CN105704810A - Mobile node positioning method based on connectivity and weighting correction - Google Patents

Abstract

The invention provides a mobile node positioning method based on connectivity and weighting correction. When positioning of a mobile unknown node is performed, first a connectivity formula is utilized to calculate connectivity between nodes to select a beacon node participating in positioning, then a weighting correction model is utilized to correct an RSSI distance-measuring error when distance between the unknown node and the beacon node is measured, finally when the unknown node obtains enough beacon information, a least square method is utilized to estimate coordinates of the unknown node. Compared with a Monte Carlo mobile positioning method, the method has a smaller calculated quantity, but has equal positioning accuracy.

Description

A kind of mobile node positioning method based on degree of communication and weighted correction

Technical field:

The invention belongs to wireless sensor network (WSN) field of locating technology, particularly to the mobile node positioning method in WSN alignment system。

Background technology:

WSN mobile node positioning method needs to weigh between amount of calculation and positioning precision, based on the WSN mobile node positioning method of weighted mass center calculate relatively easy, node energy consumption is less, but positioning precision is low, and required anchor node number is many；Be not easily susceptible to the impact of node mobility based on the mobile node positioning method of Monte Carlo, positioning precision is higher, but precision depends on substantial amounts of sample and iterative computation repeatedly, and algorithm complexity is higher。

Summary of the invention:

In order to have good positioning precision while realizing reducing amount of calculation, the present invention proposes a kind of mobile node positioning method based on degree of communication and weighted correction, it is adaptable to the mobile node positioning method in WSN alignment system。

The technical scheme is that

A kind of mobile node positioning method based on degree of communication and weighted correction, in unknown node moving process, utilize between degree of communication formula computing node degree of communication to carry out beacon selection, then utilize weighted correction Modifying model RSSI ranging information, finally with method of least square, unknown node is carried out position calculation。

The described process utilizing degree of communication formula to carry out beacon selection includes: beaconing nodes broadcasts own location information and identity information periodically towards periphery, in unknown node moving process, the broadcast from multiple beaconing nodes can be received, beacon in communication range also changes in the moment, so, we divide time into some time slots, in a time slot T, unknown node is monitored and records the broadcast message of beaconing nodes, utilizing degree of communication formula to calculate the degree of communication between each beaconing nodes and unknown node when time slot terminates, the definition of its formula is as follows:

$C_i=\frac{n_r}{n_s}\×100\%---\left(1\right)$

In formula, C_iRepresent beacon B_iAnd degree of communication between unknown node, n_rRepresent that in a time slot T, unknown node receives from beacon B_iBroadcast message number；N_sRepresent in a time slot T, beacon B_iThe positional information number sent。

Degree of communication C_iRepresent in a time slot T, the connectedness between each beaconing nodes and unknown node, when degree of communication is 0, represent beacon B_iDo not connect with unknown node；When degree of communication is 1, represent beacon B_iConnect with unknown node in time slot T always。

According to the size of connection angle value, choose connective a best beaconing nodes。Here not adopting fixing degree of communication threshold value to choose letter calibration method, but adopt the choosing method of beacon number, this is to ensure that or beacon density hour bad in connectedness also has enough beacon numbers to position calculating。

Described weighted correction Modifying model RSSI ranging information, including RSSI is converted into distance and weighted correction process。

Described weighted correction model uses logarithm-normal distribution model that RSSI is converted into distance, and its model can describe by following formula:

RSSI=P+G-PL (d) (2)

PL (d)=PL (d₀)+10nLg(d/d₀)+X(3)

In formula, P is transmitting power, and G is antenna gain, and d is the transmitting node air line distance to receiving node, and unit is km, PL (d) for signal path loss after distance d, and unit is dB, generally takes d₀=1m, n are path-loss factor, and scope is between 2～4, and X is average is 0, and standard deviation is the Gaussian random variable of 4～10。

By logarithm-normal distribution model it can be seen that along with the increase of distance, the range error that the RSSI deviation of equal size produces also is increasing, so the beacon from unknown node is nearer should be chosen participate in location Calculation as far as possible。Being affected by factors such as barriers in actual environment, RSSI can not meet distributed model, it is necessary to RSSI value correction could be reduced position error。Similar environment is had to it, it is possible to use the error produced when this beacon is found range revises the error produced when unknown node is found range from the nearest beaconing nodes of unknown node。

Described weighted correction process is: first, first obtains the correction coefficient of each beacon before location starts, after all n beaconing nodes broadcast self informations, and beaconing nodes B_i={ B₁, B₂..., B_nBeacon B in its communication range can be obtained_ij={ B_i1, B_i2..., B_imCoordinate, B_iWith B_ijBetween RSSI value, utilize formula (3-5) distance between two points formula can obtain B_iWith B_ijBetween actual range d_ij, formula (2) and formula (3) can obtain measuring distance d_ij’。

$d_{ij}=\sqrt{{(x_i-x_j)}^2+{(y_i-y_j)}^2}---\left(4\right)$

In formula, (x_i, y_i) for beaconing nodes B_iCoordinate, (x_j, y_j) for beacon B_ijCoordinate。

So beaconing nodes B_iWeighted correction coefficient lambda be

$\mathrm{\λ}=\underset{j=1}{\overset{n}{\Σ}}{\mathrm{\ρ}}_j\[(d_{ij}-{d_{ij}}^{\′})/{d_{ij}}^{\′}\]---\left(5\right)$

${\mathrm{\ρ}}_j=\frac{1}{{d_{ij}}^{\′}{\mathrm{\Σ}}_{j=1}^n\frac{1}{{d_{ij}}^{\′}}}---\left(6\right)$

In formula, ρ_jFor weighter factor, n is beaconing nodes B_jNumber, d_ij' for B_iWith B_jMeasure distance。

Then, when unknown node positions, measuring distance between unknown node and certain beaconing nodes is d ', and the correction coefficient from the nearest beaconing nodes of unknown node is λ, then correction distance d between this beaconing nodes and unknown node₀For

d_o=(1+ λ) d ' (7)

Described least square position calculation is to obtain 4 and above after the beacon message that degree of communication selects in unknown node, if unknown node coordinate is (x, y), the localizer beacon obtained after being selected by degree of communication is m, and unknown node is to i-th beacon (x_i, y_i) correction distance be d_0i, it is possible to obtain equation group and estimate as shown in formula (8):

$\left\{\begin{array}{c}{(x_1-x)}^2+{(y_1-y)}^2={d_{01}}^2\\ .\\ .\\ .\\ {(x_m-x)}^2+{(y_m-y)}^2={d_{0m}}^2\end{array}\right.---\left(8\right)$

So the estimated coordinates of unknown node is:

$\left[\begin{array}{c}x\\ y\end{array}\right]={\left(A^{T}A\right)}^{-1}{A}^{T}B---\left(9\right)$

In formula,

$A=2\left[\begin{array}{cc}{x}_1-x_m& y_1-y_m\\ x_2-x_m& y_2-y_m\\ .& .\\ .& .\\ .& .\\ x_{m-1}-x_m& y_{m-1}-y_m\end{array}\right]---\left(10\right)$

$B=\left[\begin{array}{c}{x_1}^2-{x_m}^2+{y_1}^2-{y_m}^2+{d_m}^2-{d_1}^2\\ .\\ .\\ .\\ {x_{m-1}}^2-{x_m}^2+{y_{m-1}}^2-{y_m}^2+{d_m}^2-{d_{m-1}}^2\end{array}\right]---\left(11\right)$

The work process of described mobile node positioning method is:

Before not starting location, first calculate the weighted correction coefficient of each beacon, ready for positioning:

(1) each beaconing nodes periodically broadcasts self ID and positional information with equal-wattage；

(2) beacon B_iAfter receiving the broadcast of communication range other beacons interior, record its RSSI value, it is contemplated that instantaneous disturbance problem, it is possible to use RSSI value is smoothed by formula (12) mean filter formula；

$\stackrel{\‾}{RSSI}=\frac{1}{n}\underset{j=0}{\overset{n}{\Σ}}{\mathrm{RSSI}}_j---\left(12\right)$

In formula, n is the RSSI value record number of times of same beacon, RSSI_jFor the RSSI value size recorded every time。

(3) with formula logarithm-normal distribution model, the RSSI value of record is converted into distance；

(4) known beacon coordinate, obtains two internodal actual distances；

(5) beacon B is obtained_iWeighted correction coefficient。

Need when unknown node is positioned to carry out following work:

(1) each beaconing nodes periodically broadcasts self ID, positional information and weighted correction coefficient with equal-wattage；

(2) after unknown node receives the broadcast of beacon, record its RSSI value, and profit calculates the degree of communication of unknown node and each beacon, select to participate in the beacon of location；

(3) RSSI value of localizer beacon is converted into distance；

(4) the weighting compensation coefficient of known locations beacon and measurement distance, utilize formula (7) to obtain correction distance；

(5) utilize method of least square to position calculating, obtain the coordinates computed of unknown node。

The present invention is with the amount of calculation reducing mobile node positioning method and ensures that positioning precision is for starting point, gives a kind of mobile node positioning method based on degree of communication and weighted correction。The method, by ensureing the connectedness between beaconing nodes and unknown node, corrects RSSI range error, reduces the impact of the remote big error of beacon RSSI, it is achieved that the lifting of positioning precision。

Accompanying drawing illustrates:

Fig. 1 is that position error is affected simulation curve by communication radius。

Fig. 2 is that position error is affected simulation curve by beacon number。

Detailed description of the invention:

Below in conjunction with accompanying drawing and specific embodiment, the method in the present invention is described。

Embodiment 1:

The region, location of the present embodiment is set to 100 × 100m, and fixed beacon node is evenly distributed in the zone, and mobile node moves with uniform velocity with the speed of 1m/s。The simulation parameter of acquiescence is: beaconing nodes number is 36, and communication radius is 45m, and decay factor is 2.45, and carrier frequency is 2400MHz, and the standard deviation of Gaussian random variable X is 5, and antenna gain is 1dB。Simulation result takes the meansigma methods of 100 tests。Node location Error Calculation adopts absolute error computing formula, as shown in formula (13):

$error=\frac{{\mathrm{\Σ}}_{i=1}^N\sqrt{{(x_i-x)}^2+{(y_i-y)}^2}}{Nr}---\left(13\right)$

In formula, N is simulation times, and r is node communication radius, (x_i, y_i) for the estimation position of unknown node, (x, y) for the actual position of unknown node。

Accompanying drawing 1 is that position error is affected simulation curve by communication radius, it will be apparent from this figure that along with the increase of communication radius, the position error of 4 kinds of location algorithms is all reducing。For monte carlo localization algorithm, this is owing to the increase of communication radius makes observation information increase therewith, eliminates partial invalidity particle, but when r increases to after to a certain degree, the error change of two kinds of monte carlo localization algorithms tends towards stability, and the MCL algorithm improved is significantly better than tradition MCL。

Based on difference correction and the weighted correction algorithm of degree of communication, along with the increase of r, error reduced before this, and this is owing to the increase of r makes the beaconing nodes in unknown node communication range increase gradually, obtains more useful positional information and positioning precision is increased。Afterwards, along with the increase of r, error no longer reduces, there is the trend increased on the contrary, this is owing to the increase of r has been included into more from unknown node beaconing nodes farther out, and remote node contains bigger RSSI error, and now RSSI error increase effect has exceeded the effect that positional information increases。

The positioning precision of weighted correction is better than the positioning precision of difference correction method, and the error increase tendency after r > 55m is less than difference correction, this is because weighted correction reduces the weight of remote beaconing nodes, reduce RSSI error and increase the impact on positioning precision。On the whole, it is between 35～50m, have close precision based on the mobile node location algorithm of degree of communication and weighted correction and the Monte Carlo EGS4 method of improvement at r, but the amount of calculation of the former algorithm is less than the latter, so the mobile node location algorithm based on degree of communication and weighted correction has more advantage。

Accompanying drawing 2 is that position error is affected simulation curve by beacon number, it will be apparent from this figure that increasing along with beaconing nodes, unknown node obtains more positional information, and the error of 4 kinds of location algorithms stably reduces, and tends towards stability。And, compare tradition MCL with the mobile node location algorithm of weighted correction based on degree of communication, mobile node location algorithm based on difference correction has less position error, it is possible to reach the precision close with the MCL algorithm improved。

Above one embodiment of the present of invention is described in detail, but described content has been only presently preferred embodiments of the present invention, it is impossible to be considered the practical range for limiting the present invention。All equalizations made according to the present patent application scope change and improvement etc., all should still belong within the patent covering scope of the present invention。

Claims (5)

1. the mobile node positioning method based on degree of communication and weighted correction, it is characterised in that: the method includes in order successively

1) in unknown node moving process, utilize between degree of communication formula computing node degree of communication to carry out beacon selection；

2) weighted correction Modifying model RSSI ranging information is utilized；

3) with method of least square, unknown node is carried out position calculation。

2. a kind of mobile node positioning method based on degree of communication and weighted correction according to claim 1, it is characterized in that: described step 1) in utilize between degree of communication formula computing node degree of communication to select to specifically refer to carry out beacon: beaconing nodes broadcasts own location information and identity information periodically towards periphery, in unknown node moving process, receiving the broadcast from multiple beaconing nodes, the beacon in communication range also changes in the moment；Dividing time into some time slots, in a time slot T, unknown node is monitored and records the broadcast message of beaconing nodes, utilizes degree of communication formula C when time slot terminates_i=n_r/n_s× 100% calculates the degree of communication between each beaconing nodes and unknown node, C in formula_iRepresent beacon B_iAnd degree of communication between unknown node, n_rRepresent that in a time slot T, unknown node receives from beacon B_iBroadcast message number, n_sRepresent in a time slot T, beacon B_iThe positional information number sent；

Degree of communication represents in a time slot T, the connectedness between each beaconing nodes and unknown node, and degree of communication is 0, represents that beacon Bi does not connect with unknown node；Degree of communication is 1, represents that beacon Bi connects with unknown node in T always。

3. a kind of mobile node positioning method based on degree of communication and weighted correction according to claim 2, it is characterised in that: according to the size of connection angle value, choose connective a best beaconing nodes；Beacon number is adopted or not fixing degree of communication threshold value to choose letter calibration method, in order to ensure that or beacon density hour bad in connectedness also has enough beacon numbers to position calculating。

4. a kind of mobile node positioning method based on degree of communication and weighted correction according to claim 1, it is characterized in that: utilize weighted correction Modifying model RSSI ranging information to specifically refer to: to carry out RSSI range error correction: first, the correction coefficient of each beacon is first obtained before location starts, after all n beaconing nodes broadcast self informations, beaconing nodes B_i={ B₁, B₂..., B_nBeacon B in its communication range can be obtained_ij={ B_i1, B_i2..., B_imCoordinate, B_iWith B_ijBetween RSSI value, utilize distance between two points formula can obtain B_iWith B_ijBetween actual range d_ij, by logarithm-normal distribution model RSSI=P+G-PL (d₀)-10nLg(d/d₀)+X can obtain measure distance d_ij', in formula, P is transmitting power, and G is antenna gain, and d is the transmitting node air line distance to receiving node, and unit is km, PL (d) for signal path loss after distance d, and unit is dB, generally takes d₀=1m, n are path-loss factor, and scope is between 2～4, and X is average is 0, and standard deviation is the Gaussian random variable of 4～10；It is possible to obtain beaconing nodes B_iWeighted correction coefficientWhereinN is beaconing nodes B_jNumber, d_ij' for B_iWith B_jMeasure distance；When unknown node positions, measuring distance between unknown node and certain beaconing nodes is d ', and the correction coefficient from the nearest beaconing nodes of unknown node is λ, then correction distance d between this beaconing nodes and unknown node_o=(1+ λ) d '。

5. a kind of mobile node positioning method based on degree of communication and weighted correction according to claim 1, it is characterized in that: position calculation is obtain 4 and above after the beacon message that degree of communication selects in unknown node by described least square position method, if unknown node coordinate is (x, y), the localizer beacon obtained after being selected by degree of communication is m, and unknown node is to i-th beacon (x_i, y_i) correction distance be d_0i, it is possible to obtain equation below group:

So the estimated coordinates of unknown node is:

In formula,