CN105898859A - Wireless sensing network-based target positioning method - Google Patents

Abstract

The invention discloses a wireless sensing network-based target positioning method used for determining the specific position of a target in real time, wherein the target is provided with a wireless sensing network node. According to the method of the invention, based on information interaction between a node to be positioned and a beacon node, the positional information of the beacon node in the network, per-hop distance and hop counts of the beacon node and the node to be positioned are summarized on the node to be positioned; and the node to be positioned determines the coordinates of the node to be positioned itself by using a DV-Hop algorithm according to the summarized data and acquires more accurate coordinates through performing Aitken cyclic iteration based on the above coordinates. The positioning data obtained after iteration operation is performed have high accuracy and a stable error rate.

Description

Object localization method based on wireless sense network

Technical field

The present invention relates to a kind of object localization method based on wireless sensor network.

Background technology

Wireless sensor network is made up of microsensor node substantial amounts of in being deployed in monitored area, passes through The network system of the self-organizing of the multi-hop that communication is formed, its objective is perception collaboratively, adopts Perceived object information in collection and process network's coverage area, and it is sent to observer.Wireless sensor network Sensor technology, micro electro mechanical system (MEMS) technology, embedding assembly technology, wireless communication technology and distribution are gathered The formula information processing technology is in one, mutual by sensor and the external world, data acquisition, processes, communicates And the function such as management.

In many cases, the node in wireless sensor network is it is to be appreciated that the physical location of self.Do not know The data of road sensor position and perception for a lot of application even without meaning.Obtain sensor node The purposes of the following several respects in position: first, the data that node is collected must with measure in coordinate system Position combine, there is no the data of positional information almost without value；Secondly, the one of sensor network A little systemic-functions, such as network topology control, route based on geography information etc., need positional information, separately Outer known location can optimize the scheduling mechanism on duty during the network operation makes the irregular land wheel of redundant node in network Stop to extend the life-span.Therefore, the location of sensor be properly functioning most basic of network be also most important bar Part.

In the prior art, GPS location is highly developed technology.But in sensor field, for sensing Device configuration GPS is impossible.In addition to cost factor, the also limit of sensor own vol size System, sensor location cannot receive the restriction of satellite-signal.Therefore, existing method typically can be Control centre is previously set the position of sensor, for those positions that cannot set, or often moves Dynamic sensor, then calculate this sensing by the heart in the controlling according to the network topology data of wireless senser The position of device.

Summary of the invention

For solving the sensor orientation problem of wireless sensor network, the present invention devises a kind of localization method: In specific region, random distribution wireless senser that some positions determine and some is undetermined In the network of the wireless senser composition of position, wireless senser to be positioned self initiate Location Request, according to Request obtains the residing geographical position of wireless senser topological relationship calculation self.

Implement and comprise the following steps:

The acquisition of jumping figure relation between node: for all nodes, by and surroundings nodes between information hand over Changing, obtain the jumping figure relation between node, set up the jumping figure relation table of network, jumping figure relation is can be by Unique method that the node contacts being scattered gets up；

The coarse positioning of node: utilize jumping figure relation, uses DV-Hop algorithm to carry out the coarse positioning of node, is using During DV-Hop algorithm, utilize the whole network the most often to jump away from the Average hop distance replacing nearest beaconing nodes, make not Know the estimated distance between node and each beaconing nodes closer to the actual distance between them so that Improve the average positioning precision of algorithm；

Iteration refinement: utilize Aitken iterative formula, set up model, is circulated refinement to the result of coarse positioning, Obtaining the most accurate node coordinate, during refinement, set the condition of two kinds of iteration ends, one is Reach maximum iteration time, iteration ends；Two is to reach required iteration precision, and iteration also terminates, and passes through Iteration, can be in the hope of the coordinate of the unknown node closest to actual value.

It is an advantage of the current invention that to be distributed for heterogeneous networks, adaptability is relatively strong, and registration, error rate are steady Fixed, there is the strongest robustness.

Specifically:

1. the acquisition of jumping figure relation between node:

The acquisition of jumping figure relation can obtain by the following method:

Wireless senser to be positioned sends with broadcast mode to neighbouring wireless senser after receiving positioning instruction Location Request, request can be forwarded by those sensors receiving this request, to such an extent as to the sensor of the whole network is all This request can be obtained.In order to avoid network data storm, positioning request information comprises wireless senser to be positioned Numbering and jumping figure, wherein jumping figure is initially 0.Receive the sensor after this request by following rule analysis at Reason: the numbering in this positioning request information is identical with the numbering of himself, then be left intact, otherwise will Jumping figure in this location information adds 1, then by positioning request information and the content ratio in the information storage apparatus of location Relatively, if location information storage apparatus exists corresponding Location Request sensor number and its jumping figure ratio connects The jumping figure of the positioning request information received is big, then this positioning request information is saved in location information storage apparatus In and forward this positioning request information to neighbouring wireless senser with broadcast mode, be otherwise left intact.

Above-mentioned positioning instruction is the instruction sent by the computer of control centre, it is also possible to be to have on wireless senser to determine Position request input equipment sends.This Location Request input equipment can be installed in the button on wireless senser.

Beaconing nodes receives after this request in addition to forwarding positioning request information to neighbor node, also can broadcast type Sending positional information bag, this location information includes: the numbering of wireless senser itself and transverse and longitudinal coordinate, jumping figure, Wherein jumping figure is initially 0.The node receiving this positional information can forward after adding 1 by this jumping figure.Equally in order to Avoiding network data storm and the jumping figure relation determined between each beaconing nodes, each sensor preserves this position Confidence breath bag, compares, if in the positional information received with the positional information being saved in this locality before forwarding The jumping figure that preserves than this locality of jumping figure little just forward and update preservation, be otherwise ignored as this positional information bag.

The most each beaconing nodes can obtain the positional information of other beaconing nodes in network and self and its Jumping figure relation between his beaconing nodes.Node to be positioned also can receive in network all beaconing nodes with its from The jumping figure value of body and their coordinate.Beaconing nodes, according to this jumping figure relation, utilizes following formula to calculate and often jumps Distance:

$C_i=\underset{i\≠j}{\mathrm{\Σ}}\sqrt{{(x_i-x_j)}^2+{(y_i-y_j)}^2}/\underset{i\≠j}{\mathrm{\Σ}}{\mathrm{hops}}_{\mathrm{ij}}$

Wherein, xi, yi are the position coordinates of self；Other beaconing nodes coordinates of xj, yj；Hops is self And the jumping figure value between other beaconing nodes.

After beaconing nodes obtains every hop distance, the every hop distance of himself is broadcasted in the whole network, equally in order to Avoiding network data storm, the when of broadcasting this information bag, information bag further comprises numbering and the jumping figure of self, The initial value of jumping figure is 0, and information bag can be forwarded by those sensors receiving this information bag, forwards rule such as Under: if the numbering in this information bag is identical with the numbering of himself, then it is left intact, otherwise should Jumping figure in information bag adds 1, then this information is compared, if this information bag is deposited with the content being saved in this locality And its jumping figure bigger than the jumping figure that receives, then this information is saved in this locality and to neighbouring wireless sensing Device forwards with broadcast mode, is otherwise left intact.

Through above-mentioned three kinds of data messages in the broadcast of the whole network, wireless senser the most to be positioned can receive Every hop distance that all beaconing nodes calculate and their coordinate and corresponding jumping figure.

2. the coarse positioning of node

By above-mentioned process, node to be positioned just can obtain all beaconing nodes position coordinateses, every hop distance And the jumping figure value of all beaconing nodes and node to be positioned.Afterwards, DV-Hop method can be passed through, carry out thick Location.After node to be positioned obtains the data of above-mentioned all beaconing nodes, method of least square is utilized to calculate not Know the coordinate of node, can provide equation:

$\left\{\begin{array}{c}{(x_1-x)}^2+{(y_1-y)}^2=d_1^2\\ \·\\ \·\\ \·\\ {(x_n-x)}^2+{(y_n-y)}^2=d_n^2\end{array}\right.$

Wherein n is the number of beaconing nodes, can obtain through conversion:

$\left\{\begin{array}{c}{x}_1^2-x_n^2-2(x_1-x_n)x+y_1^2-y_n^2-2(y_1-y_n)y=\\ d_1^2-d_n^2\\ \·\\ \·\\ \·\\ x_{n-1}^2-x_n^2-2(x_{n-1}-x_n)x+y_{n-1}^2-y_n^2-2(y_{n-1}-y_n)y=\\ d_{n-1}^2-d_n^2\end{array}\right.$

It is represented by equation:

AX=B

Wherein:

$A=\left[\begin{array}{cc}2(x_1-x_n)& 2(y_1-y_n)\\ \·& \·\\ \·& \·\\ \·& \·\\ 2(x_{n-1}-x_n)& 2(y_{n-1}-y_n)\end{array}\right],X=\left[\begin{array}{c}x\\ y\end{array}\right]$

$B=\left[\begin{array}{c}{x}_1^2-x_n^2+y_1^2-y_n^2+d_n^2-d_1^2\\ \·\\ \·\\ \·\\ x_{n-1}^2-x_n^2+y_{n-1}^2-y_n^2+d_n^2-d_{n-1}^2\end{array}\right]$

In above formula, x, y are coordinate to be determined, and xi, yi are for being saved in positioning information device, and each position is true Fixed wireless senser coordinate；D1 ... dn is multiplied with jumping figure the value obtained by every hop distance；

Thus can be obtained by the coordinate of node to be positioned.

In above-mentioned equation, according to traditional method, it is used for determining d1 ... every hop distance of dn, is each beacon joint Point jumping figure.When jumping figure between node increases, the length being originally broken line is inevitably mistakened as making by algorithm Air line distance between two nodes, along with the increase of jumping figure, error can be the biggest.

The preferred the whole network of this method the most often jump away from cc replace nearest beaconing nodes Average hop distance Ci:

Cc=∑ C_i/n

This makes true closer between them of estimated distance between unknown node and each beaconing nodes Distance, so that improve the average positioning precision of algorithm.

3. iteration refinement

For obtained in the previous step as a result, it is possible to set up following model:

$f(x_1,y_1,\&CenterDot;\&CenterDot;\&CenterDot;x_n,y_n)=\underset{l\&le;j\&le;m}{\underset{m<i\&le;n}{\mathrm{\&Sigma;}}}{(\sqrt{{(x_i-x_j)}^2+{(y_i-y_j)}^2}-l_{\mathrm{ij}})}^2$

N is node total number, wherein has m beaconing nodes (m ＜ n), if the communication radius of node is R, and unknown node I and beaconing nodes j distance between a hop neighbor node, i.e. i and j each other is less than R.When setting up model, I Range finding to be introduced, i.e. calculate the distance between two nodes.Node i and one hop neighbor is represented with lij Range finding between node j, represents the estimated distance between node i and j with dij, is i.e. drawn by the coarse positioning stage The value that calculated of result.

According to Aitken iterative formula:

$\left\{\begin{array}{c}{\tilde{x}}_{k+1}=g\left(x_k\right),{\stackrel{\&OverBar;}{x}}_{k+1}=g\left({\tilde{x}}_{k+1}\right)\\ x_{k+1}={\stackrel{\&OverBar;}{x}}_{k+1}-\frac{{({\stackrel{\&OverBar;}{x}}_{k+1}-{\tilde{x}}_{k+1})}^2}{{\stackrel{\&OverBar;}{x}}_{k+1}-2{\tilde{x}}_{k+1}+x_k}\end{array}\right.$

Set up equation

${\tilde{x}}_i=g\left(x_i\right)=\frac{\underset{l\&le;j\&le;m}{\mathrm{\&Sigma;}}(x_j+\frac{l_{\mathrm{ij}}(x_i-x_j)}{\sqrt{{(x_i-x_j)}^2+{(y_i-y_j)}^2}})}{m}$

${\stackrel{\&OverBar;}{x}}_i=g\left({\tilde{x}}_i\right)=\frac{\underset{l\&le;j\&le;m}{\mathrm{\&Sigma;}}(x_j+\frac{l_{\mathrm{ij}}({\tilde{x}}_i-x_j)}{\sqrt{{({\tilde{x}}_i-x_j)}^2+{(y_i-y_j)}^2}})}{m}$

Wherein x_iThe initial value of the iteration for obtaining behind first location.Afterwards, according to the second step of iterative formula, can The updated value of the coordinate obtaining unknown node i after first time iteration is:

$x_i^{\&prime;}={\stackrel{\&OverBar;}{x}}_i-\frac{{({\stackrel{\&OverBar;}{x}}_i-{\tilde{x}}_i)}^2}{{\stackrel{\&OverBar;}{x}}_i-2{\tilde{x}}_i+x_i}$

In like manner, the iterative formula of y-coordinate can be obtained.Setting the condition of two kinds of iteration ends, one is to reach Big iterations, iteration ends；Two is to reach required iteration precision, and iteration also terminates.Through iteration, Can be in the hope of the coordinate of the unknown node closest to actual value.

Claims (8)

1. a wireless sensor network node locating method, including being randomly distributed in certain area, by one Wireless senser that determined number position determines and the network of some wireless senser compositions to be positioned, its It is characterised by: its described localization method comprises the steps of

Wireless senser the most to be positioned, after receiving Location Request instruction, sends to other wireless sensers Positioning request information；

The most each wireless senser responds process to Location Request, will letter by the agreement determined each other The jumping figure relation of mark node is aggregated on wireless senser to be positioned；

The jumping figure relation that wireless senser the most to be positioned obtains according to step 2 determines location.

Wireless sensor network node locating method the most according to claim 1, it is characterised in that institute The acquisition of the node jumping figure relation stating step 2 uses distance vector exchange agreement.

Wireless sensor network node locating method the most according to claim 1, it is characterised in that institute The algorithm stating the determination location in step 3 uses DV-Hop algorithm.

Wireless sensor network node locating method the most according to claim 3, it is characterised in that institute State often jumping away from being the whole network beaconing nodes value of often jumping that anomaly obtains the most afterwards in algorithm.

5. according to the wireless sensor network node locating method described in any of the above-described claim, its feature It is, the result of obtained location is circulated iteration refinement.

Wireless sensor network node locating method the most according to claim 5, is characterised by, circulation The algorithm of iteration uses Aitken algorithm.

Wireless sensor network node locating method the most according to claim 5, is characterised by, circulation The number of times of iteration is by being previously set.

Wireless sensor network node locating method the most according to claim 5, is characterised by, circulation The number of times of iteration is determined by the precision obtaining result.