CN101221235B - Wireless sensor network location refining method based on hop count - Google Patents

Abstract

The invention provides a refinement method of a wireless sensor network location which is based on hops. After the initial location of an unknown node in a network is implemented, the method adopts a multi-hop extension algorithm and the least square method of relative weights for implementing the refinement computation of an initial estimation position and acquiring the coordinate of a final estimation position of the unknown node. The refinement method implements the refinement of the unknown node in a multi-hop range, thus increasing the redundancy of position computation and the number of valid reference points in the refinement, and improving the precision of the location refinement of the unknown node. With respect to the least square method of relative weights, the method assigns different weights to the reference nodes with different setting accuracies, thus suppressing the accumulation of distance measurement error and node location error in the refinement computation, and effectively improving the location precision of the unknown node. The refinement method can be taken as a subsequent step of the location methods which are based on distances or unrelated to distances, thus achieving the goal of improving location precision.

Description

A kind of wireless sensor network location refinement method based on jumping figure

(1) technical field:

The present invention relates to a kind of wireless sensor network location refinement method, belong to the location refinement method of radio sensing network node based on jumping figure.

(2) technical background:

Development along with sensor technology, MEMS (micro electro mechanical system) (MEMS) technology, embedded computing technique, wireless communication technology and distributed information processing, make the Multifunction Sensor of development low cost, low-power consumption, small size short distance communication become possibility, wireless sensor network occurs on these technology.Wireless sensor network is a kind of special Ad-hoc network, can be applicable to connect up and power supply is supplied with the zone of difficulty, the zone that personnel can not arrive (as be subjected to polluting, environment can not destroyed or hostile area) and some interim occasions (as the generation disaster time, fixed communication network is destroyed) etc.Characteristics such as it does not need the fixed network support, has rapid deployment, and survivability is strong can be widely used in fields such as military affairs, industry, traffic, environmental protection, have caused people's extensive concern.

Wireless sensor network exemplary operation mode is as follows: use aircraft that a large amount of sensor nodes (quantity from hundreds of to several thousand) are shed area-of-interest, node is by self-organization wireless network of formation fast.Node is the collection and the person of sending of information, also serves as the route person of information, and the data of collection arrive gateway by the multi-hop route.Gateway (being also referred to as Sink node) is a special node, can communicate by letter with Surveillance center by Internet, mobile communications network, satellite etc.; Also can utilize unmanned plane to leap the network sky, by the gateway image data.

The sensor network nodes self poisoning is as the critical support technology of wireless sensor network, and is all significant to the application and the location-based procotol research of wireless sensor network, do not have the supervisory messages of positional information normally skimble-skamble.For example during target following, sensor detects the simulation table of being followed the tracks of, and need calculate the position of target by the position of sensor self; Fire alarm needs place on fire accurately; It is which subregion is found enemy's invasion that the battlefield monitoring then need be provided to the end.On the other hand,, transmit, have good expandability, but that precondition is the coordinate of requirement node is known as geographical route implementation data under the situation of routing table for research based on node location.

Owing to be subjected to the restriction of problems such as cost, power consumption, extendability, in big-and-middle-sized wireless sensor network, often have only minority node configuration gps receiver, and can not manually dispose each node.Therefore, must adopt certain mechanism and algorithm to solve the orientation problem of node.

At present according to the location mechanism difference, sensor network locating method be divided into based on the localization method of distance and with the localization method of range-independence.

Calculate node location based on the localization method of distance distance or the angle information by point-to-point between measured node, the precision of this method is higher, but the hardware facility to network is also had relatively high expectations, and the location based on ranging technology needs repeatedly side amount, circulation refinement usually, when obtaining relative accurate localization result, can produce a large amount of calculating and communication overhead, so,, be not suitable for low-power consumption, application cheaply though this method is higher in bearing accuracy.Need not distance and angle information with the localization method of range-independence, only can realize the location of node, reduced requirement, increase to some extent but positioning error is also corresponding to node hardware according to information such as network connectivties.

Generally include distance (or angle) measurement, coordinate Calculation and the refinement three phases that optionally circulates for existing localization method in the wireless sensor network.

Phase one: range observation

Mainly contain based on reaching time (TOA, be time of arrival), based on poor (TDOA time of arrival, be time difference of arrival), arrive (AOA based on angle, be angle of arrival), based on received signal intensity indication (RSSI, i.e. received signal strength indicator) etc.

Subordinate phase: coordinate Calculation

The coordinate Calculation method of node mainly contains the trilateration and the maximum-likelihood method estimation technique.

Phase III: circulation refinement

According to the locator meams of unknown node, the circulation refinement can be divided into two kinds.A kind of is at first to utilize related positioning method to obtain unknown node initial alignment result, then with its neighbor node as the reference node, recomputate the position of unknown node, and the result that will meet qualifications is as the new location estimation of unknown node, enter cyclic process next time, up to satisfying the condition that circulation stops.Another kind of it upgraded to beaconing nodes, and enter circulation next time, location end after all unknown node that satisfy location condition are determined the position for after the part unknown node determines self-position.

In the circulation refinement process oriented unknown node is incorporated in the location Calculation of other unknown node, solved the problem of anchor node lazy weight to a certain extent, but, cause error in cyclic process, to accumulate because all there are error in the location of node itself and the measurement of euclidean distance between node pair.

(3) summary of the invention:

The purpose of this invention is to provide a kind of according to the wireless sensor network location refinement method of the message exchange between network node based on jumping figure, the accumulation that the positioning error and the euclidean distance between node pair of node itself measured (or calculating) error in the refinement that reduces to circulate improves bearing accuracy.

A kind of wireless sensor network location refinement method based on jumping figure of the present invention is applicable to self-organizing sensor network network system, can be used as the follow-up phase of general localization method (as the DV-Hop algorithm), to improve the bearing accuracy of node.In the loop iteration process, exchange positional information each other between neighbor node adopts the new position coordinates of relative weighted least-squares method estimation node.

A kind of wireless sensor network location refinement method based on jumping figure, realize on the basis of the unknown node acquisition initial estimated location coordinate in network, the neighbor node of certain limit is cooperated mutually, realize the further refinement calculating of node estimated position, improve locating accuracy, this method comprises the steps:

Step 1: each unknown node is obtained the initial estimated location coordinate in the network.This step is the basis that the refinement method realizes, can adopt based on distance or with the localization method of range-independence, obtain the initial estimated location coordinate of unknown node.

Step 2: the initial weighted value of node in the setting network.

Step 3: set refinement jumping figure (being threshold value TTL), suppose TTL=N, N＞1.

Step 4: unknown node is obtained the reference mode estimated position information of specifying in the jumping figure N by the message exchange between neighbor node;

Step 5: according to the information of the reference mode that receives, unknown node is calculated the relative weighted value of each reference mode;

Step 6: unknown node will with the distance of reference mode as constraint condition, set up the overdetermined equation group, adopt relative weighted least-squares method, obtain the estimated position coordinate of node;

Unknown node is set up the overdetermined equation group according to the position and the range information of each reference mode of gained, adopt Taylor expansion that Nonlinear System of Equations is converted into system of linear equations, adopt relative weighted least-squares method that system of equations is found the solution, obtain the estimated position coordinate of node.

Step 7: whether the estimated position coordinate of decision node meets the demands, and is then to preserve, otherwise abandons.Judging whether to satisfy the condition that refinement stops, is then to stop refinement, otherwise enters next round-robin refinement.

After unknown node obtains new estimated position coordinate, whether judging difference that new estimated position coordinate compares with former coordinate less than 1% of node communication radius, is that the refinement that then abandons new estimated position coordinate and stop this node is calculated; Otherwise the estimated position coordinate of new node more, and with the mean value of the relative weighted value of reference mode as the new weighted value of unknown node, judge whether to reach the refinement number of times of qualification then, be then to stop refinement, otherwise enter the circulation of refinement next time.

Wherein, in the described step 3, on behalf of node, the refinement jumping figure carry out the scope that reference mode calculate is collected in refinement, i.e. the maximum hop count propagated in network of packet.

Wherein, in the described step 4, unknown node is obtained the reference mode estimated position information of specifying in the refinement jumping figure by the message exchange between neighbor node, is specially: node comprises its positional information and is initially 0 counter information packet to its neighbor node broadcasting in the network; The node that receives packet with the value of counter add 1 with N relatively, then abandon this packets of information greater than N; Then travel through the reference mode tabulation of unknown node smaller or equal to N, search the ID that whether has the node that receives, otherwise the information of the node that receives added in the reference mode tabulation and with packet transmit, whether the value that is the then counter in the comparing data bag is less than the value of the counter of the reference mode of preserving in the reference mode tabulation, be the value of then upgrading the corresponding counter in the tabulation, otherwise abandon this packet.

In the circulation refinement process, unknown node receives the information of new reference mode, for initiate reference mode, its information is added in the tabulation, if not newly added node, then changes the information of corresponding reference mode.

Wherein, in the described step 5, according to the information of the reference mode that receives, unknown node is calculated the relative weighted value of each reference mode, is specially:

Wherein, in the described step 6, unknown node will with the distance of reference mode as constraint condition, set up the overdetermined equation group, adopt relative weighted least-squares method, obtain the estimated position coordinate of node, be specially:

Unknown node is calculated the distance between unknown node and reference mode according to the information of the reference mode that receives, and is divided into following two kinds according to initial alignment mechanism difference:

(1) node in the network has distance measurement function

Distance between unknown node and reference mode equal each distance of jumping section and;

(2) node in the network does not have distance measurement function

Distance between unknown node and reference mode equals the distance of jumping figure and average every jumping.

Unknown node is set up the overdetermined equation group according to the position and the range information of each reference mode of gained, adopt Taylor expansion that Nonlinear System of Equations is converted into system of linear equations, adopt relative weighted least-squares method that system of equations is found the solution, obtain the estimated position coordinate of node.

Wherein, in the described step 7, be specially: after unknown node obtains new estimated position coordinate, whether judging difference that new estimated position coordinate compares with former coordinate less than 1% of node communication radius, is that the refinement that then abandons new estimated position coordinate and stop this node is calculated; Otherwise the estimated position coordinate of new node more,, and with the mean value of the relative weighted value of reference mode as the new weighted value of unknown node, judge whether to reach the refinement number of times of qualification then, be then to stop refinement, otherwise enter the circulation of refinement next time.

The wireless sensor network location refinement method based on jumping figure that the present invention mentioned promptly is applicable to the localization method based on distance, also is applicable to the localization method with range-independence.

A kind of wireless sensor network location refinement method of the present invention based on jumping figure, its advantage and effect are:

(1) at refinement stage, unknown node adopts the information of the reference mode in the multi-hop scope, upgrades the estimated position of self.Avoid the problem of tradition effective reference mode deficiency of refinement in the single-hop scope, improved the coverage rate of refinement.

(2) adopt relative weighted value to weigh the precision of reference mode with respect to unknown node, relative weighted least-squares method is used for the new position coordinates of solution node, reduces the accumulation that the positioning error and the euclidean distance between node pair of reference mode itself are measured (or calculating) error.

(4) description of drawings:

Fig. 1 is the wireless sensor network location refinement method flow diagram that the present invention is based on jumping figure;

Fig. 2 obtains the process flow diagram of specifying the reference mode estimated position information in the jumping figure for node;

(5) embodiment:

Below in conjunction with the drawings and specific embodiments, further describe technical scheme of the present invention.

A kind of wireless sensor network location refinement method of the present invention based on jumping figure, please in conjunction with Figure 1 and Figure 2, its concrete steps are as follows:

Among Fig. 1 101: the unknown node in the network utilizes localization method to calculate the initial estimated location coordinate.

This step is the basis that the refinement method realizes, can adopt based on distance or with the localization method of range-independence.

In wireless sensor network, be meant that based on the localization method of distance actual range or orientation by measuring between adjacent node position.The method that adopts according to measured node distance or orientation can be divided into: based on the location of TOA, based on the location of TDOA, based on the location of AOA with based on the location of RSSI etc.And the localization method of range-independence is meant the absolute distance that need not between measured node, utilizes the position of internodal estimated distance computing node.Localization method typical and range-independence has interior point (APIT, the i.e. approximatepoint-in-triangulation test) algorithm of centroid algorithm, distance vector (DV-Hop, i.e. distance vector-hop) algorithm and triangle etc.

Among Fig. 1 102: the initial weighted value of node in the setting network.

Node in the network is divided into anchor node and unknown node:

(1) weighted value of anchor node is 1.0

(2) since the positioning error of unknown node mainly by the decision of the range error between unknown node and anchor node, so the weighted value of unknown node is determined with distance error between relevant beaconing nodes by unknown node.

Suppose that anchor node arrives in the path of unknown node, the neighbor node number of unknown node is N; The anchor node number that unknown node receives is n; Beaconing nodes is hop to the jumping figure of unknown node ₁, hop ₂..., hop _nThe initial degree of confidence w of unknown node then ₀Computing method as shown in the formula:

$w_0=\left\{\begin{array}{cc}0.01& N<3\\ \frac{\underset{i=1}{\overset{n}{\mathrm{\&Sigma;}}}\sqrt{{\mathrm{hop}}_i}}{N*n}& N\&GreaterEqual;3\end{array}\right.---\left(1\right)$

Among Fig. 1 103: set the jumping figure N of refinement, promptly node is to carry out refinement in the N jumping scope as starting point.

Among Fig. 1 104: unknown node is obtained the reference mode estimated position information of specifying in the jumping figure N.Detailed process is as shown in Figure 2:

Step 1: node comprises its positional information and is initially 0 counter information packet to its neighbor node broadcasting in the network;

Step 2: the node that receives packet with the value of counter add 1 and with N relatively;

Step 3: if greater than N then abandon this packets of information;

Step 4:, search the ID that whether has the node that receives: transmit otherwise the information of the node that receives added in the reference mode tabulation and with packet if smaller or equal to N then travel through the reference mode tabulation of unknown node; Whether the value that is the then counter in the comparing data bag is the value of then upgrading the corresponding counter in the tabulation, otherwise abandons this packet less than the value of the counter of the reference mode of preserving in the reference mode tabulation.

If in the circulation refinement process, unknown node receives the information of new reference mode, for initiate reference mode, its information is added in the tabulation; If not initiate reference mode, then change the information of corresponding reference mode.

Among Fig. 1 105: according to the information of the reference mode that receives, calculate reference mode relatively with the relative weighted value of unknown node;

Weighted value is determined jointly by the bearing accuracy of node itself and the precision of internodal range observation relatively.Here suppose that nodes X receives the information of n reference mode S, reference mode S _iBearing accuracy be δ _i, and the precision of the range observation between nodes X is μ _i, reference mode S so _iDegree of confidence with respect to nodes X is

$w_i=\frac{1}{\sqrt{{\mathrm{\&delta;}}_i^2+{\mathrm{\&mu;}}_i^2}}.$

Among Fig. 1 106: according to the distance restraint condition, set up system of equations, adopt the estimated position coordinate of relative weighted least-squares method computing node;

Step 1: according to distance restraint, set up system of equations, different according to the mechanism of initial alignment method, be divided into following two kinds of situations:

1. the node in the network has distance measurement function

The coordinate of supposing unknown node is for (x, y), the coordinate of its reference mode is (Rx _i, Ry _i), unknown node to each reference mode through the distance of each section in path be R _i, then

$\left\{\begin{array}{c}\sqrt{{(x-{\mathrm{Rx}}_1)}^2+{(y-{\mathrm{Ry}}_1)}^2}=R_1\\ \&CenterDot;\\ \&CenterDot;\\ \&CenterDot;\\ \sqrt{{(x-{\mathrm{Rx}}_n)}^2+{(y-{\mathrm{Ry}}_n)}^2}=R_n\end{array}\right.---\left(2\right)$

2. the node in the network does not have distance measurement function

The coordinate of supposing unknown node is for (x, y), the coordinate of its reference mode is (Rx _i, Ry _i), unknown node is hop to the jumping figure of each reference mode _i, average every distance is r in the network, then

$\left\{\begin{array}{c}\sqrt{{(x-{\mathrm{Rx}}_1)}^2+{(y-{\mathrm{Ry}}_1)}^2}={\mathrm{hop}}_1*r\\ \&CenterDot;\\ \&CenterDot;\\ \&CenterDot;\\ \sqrt{{(x-{\mathrm{Rx}}_n)}^2+{(y-{\mathrm{Ry}}_n)}^2}={\mathrm{hop}}_n*r\end{array}\right.---\left(3\right)$

Step 2: adopt Taylor expansion that Nonlinear System of Equations is transformed the system of linear equations order $f(x,y)=\sqrt{{(x-x_i)}^2+{(y-y_i)}^2}---\left(4\right)$

To (3) at (x ₀, y ₀) point carries out Taylor and minute solves:

$f(x,y)=f(x_0+h,y_0+k)$

$=\sqrt{{(x_0-x_i)}^2+{(y_0-y_i)}^2}$

$+\frac{(x_0-x_i)}{\sqrt{{(x_0-x_i)}^2+{(y_0-y_i)}^2}}h+\frac{(y_0-y_i)}{\sqrt{{(x_0-x_i)}^2+{(y_0-y_i)}^2}}k---\left(5\right)$

For formula (2), (3) can be converted into

$\left\{\begin{array}{c}\frac{(x_0-{\mathrm{Rx}}_1)}{\sqrt{{(x_0-{\mathrm{Rx}}_1)}^2+{(y_0-{\mathrm{Ry}}_1)}^2}}h+\frac{(y_0-{\mathrm{Ry}}_1)}{\sqrt{{(x_0-R{x}_1)}^2+{(y_0-{\mathrm{Ry}}_1)}^2}}k=d_1-\sqrt{{(x_0-{\mathrm{Rx}}_1)}^2+{(y_0-{\mathrm{Ry}}_1)}^2}\\ \frac{(x_0-{\mathrm{Rx}}_2)}{\sqrt{{(x_0-{\mathrm{Rx}}_2)}^2+{(y_0-{\mathrm{Ry}}_2)}^2}}h+\frac{(y_0-{\mathrm{Ry}}_2)}{\sqrt{{(x_0-{\mathrm{Rx}}_2)}^2+{(y_0-{\mathrm{Ry}}_2)}^2}}k=d_2-\sqrt{{(x_0-{\mathrm{Rx}}_2)}^2+{(y_0-{\mathrm{Ry}}_2)}^2}\\ M\\ \frac{(x_0-{\mathrm{Rx}}_n)}{\sqrt{{(x_0-{\mathrm{Rx}}_n)}^2+{(y_0-{\mathrm{Ry}}_n)}^2}}h+\frac{(y_0-{\mathrm{Ry}}_n)}{\sqrt{{(x_0-{\mathrm{Rx}}_n)}^2+{(y_0-{\mathrm{Ry}}_n)}^2}}k=d_n-\sqrt{{(x_0-{\mathrm{Rx}}_n)}^2+{(y_0-{\mathrm{Ry}}_n)}^2}\end{array}-\left(6\right)\right.$

(x wherein ₀, y ₀), be the estimated position coordinate of unknown node, d _iBe the estimated distance between unknown node and reference mode, have the situation d of distance measurement function for the node in the network _i=R _i, do not have the situation d of distance measurement function for the node in the network _i=hop _i* r.

Formula (6) is expressed as with matrix

AΔX＝B (7)

Wherein

$A=\left[\begin{array}{cc}\frac{(x_0-{\mathrm{Rx}}_1)}{\sqrt{{(x_0-{\mathrm{Rx}}_1)}^2+{(y_0-{\mathrm{Ry}}_1)}^2}}& \frac{(y_0-{\mathrm{Ry}}_1)}{\sqrt{{(x_0-{\mathrm{Rx}}_1)}^2+{(y_0-{\mathrm{Ry}}_1)}^2}}\\ M& M\\ \frac{(x_0-{\mathrm{Rx}}_n)}{\sqrt{{(x_0-{\mathrm{Rx}}_n)}^2+{(y_0-{\mathrm{Ry}}_n)}^2}}& \frac{(y_0-{\mathrm{Ry}}_n)}{\sqrt{{(x_0-{\mathrm{Rx}}_n)}^2+{(y_0-{\mathrm{Ry}}_n)}^2}}\end{array}\right]$

$B=\left[\begin{array}{c}{d}_1-\sqrt{{(x_0-{\mathrm{Rx}}_1)}^2+{(y_0-{\mathrm{Ry}}_1)}^2}\\ M\\ d_n-\sqrt{{(x_0-{\mathrm{Rx}}_n)}^2+{(y_0-{\mathrm{Ry}}_n)}^2}\end{array}\right]$

$\mathrm{\&Delta;X}=\left[\begin{array}{c}h\\ k\end{array}\right]$

Step 3: adopt relative weighted least-squares method solving equation group

Hypothetical reference node (Rx _i, Ry _i) (x, relative weighted value y) is w with respect to unknown node _Ri, then weighting matrix is relatively:

To can get in the relative weighting matrix substitution formula (7)

WAΔX＝WB (8)

(8) are found the solution

ΔX＝(A ^TW ^TWA) ^-1A ^TW ^TWB

Among Fig. 1 107: judge whether formula (9) is set up

$\sqrt{h^2+k^2}<{\mathrm{\&epsiv;}}_{\mathrm{threshold}}---\left(9\right)$

If set up the node stop refinement, otherwise enter next Rule of judgment.

Among Fig. 1 108: whether decision node reaches predefined refinement number of times, is then to stop refinement, otherwise unknown node enters the refinement process that next time circulates to the new estimated position coordinate of its neighbor node broadcasting.

Among Fig. 1 109: unknown node will $(x_0+\frac{h}{2},y_0+\frac{k}{2})$ As new estimated position coordinate, and with the relative weighted value of its reference mode as the new weighted value of node, promptly $w_0=\frac{\underset{i=1}{\overset{n}{\mathrm{\&Sigma;}}}{w}_{\mathrm{Ri}}}{n}.$

In sum, the present invention proposes a kind of location refinement method based on wireless sensor network, unknown node is being obtained on the basis of initial estimated location, utilizes the information of the reference mode in the multi-hop scope to carry out refinement calculating with relative weighted least-squares method, to improve the precision of node locating.This method improves the redundance of joint location Calculation, improved the coverage rate of refinement node, weighted least-squares method is used and has been effectively reduced the accumulation that node locating sum of errors euclidean distance between node pair is measured (or calculating) error in the refinement relatively, has improved the bearing accuracy of node.

It should be noted last that; above embodiment is only unrestricted in order to technical scheme of the present invention to be described; although the present invention is had been described in detail with reference to preferred embodiment; those of ordinary skill in the art is to be understood that; can make amendment or be equal to replacement technical scheme of the present invention; and do not break away from the spirit and scope of technical solution of the present invention, still play and ask within the protection domain in the described right of present patent application.

Claims (6)

1. the wireless sensor network based on jumping figure is located the refinement method, and it is characterized in that: this method comprises the steps:

Step 1: each unknown node is obtained the initial estimated location coordinate in the network;

Step 2: the initial weighted value of node in the setting network;

Step 3: setting the refinement jumping figure is N, N＞1;

Step 4: unknown node is obtained the reference mode estimated position information of specifying in the refinement jumping figure by the message exchange between neighbor node;

Step 5: according to the information of the reference mode that receives, unknown node is calculated the relative weighted value of each reference mode;

Step 6: unknown node will with the distance of reference mode as constraint condition, set up the overdetermined equation group, adopt relative weighted least-squares method, obtain the estimated position coordinate of node;

Step 7: whether the estimated position coordinate of at first judging unknown node meets the demands, and is then to preserve and stop refinement; Otherwise abandoning, judge whether then to satisfy the condition that refinement stops, is then to stop refinement, otherwise enters next round-robin refinement.

2. a kind of wireless sensor network location refinement method according to claim 1 based on jumping figure, it is characterized in that: on behalf of node, the refinement jumping figure in this step 3 carry out the scope that reference mode calculate is collected in refinement, i.e. the maximum hop count propagated in network of packet.

3. a kind of wireless sensor network location refinement method according to claim 1 based on jumping figure, it is characterized in that: this step 4 is specially: node comprises its positional information and is initially 0 counter information packet to its neighbor node broadcasting in the network; The node that receives packet with the value of counter add 1 with N relatively, then abandon this packets of information greater than N; Then travel through the reference mode tabulation of unknown node smaller or equal to N, search the ID that whether has the node that receives, otherwise the information of the node that receives added in the reference mode tabulation and with packet transmit, whether the value that is the then counter in the comparing data bag is less than the value of the counter of the reference mode of preserving in the reference mode tabulation, be the value of then upgrading the corresponding counter in the tabulation, otherwise abandon this packet.

4. a kind of wireless sensor network location refinement method according to claim 1 based on jumping figure, it is characterized in that: this step 5 is specially:

5. a kind of wireless sensor network location refinement method according to claim 1 based on jumping figure, it is characterized in that: this step 6 is specially:

Unknown node is calculated the distance between unknown node and reference mode according to the information of the reference mode that receives, and is divided into following two kinds according to initial alignment mechanism difference:

(1) node in the network has distance measurement function

Distance between unknown node and reference mode equal each distance of jumping section and;

(2) node in the network does not have distance measurement function

Distance between unknown node and reference mode equals the product of jumping figure and average every hop distance;

Unknown node is set up the overdetermined equation group according to the position and the range information of each reference mode of gained, adopt Taylor expansion that Nonlinear System of Equations is converted into system of linear equations, adopt relative weighted least-squares method that system of equations is found the solution, obtain the estimated position coordinate of node.

6. a kind of wireless sensor network location refinement method according to claim 1 based on jumping figure, it is characterized in that: this step 7 is specially:

After unknown node obtains new estimated position coordinate, whether judging difference that new estimated position coordinate compares with former coordinate less than 1% of node communication radius, is that the refinement that then abandons new estimated position coordinate and stop this node is calculated; Otherwise the estimated position coordinate of new node more, and with the mean value of the relative weighted value of reference mode as the new weighted value of unknown node, judge whether to reach the refinement number of times of qualification then, be then to stop refinement, otherwise enter the circulation of refinement next time.

Images