayes method can be very good to solve the problems, such as dynamic state estimator.Assuming that state model is xk=fk
(xk-1,vk-1), wherein xkIt is dbjective state, vkIt is process noise.At the same time, measurement model zk=hk(xk,wk), wherein wk
It is measurement noise.So bayes method can be according to measurement z1:k={ z1,z2,…zkEstimation xkProbability density function
(PDF)：p(xk| z1:k).In the application of Sensor Network multiple target tracking, the motion model of target is typically uncertain and unstable
, and computing resource is limited.So generally select motion model of the Brownian Model as target.In fact, in wireless sensing
In the MTT applications of device network, the variation of target movement is more much smaller than the sample rate of sensing node, and (for example the sample rate of MICAz can
With more than 100 hertz or more), therefore constant speed (CV) model can be applied in the sampling interval.
Assuming that in moment k-1, probability density function (PDF) p (x are givenk-1| z1:k-1).So PDF of dbjective state is p
(xk| z1:k-1)=∫ p (xk| xk-1)p(xk-1| z1:k-1)dxk-1.At the k moment, when obtaining measuring zkWhen, it can according to bayes method
To calculate the estimate of target location, computational methods such as following formula (9)：
4) Posterior probability distribution of target is calculated, it will be used to carry out data correlation in next estimation.
Assuming that there is m sensor to participate in state estimations of the target k in t moment, the measurement of all the sensors is mutual indepedent,
The Posterior probability distribution of target k can calculate as follows：
Wherein, α is a normalization factor,K=1 ..., N is increasings of the target k in t moment
Dominant vector, wherein xk(t) it is position vector,It is velocity vector.So, state estimations of the target k in t moment is obtained
Afterwards,It can it is expected to obtain by calculating.
S2. judge whether to continue to track, if it is, entering step S1, otherwise terminate.
Claims (6)
1. a kind of data correlation method that multiple target tracking is carried out using Sensor Network, is included the following steps：
S1. the sensor in Sensor Network is subjected to sub-clustering；
S2. when target enters the monitoring region of Sensor Network, the measurement information of target is gathered by sensor node；
S3. the state discrete time equation of target is calculated；
S4. the result of calculation obtained according to step S3 calculates the status predication result of target using following formula：
<mrow>
<mover>
<mi>p</mi>
<mo>^</mo>
</mover>
<mrow>
<mo>(</mo>
<msub>
<mi>y</mi>
<mi>k</mi>
</msub>
<mo>(</mo>
<mi>t</mi>
<mo>)</mo>
<mo>)</mo>
</mrow>
<mo>=</mo>
<msub>
<mo>&Integral;</mo>
<mrow>
<mi>&psi;</mi>
<mrow>
<mo>(</mo>
<mi>t</mi>
<mo>-</mo>
<mn>1</mn>
<mo>)</mo>
</mrow>
</mrow>
</msub>
<mi>p</mi>
<mrow>
<mo>(</mo>
<msub>
<mi>y</mi>
<mi>k</mi>
</msub>
<mo>(</mo>
<mi>t</mi>
<mo>)</mo>
<mo>|</mo>
<msub>
<mi>y</mi>
<mi>k</mi>
</msub>
<mo>(</mo>
<mrow>
<mi>t</mi>
<mo>-</mo>
<mn>1</mn>
</mrow>
<mo>)</mo>
<mo>)</mo>
</mrow>
<mo>&CenterDot;</mo>
<mover>
<mi>p</mi>
<mo>~</mo>
</mover>
<mrow>
<mo>(</mo>
<msub>
<mi>y</mi>
<mi>k</mi>
</msub>
<mo>(</mo>
<mrow>
<mi>t</mi>
<mo>-</mo>
<mn>1</mn>
</mrow>
<mo>)</mo>
<mo>)</mo>
</mrow>
<msub>
<mi>dy</mi>
<mi>k</mi>
</msub>
<mrow>
<mo>(</mo>
<mi>t</mi>
<mo>-</mo>
<mn>1</mn>
<mo>)</mo>
</mrow>
</mrow>
WhereinIt is distributed for target k in the state at t-1 moment；
And the prediction calculating for measuring distribution is as follows：
<mrow>
<mover>
<mi>p</mi>
<mo>^</mo>
</mover>
<mrow>
<mo>(</mo>
<msubsup>
<mi>z</mi>
<mi>k</mi>
<mi>i</mi>
</msubsup>
<mo>(</mo>
<mi>t</mi>
<mo>)</mo>
<mo>)</mo>
</mrow>
<mo>=</mo>
<msub>
<mo>&Integral;</mo>
<mrow>
<mi>&psi;</mi>
<mrow>
<mo>(</mo>
<mi>t</mi>
<mo>)</mo>
</mrow>
</mrow>
</msub>
<mi>p</mi>
<mrow>
<mo>(</mo>
<msubsup>
<mi>z</mi>
<mi>k</mi>
<mi>i</mi>
</msubsup>
<mo>(</mo>
<mi>t</mi>
<mo>)</mo>
<mo>|</mo>
<msub>
<mi>y</mi>
<mi>k</mi>
</msub>
<mo>(</mo>
<mi>t</mi>
<mo>)</mo>
<mo>)</mo>
</mrow>
<mo>&CenterDot;</mo>
<mover>
<mi>p</mi>
<mo>^</mo>
</mover>
<mrow>
<mo>(</mo>
<msub>
<mi>y</mi>
<mi>k</mi>
</msub>
<mo>(</mo>
<mi>t</mi>
<mo>)</mo>
<mo>)</mo>
</mrow>
<msub>
<mi>dy</mi>
<mi>k</mi>
</msub>
<mrow>
<mo>(</mo>
<mi>t</mi>
<mo>)</mo>
</mrow>
</mrow>
Wherein,It is likelihood function, zi(t) it is signal strength that node i is received in t moment；
S5. according to the result of calculation of step S4, posterior probability of the target k in node i is calculated：
<mrow>
<mover>
<mi>p</mi>
<mo>~</mo>
</mover>
<mrow>
<mo>(</mo>
<msubsup>
<mi>z</mi>
<mi>k</mi>
<mi>i</mi>
</msubsup>
<mo>(</mo>
<mi>t</mi>
<mo>)</mo>
<mo>)</mo>
</mrow>
<mo>=</mo>
<msub>
<mo>&Integral;</mo>
<msup>
<mi>R</mi>
<mrow>
<mi>N</mi>
<mo>-</mo>
<mn>1</mn>
</mrow>
</msup>
</msub>
<mi>p</mi>
<mrow>
<mo>(</mo>
<msup>
<mi>z</mi>
<mi>i</mi>
</msup>
<mo>|</mo>
<msubsup>
<mi>z</mi>
<mn>1</mn>
<mi>i</mi>
</msubsup>
<mo>,</mo>
<mn>...</mn>
<mo>,</mo>
<msubsup>
<mi>z</mi>
<mi>N</mi>
<mi>i</mi>
</msubsup>
<mo>)</mo>
</mrow>
<mo>&lsqb;</mo>
<munderover>
<mo>&Pi;</mo>
<mrow>
<mi>i</mi>
<mo>=</mo>
<mn>1</mn>
</mrow>
<mi>N</mi>
</munderover>
<mover>
<mi>p</mi>
<mo>^</mo>
</mover>
<mrow>
<mo>(</mo>
<msubsup>
<mi>z</mi>
<mi>k</mi>
<mi>i</mi>
</msubsup>
<mo>(</mo>
<mi>t</mi>
<mo>)</mo>
<mo>)</mo>
</mrow>
<mo>&rsqb;</mo>
<msubsup>
<mi>dz</mi>
<mn>1</mn>
<mi>i</mi>
</msubsup>
<mn>...</mn>
<msubsup>
<mi>dz</mi>
<mrow>
<mi>k</mi>
<mo>-</mo>
<mn>1</mn>
</mrow>
<mi>i</mi>
</msubsup>
<msubsup>
<mi>dz</mi>
<mrow>
<mi>k</mi>
<mo>+</mo>
<mn>1</mn>
</mrow>
<mi>i</mi>
</msubsup>
<mn>...</mn>
<msubsup>
<mi>dz</mi>
<mi>N</mi>
<mi>i</mi>
</msubsup>
</mrow>
Wherein N is the quantity of target；
S6. data correlation is carried out according to step S5, and target following is carried out using Bayesian frame；
S7. the Posterior probability distribution of target is calculated using following formula：
<mrow>
<mover>
<mi>p</mi>
<mo>~</mo>
</mover>
<mrow>
<mo>(</mo>
<msub>
<mi>y</mi>
<mi>k</mi>
</msub>
<mo>(</mo>
<mi>t</mi>
<mo>)</mo>
<mo>)</mo>
</mrow>
<mo>=</mo>
<mi>&alpha;</mi>
<mo>&CenterDot;</mo>
<mover>
<mi>p</mi>
<mo>^</mo>
</mover>
<mrow>
<mo>(</mo>
<msub>
<mi>y</mi>
<mi>k</mi>
</msub>
<mo>(</mo>
<mi>t</mi>
<mo>)</mo>
<mo>)</mo>
</mrow>
<mo>&CenterDot;</mo>
<mo>&Integral;</mo>
<mo>&lsqb;</mo>
<munderover>
<mo>&Pi;</mo>
<mrow>
<mi>i</mi>
<mo>=</mo>
<mn>1</mn>
</mrow>
<mi>m</mi>
</munderover>
<mover>
<mi>p</mi>
<mo>~</mo>
</mover>
<mrow>
<mo>(</mo>
<msubsup>
<mi>z</mi>
<mi>k</mi>
<mi>i</mi>
</msubsup>
<mo>(</mo>
<mi>t</mi>
<mo>)</mo>
<mo>)</mo>
</mrow>
<mo>&CenterDot;</mo>
<mi>p</mi>
<mrow>
<mo>(</mo>
<msubsup>
<mi>z</mi>
<mi>k</mi>
<mi>i</mi>
</msubsup>
<mo>(</mo>
<mi>t</mi>
<mo>)</mo>
<mo>|</mo>
<msub>
<mi>y</mi>
<mi>k</mi>
</msub>
<mo>(</mo>
<mi>t</mi>
<mo>)</mo>
<mo>)</mo>
</mrow>
<mo>&rsqb;</mo>
<msubsup>
<mi>dz</mi>
<mi>k</mi>
<mn>1</mn>
</msubsup>
<mn>...</mn>
<msubsup>
<mi>dz</mi>
<mi>k</mi>
<mi>m</mi>
</msubsup>
</mrow>
Wherein, α is a normalization factor, and the purpose is to willValue be transformed between 0~1；It is target k vectorial in the enhancing of t moment, wherein xk(t) it is position vector,It is
Velocity vector,It is target k in the expectation of the state estimation of t moment, m is the state estimation for participating in target k in t moment
Number of sensors；
S8. step S2~S7 is repeated, completes the tracking of target.
2. the data correlation method according to claim 1 that multiple target tracking is carried out using Sensor Network, it is characterised in that step
Sub-clustering described in rapid S1 carries out sub-clustering for node is pressed distance.
3. the data correlation method according to claim 1 that multiple target tracking is carried out using Sensor Network, it is characterised in that step
Sensor described in rapid S1 is receiving intensity sensor.
4. the data correlation method that multiple target tracking is carried out using Sensor Network according to one of claims 1 to 3, feature
It is the measurement information of the acquisition target described in step S2, to use following formula calculate node i strong in the signal that t moment receives
Degree：
<mrow>
<msup>
<mi>z</mi>
<mi>i</mi>
</msup>
<mrow>
<mo>(</mo>
<mi>t</mi>
<mo>)</mo>
</mrow>
<mo>=</mo>
<munderover>
<mo>&Sigma;</mo>
<mrow>
<mi>k</mi>
<mo>=</mo>
<mn>1</mn>
</mrow>
<mi>N</mi>
</munderover>
<mfrac>
<mrow>
<msub>
<mi>a</mi>
<mi>k</mi>
</msub>
<mrow>
<mo>(</mo>
<mi>t</mi>
<mo>)</mo>
</mrow>
</mrow>
<mrow>
<mo>|</mo>
<mo>|</mo>
<msub>
<mi>x</mi>
<mi>k</mi>
</msub>
<mrow>
<mo>(</mo>
<mi>t</mi>
<mo>)</mo>
</mrow>
<mo>-</mo>
<msub>
<mi>&xi;</mi>
<mi>i</mi>
</msub>
<mo>|</mo>
<mo>|</mo>
</mrow>
</mfrac>
<mo>+</mo>
<msub>
<mi>&upsi;</mi>
<mi>i</mi>
</msub>
</mrow>
Wherein, zi(t) it is signal strength that node i is received in t moment；vi(t) it is ambient noise, N is the quantity of target, ak
(t) it is signal strengths of the target k in t moment, xk(t) it is position vectors of the target k in t moment, ξiIt is the position of sensor node i
Put vector.
5. the data correlation method that multiple target tracking is carried out using Sensor Network according to one of claims 1 to 3, feature
It is the state discrete time equation described in step S3, to calculate state discrete time equation using following formula：
yk(t)=Ayk(t-1)+ωk
Wherein,It is state-transition matrix, ωkIt is the Gaussian noise sequence that average is 0；In same sampling period,
The variation of speed can be expressed asAnd using it as the selection criteria of the value of q, wherein Q is covariance matrixtsFor the sampling interval.
6. the data correlation method that multiple target tracking is carried out using Sensor Network according to one of claims 1 to 3, feature
It is to carry out target following using Bayesian frame described in step S6, to calculate the estimate of target location using following formula：
<mrow>
<mi>p</mi>
<mrow>
<mo>(</mo>
<msub>
<mi>x</mi>
<mi>k</mi>
</msub>
<mo>|</mo>
<msub>
<mi>z</mi>
<mrow>
<mn>1</mn>
<mo>:</mo>
<mi>k</mi>
</mrow>
</msub>
<mo>)</mo>
</mrow>
<mo>=</mo>
<mfrac>
<mrow>
<mi>p</mi>
<mrow>
<mo>(</mo>
<msub>
<mi>z</mi>
<mi>k</mi>
</msub>
<mo>|</mo>
<msub>
<mi>x</mi>
<mi>k</mi>
</msub>
<mo>)</mo>
</mrow>
<mi>p</mi>
<mrow>
<mo>(</mo>
<msub>
<mi>x</mi>
<mi>k</mi>
</msub>
<mo>|</mo>
<msub>
<mi>z</mi>
<mrow>
<mn>1</mn>
<mo>:</mo>
<mi>k</mi>
<mo>-</mo>
<mn>1</mn>
</mrow>
</msub>
<mo>)</mo>
</mrow>
</mrow>
<mrow>
<mi>p</mi>
<mrow>
<mo>(</mo>
<msub>
<mi>z</mi>
<mi>k</mi>
</msub>
<mo>|</mo>
<msub>
<mi>z</mi>
<mrow>
<mn>1</mn>
<mo>:</mo>
<mi>k</mi>
<mo>-</mo>
<mn>1</mn>
</mrow>
</msub>
<mo>)</mo>
</mrow>
</mrow>
</mfrac>
</mrow>
Wherein p (xk|z1:k) represent according to measurement z1:k={ z1,z2,…zkEstimation xkProbability density function；Measure zk=hk
(xk,wk) it is state model xk=fk(xk-1,vk-1) measured value；ωkIt is the Gaussian noise sequence that average is 0；vkIt referred to
Journey noise.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610088231.1A CN105717505B (en) | 2016-02-17 | 2016-02-17 | The data correlation method of multiple target tracking is carried out using Sensor Network |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610088231.1A CN105717505B (en) | 2016-02-17 | 2016-02-17 | The data correlation method of multiple target tracking is carried out using Sensor Network |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN105717505A CN105717505A (en) | 2016-06-29 |
| CN105717505B true CN105717505B (en) | 2018-06-01 |
Family
ID=56156795
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201610088231.1A Active CN105717505B (en) | 2016-02-17 | 2016-02-17 | The data correlation method of multiple target tracking is carried out using Sensor Network |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN105717505B (en) |
Families Citing this family (14)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107271991B (en) * | 2017-05-25 | 2019-05-24 | 北京环境特性研究所 | A kind of optical electrical sensor target correlating method based on state estimation |
| CN108152790A (en) * | 2018-01-05 | 2018-06-12 | 燕山大学 | A kind of non-cooperation multi-target traces projectional technique based on distributed structure/architecture |
| DE102018201570A1 (en) * | 2018-02-01 | 2019-08-01 | Robert Bosch Gmbh | Multiple target object tracking method, apparatus and computer program for performing multiple target object tracking on moving objects |
| CN110609254B (en) * | 2018-06-15 | 2022-09-27 | 富士通株式会社 | Action detection method and device based on wireless signals and electronic equipment |
| CN109214432B (en) * | 2018-08-16 | 2022-02-08 | 上海交通大学 | Multi-sensor multi-target joint detection, tracking and classification method |
| CN110913344B (en) * | 2018-08-27 | 2023-09-01 | 香港科技大学 | Collaborative target tracking system and method |
| CN109212519B (en) * | 2018-08-27 | 2023-04-07 | 西安电子科技大学 | Narrow-band radar target tracking method based on BF-DLSTM |
| CN109996205B (en) * | 2019-04-12 | 2021-12-07 | 成都工业学院 | Sensor data fusion method and device, electronic equipment and storage medium |
| CN111132026B (en) * | 2019-11-25 | 2021-07-30 | 成都工业学院 | Target detection method, device, network system and readable storage medium |
| WO2021102676A1 (en) * | 2019-11-26 | 2021-06-03 | 深圳市大疆创新科技有限公司 | Object state acquisition method, mobile platform and storage medium |
| CN112285698B (en) * | 2020-12-25 | 2021-04-20 | 四川写正智能科技有限公司 | Multi-target tracking device and method based on radar sensor |
| CN113514824B (en) * | 2021-07-06 | 2023-09-08 | 北京信息科技大学 | Multi-target tracking method and device for safety and lightning protection |
| CN114172935A (en) * | 2021-12-08 | 2022-03-11 | 深圳市宏电技术股份有限公司 | Physical examination method and device for Internet of things equipment, Internet of things platform and storage medium |
| CN117495917B (en) * | 2024-01-03 | 2024-03-26 | 山东科技大学 | Multi-target tracking method based on JDE multi-task network model |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103648108A (en) * | 2013-11-29 | 2014-03-19 | 中国人民解放军海军航空工程学院 | Sensor network distributed consistency object state estimation method |
| CN104168648A (en) * | 2014-01-20 | 2014-11-26 | 中国人民解放军海军航空工程学院 | Sensor network multi-target distributed consistency tracking device |
Family Cites Families (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US7698108B2 (en) * | 2006-10-10 | 2010-04-13 | Haney Philip J | Parameterization of non-linear/non-Gaussian data distributions for efficient information sharing in distributed sensor networks |
| US8989442B2 (en) * | 2013-04-12 | 2015-03-24 | Toyota Motor Engineering & Manufacturing North America, Inc. | Robust feature fusion for multi-view object tracking |
-
2016
- 2016-02-17 CN CN201610088231.1A patent/CN105717505B/en active Active
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103648108A (en) * | 2013-11-29 | 2014-03-19 | 中国人民解放军海军航空工程学院 | Sensor network distributed consistency object state estimation method |
| CN104168648A (en) * | 2014-01-20 | 2014-11-26 | 中国人民解放军海军航空工程学院 | Sensor network multi-target distributed consistency tracking device |
Also Published As
| Publication number | Publication date |
|---|---|
| CN105717505A (en) | 2016-06-29 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN105717505B (en) | The data correlation method of multiple target tracking is carried out using Sensor Network | |
| CN102147468B (en) | Bayesian theory-based multi-sensor detecting and tracking combined processing method | |
| Sabek et al. | ACE: An accurate and efficient multi-entity device-free WLAN localization system | |
| Wang et al. | Robust device-free wireless localization based on differential RSS measurements | |
| CN103885057B (en) | Adaptive strain sliding window multi-object tracking method | |
| CN107066806B (en) | Data Association and device | |
| CN107526070A (en) | The multipath fusion multiple target tracking algorithm of sky-wave OTH radar | |
| CN104778358B (en) | The partly overlapping extension method for tracking target in monitored area be present in multisensor | |
| CN107102295A (en) | The multisensor TDOA passive location methods filtered based on GLMB | |
| CN108333569A (en) | A kind of asynchronous multiple sensors fusion multi-object tracking method based on PHD filtering | |
| CN103298156B (en) | Based on the passive multi-target detection tracking of wireless sensor network | |
| CN116128932B (en) | Multi-target tracking method | |
| CN105444763A (en) | IMU indoor positioning method | |
| CN105022055A (en) | IMU indoor positioning method | |
| CN102056192A (en) | WSN (wireless sensor network) intra-network data fusion method based on kernel density estimation and non-parameter belief propagation | |
| CN107436434B (en) | Track starting method based on bidirectional Doppler estimation | |
| Sabek et al. | Multi-entity device-free WLAN localization | |
| CN107064865A (en) | The passive co-located method of polar coordinates Dynamic Programming clustered based on depth | |
| CN111711432B (en) | Target tracking algorithm based on UKF and PF hybrid filtering | |
| CN107544066A (en) | One kind is based on the distributed asynchronous iteration Wave filter merging method of tracking before detection | |
| CN102685772B (en) | Tracking node selection method based on wireless all-around sensor network | |
| Zhao et al. | A novel measurement data classification algorithm based on SVM for tracking closely spaced targets | |
| CN104021285B (en) | A kind of interactive multi-model method for tracking target with optimal motion pattern switching parameter | |
| He et al. | A robust CSI-based Wi-Fi passive sensing method using attention mechanism deep learning | |
| Li et al. | Intelligent fusion of information derived from received signal strength and inertial measurements for indoor wireless localization |
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 | ||
| GR01 | Patent grant |