CN101742642B - Area division and coordinate welting-based wireless sensor network semiautomatic node positioning method - Google Patents

Abstract

The invention discloses an area division and coordinate welting-based semiautomatic node positioning method, which comprises the following steps: 1) parameter configuration, namely dividing a rectangular area for a target network according to a total topological graph of the network, determining an area where each beacon node is positioned, setting an area node density rho, node communication radius r, correction factor m, welting threshold value k and beacon node coordinates, and calculating the average size per hop of the network; and 2) a positioning process, (2.1) calculating minimum hop numbers of an unknown node and the beacon node, (2.2) calculating the hop size between the unknown node and the beacon node, (2.3) calculating the initial coordinate value of the unknown node, (2.4) welting the coordinates, and acquiring node coordinate correction values serving as final position coordinates of the unknown node after the welting is finished. The method has the advantages of high positioning precision and strong practicability, and can be applied to the practical positioning environment.

Description

Wireless sensor network semiautomatic node positioning method based on area dividing and coordinate welt

Technical field

The invention belongs to the radio sensing network field, especially a kind of wireless sensor network node positioning method.

Background technology

Development along with sensor technology, embedded technology, distributed information processing and wireless communication technique; Radio sensing network (Wireless Sensor Network with a large amount of microsensor node compositions with microprocessing capability; WSN) become the research focus of academia gradually, also become industrial quarters and expand the focus of using.

The key issue that much needs to be resolved hurrily and study is arranged in the WSN field, and the node locating problem is exactly one of them.Node locating refers to the sensor node in the radio sensing network, according to the node of minority known location, gets access to from the physical location in network through carrying out someway or move certain algorithm.According to node known self-position whether, node can be divided into beaconing nodes (Anchor) and unknown node.Beaconing nodes is through the exact position of means acquisitions self such as carrying the GPS positioning equipment or manually write, and the ratio that in network node, accounts for is very little; The node of other non-beacon is exactly a unknown node, and they confirm self-position through beaconing nodes.Because in radio sensing network, the node distribution space is wide, and position randomness is very big, and the collected information of network has very big position dependence, so the positioning oneself to use for WSN significance is arranged of node.

Up to the present, though existing a lot of scholar proposes relatively more typical WSN node locating algorithm, obtained certain achievement, still had many weak points, this mainly shows following aspects:

(1). rest on theory stage mostly, practical application is seldom.This is a present topmost weak point to the research of wireless sensor network node orientation problem.Now there have been a large amount of node positioning methods and location algorithm to be suggested, the case in certain actual system has been realized and be successfully applied to these algorithms yet seldom have.To these Locating Algorithm and performance evaluation etc., majority all is to adopt ways such as emulation and simple program realization, does not have practical significance.

(2). positioning accuracy is not high.Though a lot of wireless sensor network node positioning methods and algorithm have been arranged at present, the positioning accuracy of these methods in practical application is also not really high, even use special distance-measuring equipment to obtain the position, its error is also relatively large.This is because the problem of localization method or algorithm itself on the one hand, more these localization methods of chief reason combine with practical application not tight, less than position fixing process not being made optimization according to different WSN application characteristics separately.

(3). combine not enough with other problem.Orientation problem is not an isolated problem; In the positioning accuracy of considering how to improve node, location efficiency, locating accuracy; Problems such as the energy consumption growth that also will consider thus to be brought, information security, and existing Position Research is few for the consideration of this respect.

Indoor emergency escape guiding (the IndoorEmergency-escape-guide System of system based on radio sensing network; IEEGS); Its basic thought is: the wireless sensing node of in the building of a planar structure, arranging some; Form radio sensing network, environmental parameters such as the temperature around these node monitorings, illumination take place to have judged whether fire.In case the data identification that comes according to collection when some node has fire to take place, these nodes will get into abnormality, and broadcast to whole network.The node of receiving abnormality is according to fire location, self-position, exit position; Calculate a safe escape direction in real time; And using Warning Mark that this direction is pointed out, the trapped personnel in the building can be escaped according to these Way outs to nearby that points to step by step.There is very big dependence in the IEEGS system to the positional information of node in the network, and therefore the same with most WSN application, IEEGS need solve orientation problem.Except general WSN node locating demand, the IEEGS system also has the characteristics of himself aspect node locating:

(1). require certain precision.In the IEEGS system, the fire place needs to judge according to the position of the node that sends out an emergency signal, if the location of these nodes is not accurate enough, not only influences the calculating of best-effort path, and can influence rescuer's correct judgement, even cause serious consequence.This node locating algorithm that just requires to run among the IEEGS can reach certain precision, if needed, can sacrifice some and carry out efficient and energy consumption.

(2). node distributes and presents region characteristic, local edge.The IEEGS system can be deployed in the building of plane usually, sensor node or be placed in the aisle and the wall in room on, or be suspended on the ceiling, its distributing position mostly is in the edge of a rectangular area.

Summary of the invention

Low for the positioning accuracy that overcomes the existing wireless sensor network locating method, can't be applied to actual environment, the relatively poor deficiency of practicality, the present invention provide a kind of positioning accuracy high, can be suitable for actual location environment, practical wireless sensor network semiautomatic node positioning method based on area dividing and coordinate welt.

The technical solution adopted for the present invention to solve the technical problems is:

A kind of wireless sensor network semiautomatic node positioning method based on area dividing and coordinate welt, said semiautomatic node positioning method may further comprise the steps:

1), parameter configuration:

According to total topological diagram of network, objective network to be carried out the rectangular area divide, and confirm each beaconing nodes zone of living in, the area coordinate parameter after the division is (x ₁, y ₁)-(x ₂, y ₂), (x wherein ₁, y ₁) be regional left upper apex coordinate, (x ₂, y ₂) be regional bottom right apex coordinate;

And regions node density ρ, node communication radius, modifying factor m, welt threshold value k and beaconing nodes coordinate;

The span of modifying factor m be (0,5], beaconing nodes is a reference point, its coordinate is a known quantity;

According to Area Node density p, node communication radius, modifying factor m, the average every hop distance of computing network specifically has:

$\mathrm{ASPH}=\left\{\begin{array}{cc}\frac{2m}{\sqrt{\mathrm{\ρ}}-\frac{1}{r}},& \sqrt{\mathrm{\ρ}}>\frac{i}{r}\\ \mathrm{mr},& \sqrt{\mathrm{\ρ}}\≤\frac{1}{r}\end{array}\right.---\left(6\right)$

Wherein, ASPH is the average every hop distance of network;

2), position fixing process, specifically have:

(2.1) minimum hop count of calculating unknown node and beaconing nodes;

(2.2) calculate the jumping figure distance of unknown node and beaconing nodes: unknown node utilizes computes to arrive the geometric distance of these beaconing nodes after receiving the jumping figure information with beaconing nodes:

d _i＝ASPH×hop _i

d _iBe the distance of node to beaconing nodes i, hop _iBe the jumping figure of node to beaconing nodes i;

(2.3) calculate unknown node coordinate initial value: when a unknown node obtain with 3 or more internodal jumpings of multi-beacon apart from after, use trilateration calculating self-position, obtain the initial value of its position coordinates

(2.4) coordinate welt: according to coordinate, the welt threshold value k in unknown node self coordinate initial value, zone of living in; Carry out coordinate welt process:, then just use the actual coordinate of this coordinate as unknown node if this coordinate arrives the distance at regional 4 edges up and down all greater than k; If the X value of this coordinate, is then used apart from the X coordinate of its nearest vertical edge edge X coordinate as unknown node smaller or equal to k to the distance of the left hand edge in zone or right hand edge; If this coordinate Y value smaller or equal to k, then uses Y coordinate apart from its nearest lateral edge as unknown node Y coordinate to the distance of the top edge in zone or lower limb; If this coordinate is positioned at outside the zone, then use X coordinate apart from its nearest vertical edge edge as unknown node X coordinate, use Y coordinate apart from its nearest lateral edge as unknown node Y coordinate;

After accomplishing, welt obtains the final position coordinate of node coordinate corrected value as unknown node.

Further, in said step (2.3), the trilateration process is following: the coordinate of supposing 3 beaconing nodes is respectively (x ₁, y ₁), (x ₂, y ₂), (x ₃, y ₃), the coordinate of the unknown node of position to be determined is (x _u, y _u), the distance of this node to 3 reference node is respectively d ₁, d ₂, d ₃, according to two-dimensional space distance calculation formula, obtain equation group:

$\left\{\begin{array}{c}{d}_1=\sqrt{{(x_1-x_u)}^2{+(y_1-y_u)}^2}\\ d_2=\sqrt{{(x_2-x_u)}^2+{(y_2-y_u)}^2}\\ d_3=\sqrt{{(x_3-x_u)}^2+{(y_3-y_u)}^2}\end{array}\right.$

In above-mentioned equation group, x _u, y _uBe unknown quantity, find the solution this equation group, can obtain unknown node (x _u, y _u) position coordinates.

Further again, in said step (2.1), the computational process of the minimum hop count of unknown node and beaconing nodes is:

Service range vector exchange agreement, through internodal information exchange, all nodes are obtained and beaconing nodes between the jumping distance; The node that receives grouping is at first according to (X ₁, Y ₁)-(X ₂, Y ₂) judge whether this beaconing nodes belongs to this node region; If not, then abandon this grouping, otherwise note the minimum hop count of this beaconing nodes; Ignore from the bigger grouping of the jumping figure value of same beaconing nodes; Hop field value in will dividing into groups then adds 1, and is transmitted to neighbor node, finally makes unknown node in the network can both note the minimum hop count of each beaconing nodes in self region.

Technical conceive of the present invention is: radio sensing network is made up of a large amount of nodes of disposing at random, so the position of most of node is unknown in the network.Yet use for great majority, do not know that the perception data of node location has no meaning.Sensor node must be known just illustrative " where or zone what specific thing has taken place " of self-position.So location technology is one of key technology of radio sensing network, play a supporting role in the every field of radio sensing network.

The DV-Hop algorithm is a kind of Distributed localization algorithm that is utilized distance vector route (Distance vector routing) and the proposition of GPS Positioning Principle by the people such as DragosNiculescu of U.S. rutgers university (Rutgers University).Basic thought is; At first calculate the minimum hop count of unknown node to reference point; The average every hop distance of reference point computing network then; Represent unknown node to the estimated distance between the reference point with the average every hop distance of network with to the product of the minimum hop count of reference point again, utilize trilateration to obtain the position of node at last.

The IEEGS system is one and is structured on the radio sensing network, carries out the system of disaster monitoring, escape guiding, and it has very big dependence to the positional information of node in the network.Except general radio sensing network node location requirement, the IEEGS system also has the characteristics of himself aspect node locating, and, node high like positioning accuracy request distributes and present region characteristic and local edge etc.According to the special location requirement of IEEGS system, propose semi-automatic average every hop distance and obtain, promptly in acquisition process, add the artificial operation of a part, exchange the algorithm positioning performance for workload, improve the node locating precision; Node distributed areas characteristic according to the IEEGS grid is had proposes area dividing, promptly is divided into plurality of single zone to whole network, on each zone, uses different parameters to move location algorithm respectively; According to the distinctive node edge distribution character of IEEGS network, the coordinate welt is proposed, make and can suitable node coordinate directly be abutted on the edges of regions according to the relation of node coordinate and edges of regions.

The present invention is based on the DV-Hop algorithm, its core concept is: the average every hop distance of minimum hop count between the computing node and network at first is shown as between the two jumping figure and the product of the average every hop distance of network to the distance table between the node.Obtain unknown node after the distance of reference point, utilizing trilateration to obtain the position coordinates initial value of node.On this basis, ASBS has merged that semi-automatic average every hop distance obtains, the thought of area dividing, coordinate welt, to adapt to the location needs of IEEGS, improves the practical application performance of algorithm.

The proposition that semi-automatic average every hop distance obtains: establish a network that node density is average, its node density is ρ, and the node communication radius is r, can know nearly n=ρ π r in the single-hop scope of a node ²Individual node.General, when n node that is evenly distributed arranged in the single-hop scope at a node, because density is average, this n node distribution situation was as shown in Figure 2 under the ideal situation.Great circle is the direct communications range of Centroid N among the figure, is called the single-hop range circle.This n node will have the 6i node around the Centroid storied placement on the i layer, every node layer is formed regular hexagon; All 2 adjacent nodal distances are identical; Per 3 adjacent in twos nodes are formed equilateral triangle; All two adjacent interlamellar spacings are identical, for number of plies single-hop range circle radius divided by the number of plies, be designated as h.For for simplicity, and do not lose accuracy, can estimate average every hop distance ASPH value with the h value.

If n node that is evenly distributed arranged in the single-hop range circle of a node, these nodes are lined up the m layer according to top rule, then just like lower inequality:

$\frac{m(6+6m)}{2}\≥n---\left(1\right)$

Try to achieve the span of m

$m\≥\frac{-1+\sqrt{1+\frac{4n}{3}}}{2}$ Or $m\≤\frac{-1-\sqrt{1+\frac{4n}{3}}}{2}$

Because the latter is cast out in m＞0, substitution n=ρ π r ², get π ≈ 3, and get the suitable scaling of result

$m\≥r\sqrt{\mathrm{\ρ}}-\frac{1}{2}---\left(2\right)$

Get

$m\≈r\sqrt{\mathrm{\ρ}}-\frac{1}{2}---\left(3\right)$

Promptly; According to the described rule of preamble; N around Centroid; Lined up

layer approximately, then

$h=\frac{r}{m}\≈\frac{r}{r\sqrt{\mathrm{\ρ}}-\frac{1}{2}}=\frac{2}{\sqrt{\mathrm{\ρ}}-\frac{1}{r}}---\left(4\right)$

Consider h＞0, therefore work as $\sqrt{\mathrm{\ρ}}-\frac{1}{r}\≤0$ The time, directly get h=r,

Then the average every hop distance of estimation network is following:

$\mathrm{ASPH}=h=\left\{\begin{array}{cc}\frac{2}{\sqrt{\mathrm{\ρ}}-\frac{1}{r}},& \sqrt{\mathrm{\ρ}}>\frac{i}{r}\\ r,& \sqrt{\mathrm{\ρ}}\≤\frac{1}{r}\end{array}\right.---\left(5\right)$

Following formula is under the most desirable, the most theoretical situation, to derive and come, and in reality, can not have such situation to occur, and calculates more closing to reality in order to make; Introduce a modifying factor m, its span be (0,5]; Result of calculation being carried out suitable amplification or dwindle, obtain correction formula:

$\mathrm{ASPH}=\left\{\begin{array}{cc}\frac{2m}{\sqrt{\mathrm{\ρ}}-\frac{1}{r}},& \sqrt{\mathrm{\ρ}}>\frac{i}{r}\\ \mathrm{mr},& \sqrt{\mathrm{\ρ}}\≤\frac{1}{r}\end{array}\right.---\left(6\right)$

Based on above theory, propose semi-automatic (Semi-auto, SA) average every hop distance obtains: it is average to establish in the network node density; Before location algorithm operation, in each node, manually write network node density ρ, two parameters of node communication radius, when algorithm need use average every hop distance; Utilize formula (6) to calculate the ASPH value automatically and participate in the location; Wherein ASPH be network average every hop distance (Average Size per Hop, ASPH), m is a modifying factor; Its span be (0,5].Automanual implication is, in obtaining the process of average every hop distance, is not that each step is all accomplished by node automatically, and needs artificial the participation, in node, manually writes necessary parameter in advance.

DV-Hop thought based on area dividing (Area Division) a: if wireless sensor network; Can it be divided into several regions; Make that each regional interior nodes distribution density is even; Isotropism can be divided this network area so, and in each zone, used the DV-Hop algorithm to carry out node locating respectively." C " type network topology shown in Fig. 3 (a), node distributes extremely unevenly in this network, is in two nodes in two in the north and south respectively, calculate by DV-Hop they between distance geometric distance value actual with it differ greatly.Through area dividing, shown in Fig. 3 (b), in A, B, three rectangular areas of C, the node distribution density is even, isotropism, and proterties is good, is fit to very much use the DV-Hop algorithm.

Coordinate welt thought: in the IEEGS system, node distributes except presenting region characteristic, and a kind of local edge is also arranged, and promptly most nodes all are distributed in the edge of region.Based on this specific character; Coordinate welt (Stick To Border is proposed; STB) thought: in a zone, set a minimum range threshold value k (k >=0); After the approximation of node through location algorithm acquisition self coordinate, the actual coordinate of this coordinate as node then just used if this coordinate arrives the distance at regional 4 edges up and down all greater than k in (1); (2) if the X value of this coordinate to the distance of the left hand edge in zone or right hand edge smaller or equal to k, then use X coordinate apart from its nearest vertical edge edge as the nodes X coordinate; (3) if this coordinate Y value to the distance of the top edge in zone or lower limb smaller or equal to k, then use Y coordinate apart from its nearest lateral edge as node Y coordinate; (4) if this coordinate is positioned at outside the zone, then use X coordinate apart from its nearest vertical edge edge as the nodes X coordinate, use Y coordinate apart from its nearest lateral edge as node Y coordinate.

Beneficial effect of the present invention mainly shows: 1). according to the location needs of IEEGS system, proposed that semi-automatic average every hop distance obtains, the thought of area dividing, coordinate welt, and it has successfully been introduced in the radio sensing network node location algorithm; 2). solve the node locating problem of existing IEEGS system, made wireless sensing node in the system can realize that high accuracy positions oneself; 3) the .ASBS location algorithm has carried out multiple improvement to DV-Hop, comprises that semi-automatic average every hop distance obtains, area dividing, coordinate welt, performances such as the positioning accuracy that has not only improved algorithm and the rate that runs succeeded, but also improved the practical application property of algorithm.

Description of drawings

Fig. 1 is an ASBS algorithm general flow chart.

Fig. 2 is single-hop scope interior nodes distribution situation figure in the density averaging network.

Fig. 3 is unbalanced network topology area division figure.

Embodiment

Below in conjunction with accompanying drawing the present invention is further described.

With reference to Fig. 1～Fig. 3, a kind of wireless sensor network semiautomatic node positioning method based on area dividing and coordinate welt, said semiautomatic node positioning method may further comprise the steps:

1), parameter configuration:

According to total topological diagram of network, objective network to be carried out the rectangular area divide, and confirm each beaconing nodes zone of living in, the area coordinate parameter after the division is (x ₁, y ₁)-(x ₂, y ₂), (x wherein ₁, y ₁) be regional left upper apex coordinate, (x ₂, y ₂) be regional bottom right apex coordinate;

And regions node density ρ, node communication radius, modifying factor m, welt threshold value k and beaconing nodes coordinate;

The span of modifying factor m be (0,5], beaconing nodes is a reference point, its coordinate is a known quantity;

According to Area Node density p, node communication radius, modifying factor m, the average every hop distance of computing network specifically has:

$\mathrm{ASPH}=\left\{\begin{array}{cc}\frac{2m}{\sqrt{\mathrm{\ρ}}-\frac{1}{r}},& \sqrt{\mathrm{\ρ}}>\frac{i}{r}\\ \mathrm{mr},& \sqrt{\mathrm{\ρ}}\≤\frac{1}{r}\end{array}\right.---\left(6\right)$

Wherein, ASPH is the average every hop distance of network;

2), position fixing process, specifically have:

(2.1) minimum hop count of calculating unknown node and beaconing nodes;

(2.2) calculate the jumping figure distance of unknown node and beaconing nodes: unknown node utilizes computes to arrive the geometric distance of these beaconing nodes after receiving the jumping figure information with beaconing nodes:

d _i＝ASPH×hop _i

d _iBe the distance of node to beaconing nodes i, hop _iBe the jumping figure of node to beaconing nodes i;

(2.3) calculate unknown node coordinate initial value: when a unknown node obtain with 3 or more internodal jumpings of multi-beacon apart from after, use trilateration calculating self-position, obtain the initial value of its position coordinates

(2.4) coordinate welt: according to coordinate, the welt threshold value k in unknown node self coordinate initial value, zone of living in; Carry out coordinate welt process:, then just use the actual coordinate of this coordinate as unknown node if this coordinate arrives the distance at regional 4 edges up and down all greater than k; If the X value of this coordinate, is then used apart from the X coordinate of its nearest vertical edge edge X coordinate as unknown node smaller or equal to k to the distance of the left hand edge in zone or right hand edge; If this coordinate Y value smaller or equal to k, then uses Y coordinate apart from its nearest lateral edge as unknown node Y coordinate to the distance of the top edge in zone or lower limb; If this coordinate is positioned at outside the zone, then use X coordinate apart from its nearest vertical edge edge as unknown node X coordinate, use Y coordinate apart from its nearest lateral edge as unknown node Y coordinate;

After accomplishing, welt obtains the final position coordinate of node coordinate corrected value as unknown node.

In said step (2.3), the trilateration process is following: the coordinate of supposing 3 beaconing nodes is respectively (x ₁, y ₁), (x ₂, y ₂), (x ₃, y ₃), the coordinate of the unknown node of position to be determined is (x _u, y _u), the distance of this node to 3 reference node is respectively d ₁, d ₂, d ₃, according to two-dimensional space distance calculation formula, obtain equation group:

$\left\{\begin{array}{c}{d}_1=\sqrt{{(x_1-x_u)}^2{+(y_1-y_u)}^2}\\ d_2=\sqrt{{(x_2-x_u)}^2+{(y_2-y_u)}^2}\\ d_3=\sqrt{{(x_3-x_u)}^2+{(y_3-y_u)}^2}\end{array}\right.$

In above-mentioned equation group, x _u, y _uBe unknown quantity, find the solution this equation group, can obtain unknown node (x _u, y _u) position coordinates.

In said step (2.1), the computational process of the minimum hop count of unknown node and beaconing nodes is: service range vector exchange agreement, through internodal information exchange, all nodes are obtained and beaconing nodes between the jumping distance; The node that receives grouping is at first according to (X ₁, Y ₁)-(X ₂, Y ₂) judge whether this beaconing nodes belongs to this node region; If not, then abandon this grouping, otherwise note the minimum hop count of this beaconing nodes; Ignore from the bigger grouping of the jumping figure value of same beaconing nodes; Hop field value in will dividing into groups then adds 1, and is transmitted to neighbor node, finally makes unknown node in the network can both note the minimum hop count of each beaconing nodes in self region.

The semiautomatic node positioning method of present embodiment comprises following two parts: (1) parameter configuration part.(2) actual motion part.

(1) parameter configuration part.

Parameter in that ASBS needs in service use must write in each node in advance, and these parameters comprise:

(1.1) node area coordinate of living in:, objective network is carried out need being provided with each node, so that it understands self zone of living in after the rectangular area divides according to total topological diagram of network.The area coordinate parameter is (x ₁, y ₁)-(x ₂, y ₂) form, (x wherein ₁, y ₁) be regional left upper apex coordinate, (x ₂, y ₂) be regional bottom right apex coordinate.

(1.2) Area Node density p and node communication radius: in the different zone of network; Node density and node communication radius all are not quite similar; In order to use semi-automatic mode to obtain average every hop distance, the node density ρ that need in node, write its region is (individual/m ²) and node communication radius.

(1.3) modifying factor m: semi-automatic when obtaining the average every hop distance of network, in order to make computing environment closing to reality situation, need to use modifying factor m that result of calculation is suitably adjusted; Its span is (0; 5], in application, this factor can provide according to experiment or experience.

(1.4) welt threshold value k: welt threshold value k; Described when node initial position distance areas edge and carried out the operation of coordinate welt how far the time; If it is excessive that the k value is got, can be wrong in the algorithm running the node welt that much is not distributed in edges of regions; If the k value is too small, then possibly leak the node that much needs welt.Need suitably choosing of k value according to parameters such as the edge distribution situation of network topology, the actual floor space of network, node, node density degree.Can't use mathematical formulae to confirm the value of k, can only be in actual application, through repeatedly experiment, according to the roughly span of estimation k such as artificial experience.

(1.5) position fixing process of beaconing nodes coordinate: ASBS depends on the beaconing nodes position coordinates.The manual work in advance of beaconing nodes coordinate writes, and obtains in the time of also can carrying equipment operations such as GPS to beaconing nodes.

(2) actual motion part.

The ASBS location algorithm was made up of the stage of 4 non-overlapping copies:

(2.1) the 1st stages: unknown node is calculated the minimum hop count with beaconing nodes.Use typical distance vector exchange agreement, through internodal information exchange, all nodes are obtained and beaconing nodes between the jumping distance.Detailed process is: all beaconing nodes are to neighbor node broadcasting self-position information block, and packet format is following:

ID

Hop

X

Y

X 1

Y 1

X 2

Y 2

Wherein ID is the unique identification of each beaconing nodes; Hop is the jumping figure to this beaconing nodes, and the beaconing nodes that the hop count field value is issued this grouping is initialized as 1; (X Y) is the beaconing nodes coordinate; (X ₁, Y ₁)-(X ₂, Y ₂) be the beaconing nodes region.The node that receives grouping is at first according to (X ₁, Y ₁)-(X ₂, Y ₂) judge whether this beaconing nodes belongs to this node region; If not, then abandon this grouping, otherwise note the minimum hop count of this beaconing nodes; Ignore from the bigger grouping of the jumping figure value of same beaconing nodes; Hop field value in will dividing into groups then adds 1, and is transmitted to neighbor node, finally makes all nodes in the network can both note the minimum hop count of each beaconing nodes in self region.

(2.2) the 2nd stages: the jumping figure distance of calculating unknown node and beaconing nodes.Unknown node just can utilize computes to arrive the geometric distance of these beaconing nodes after receiving the jumping figure information with beaconing nodes:

d _i＝ASPH×hop _i

d _iBe the distance of node to beaconing nodes i, ASPH is the average every hop distance of network (calculating according to formula 6), hop _iBe the jumping figure of node to beaconing nodes i.

(2.3) the 3rd stages: calculating node coordinates initial value.When unknown node obtain with 3 or more internodal jumpings of multi-beacon apart from after, use trilateration calculating self-position, obtain the initial value of its position coordinates.

Trilateration is described below: in two-dimensional space, know the distance of a point at least 3 known Beacon Points, just can confirm the coordinate of this point.Suppose that 3 coordinates of reference points are respectively (x ₁, y ₁), (x ₂, y ₂), (x ₃, y ₃), the coordinate of nodes of locations to be determined is (x _u, y _u), the distance of this node to 3 reference node is respectively d ₁, d ₂, d ₃, according to two-dimensional space distance calculation formula, can obtain equation group:

$\left\{\begin{array}{c}{d}_1=\sqrt{{(x_1-x_u)}^2{+(y_1-y_u)}^2}\\ d_2=\sqrt{{(x_2-x_u)}^2+{(y_2-y_u)}^2}\\ d_3=\sqrt{{(x_3-x_u)}^2+{(y_3-y_u)}^2}\end{array}\right.$

In above-mentioned equation group, x _u, y _uBe unknown quantity, find the solution this equation group, can obtain unknown node (x _u, y _u) position coordinates.

(2.4) the 4th stages: coordinate welt.After accomplishing the unknown node in the 3rd stage; According to self coordinate initial value, the coordinate in zone of living in, welt threshold value k; Carry out coordinate welt process:, then just use the actual coordinate of this coordinate as node if this coordinate arrives the distance at regional 4 edges up and down all greater than k; If the X value of this coordinate smaller or equal to k, then uses X coordinate apart from its nearest vertical edge edge as the nodes X coordinate to the distance of the left hand edge in zone or right hand edge; If this coordinate Y value smaller or equal to k, then uses Y coordinate apart from its nearest lateral edge as node Y coordinate to the distance of the top edge in zone or lower limb; If this coordinate is positioned at outside the zone, then use X coordinate apart from its nearest vertical edge edge as the nodes X coordinate, use Y coordinate apart from its nearest lateral edge as node Y coordinate.Welt is accomplished back acquisition node coordinate corrected value, and as the final position coordinate of node.

Claims (3)

1. wireless sensor network semiautomatic node positioning method based on area dividing and coordinate welt, it is characterized in that: said semiautomatic node positioning method may further comprise the steps:

1), parameter configuration:

According to total topological diagram of network, objective network to be carried out the rectangular area divide, and confirm each beaconing nodes zone of living in, the area coordinate parameter after the division is (x ₁, y ₁)-(x ₂, y ₂), (x wherein ₁, y ₁) be regional left upper apex coordinate, (x ₂, y ₂) be regional bottom right apex coordinate;

And regions node density ρ, node communication radius, modifying factor m, welt threshold value k and beaconing nodes coordinate;

The span of modifying factor m be (0,5], beaconing nodes is a reference point, its coordinate is a known quantity;

According to Area Node density p, node communication radius, modifying factor m, the average every hop distance of computing network specifically has:

Wherein, ASPH is the average every hop distance of network;

2), position fixing process, specifically have:

(2.1) minimum hop count of calculating unknown node and beaconing nodes;

(2.2) calculate the jumping figure distance of unknown node and beaconing nodes: unknown node utilizes computes to arrive the geometric distance of these beaconing nodes after receiving the jumping figure information with beaconing nodes:

d _i＝ASPH×hop _i

d _iBe the distance of node to beaconing nodes i, hop _iBe the jumping figure of node to beaconing nodes i;

(2.3) calculate unknown node coordinate initial value: when a unknown node obtain with 3 or more internodal jumpings of multi-beacon apart from after, use trilateration calculating self-position, obtain the initial value of its position coordinates

(2.4) coordinate welt:, carry out coordinate welt process according to coordinate, the welt threshold value k in unknown node self coordinate initial value, zone of living in: if the coordinate in this zone of living in to the distance at regional 4 edges up and down all greater than _k, then just use the actual coordinate of the coordinate in this zone of living in as unknown node; If the X value of the coordinate in this zone of living in to the distance of the left hand edge in zone or right hand edge smaller or equal to _k, then use apart from the X coordinate of its nearest vertical edge edge X coordinate as unknown node; If the coordinate Y value in this zone of living in to the distance of the top edge in zone or lower limb smaller or equal to _k, then use Y coordinate apart from its nearest lateral edge as unknown node Y coordinate; If the coordinate in this zone of living in is positioned at outside the zone, then use X coordinate apart from its nearest vertical edge edge as unknown node X coordinate, use Y coordinate apart from its nearest lateral edge as unknown node Y coordinate;

After accomplishing, welt obtains the final position coordinate of node coordinate corrected value as unknown node.

2. the wireless sensor network semiautomatic node positioning method based on area dividing and coordinate welt as claimed in claim 1 is characterized in that: in said step (2.3), the trilateration process is following: the coordinate of supposing 3 beaconing nodes is respectively (x ₁, y ₁), (x ₂, y ₂), (x ₃, y ₃), the coordinate of the unknown node of position to be determined is (x _u, y _u), the distance of this node to 3 reference node is respectively d ₁, d ₂, d ₃, according to two-dimensional space distance calculation formula, obtain equation group:

In above-mentioned equation group, x _u, y _uBe unknown quantity, find the solution this equation group, can obtain unknown node (x _u, y _u) position coordinates.

3. according to claim 1 or claim 2 wireless sensor network semiautomatic node positioning method based on area dividing and coordinate welt, it is characterized in that: in said step (2.1), the computational process of the minimum hop count of unknown node and beaconing nodes is:

Service range vector exchange agreement, through internodal information exchange, all nodes are obtained and beaconing nodes between the jumping distance; The node that receives grouping is at first according to (X ₁, Y ₁)-(X ₂, Y ₂) judge whether this beaconing nodes belongs to this node region; If not, then abandon this grouping, otherwise note the minimum hop count of this beaconing nodes; Ignore from the bigger grouping of the jumping figure value of same beaconing nodes; Hop field value in will dividing into groups then adds 1, and is transmitted to neighbor node, finally makes unknown node in the network can both note the minimum hop count of each beaconing nodes in self region.

Images