CN102752850A - Range-free based device and method for screening network anchor nodes - Google Patents

Abstract

The invention discloses a non-distance-based network anchor node screening and positioning device and method thereof. The average hop distance distribution function information to the anchor node stored in the unknown node is used to screen qualified anchor nodes for positioning, and a computer terminal is used to locate the anchor node. Algorithm evaluation belongs to the field of wireless sensor networks. The present invention mainly includes: the hardware design of the communicable node module; using the screening algorithm to select qualified anchor nodes, and then using the least square principle to calculate the position of the unknown node; building an algorithm evaluation system at the computer terminal, and using the algorithm in the present invention Quantitatively compare with other non-ranging algorithms, and can make a more accurate judgment on the network characteristics from the comparison results, adjust the positioning algorithm according to the network characteristics, so as to realize the optimization of system performance. The device and method of the invention can improve the positioning accuracy of unknown nodes, and the improvement effect is more obvious as the distribution unevenness of network nodes increases.

Description

A kind of network anchor node screening plant and method thereof based on non-range finding

Technical field

The invention belongs to the wireless sensor network non-range field of locating technology, be specifically related to a kind of network anchor node screening plant and method thereof based on non-range finding.

Background technology

Wireless sensor network (WSN, Wireless Sensor Networks) is by being distributed in the self-organizing network that the low-power consumption in the guarded region, small-sized sensor node are formed.Because advantages such as its low-power consumption, low price, fault freedom are strong, wireless sense network are one of focuses of wireless communication field research, and are applied to fields such as environmental monitoring, disaster scene, military battlefield and two packing spaces Journal of Sex Research always.The positional information of node has important effect in WSN, it is significant to the validity and the practicality that improve network therefore to improve positioning accuracy.

Location algorithm is of a great variety, according to the distance or the angle information that whether need in the position fixing process between the node, can it be divided into two types of location algorithm and non-location algorithms.Concerning location algorithm, generally need know the measuring distance or the angle of node and its neighbor node, obtain the position of unknown node in view of the above.Location algorithm comprises based on (the TOA time of advent; Time of Arrival), the time of advent poor (TDOA; Time Difference of Arrival), arrive angle (AOA; Angle of Arrival) and receive the algorithm of signal strength signal intensity indication (RSSI, Received Signal Strength Indicator).Non-location algorithm need not to understand distance or angle information, only needs to understand the connection characteristic information of whole network, has therefore obtained very big concern in the WSN field.Classical non-location algorithm has APIT (Approximate Point-In Triangulation test), DV-Hop (Distance Vector-Hop), MDS (Multi-dimensional Scaling) and Centroid etc.

Summary of the invention

Goal of the invention: to the relatively poor problem of non-location algorithm positioning accuracy; The present invention proposes a kind of network anchor node screening plant and method thereof based on non-range finding; This devices and methods therefor can not only improve positioning accuracy, and along with the increase of network unevenness, it is obvious more that effect is improved in the location.

Technical scheme: for realizing above-mentioned purpose; A kind of network anchor node screening plant of the present invention based on non-range finding; The ordinary node that comprises random distribution; Ordinary node comprises anchor node and unknown node; Said anchor node and unknown node include the embedded node module and the power supply of actuating force are provided for the embedded node module; Said embedded node module comprises programming chip, Flash storage chip, communication chip and antenna, and said programming chip is analyzed, screened and calculate to the jumping figure and the position data of this node each anchor node that imports in the network, and through jumping figure and the position data of each anchor node of order control communication chip broadcasting to this node; Said communication chip and antenna receive and transmission the jumping figure of wireless transmission and position data under the control of programming chip, and can the control transmission coverage; Said Flash storage chip is preserved the position and the jumping figure information of anchor nodes all in the network; Also be equipped with the GPS module in the said anchor node, the GPS module positions anchor node, accurately obtains the anchor node positional information, and positional information will be stored in the Flash storage chip of anchor node, and in general red broadcasting, outwards transmits; Communication node be used for unknown node unknown calculate finish after, all node coordinates of network are sent to terminal, communication node is made up of transmission range wireless communication module far away, that power is bigger.

Network anchor node screening plant based on non-range finding of the present invention comprises: based on the anchor node module (unknown node does not have GPS) of MSP430 programming chip, NRF2401 communication chip and GPS, and according to the information of terminal database these nodes are distributed in the zone; In the position fixing process, utilize the AFDV-Hop filtering algorithm at first to simulate normal distribution curve, calculate the unknown node coordinate through qualified anchor node again, and should estimate that coordinate was kept at intranodal according to the information that unknown node is preserved; At the computer evaluation end, utilize software to obtain the estimation coordinate of all nodes that communication node transmits, and algorithm effects is assessed.Specific as follows:

(1) the on-the-spot node module in location

A. programming Control chip part.This module comprises MSP430 programming chip, Flash storage chip and peripheral circuit thereof.Can analyze and determine to abandon still forwarding to the data that receive through the MSP430 chip, can do the least square computing the anchor node that final participation is calculated.The Flash storage chip has been preserved the position and the jumping figure information of all anchor nodes in the network.Functions such as peripheral circuit is decoupled, protection.

B. wireless communication section.This part comprises NRF2401 chip and antenna.Through writing control command to NRF2401, can realize that module receives and sends wireless data, and can the control transmission coverage.

C. supply module.Power supply is responsible for the electric energy that equipment provides various needs, is generally the lithium battery that PDA or mobile phone adopt.

D. debugging module.When the terminal need be detected or repair, this module capable of using detected.

(2) a kind of anchor node screening technique based on above-mentioned device may further comprise the steps:

Step 1; All anchor nodes that receive according to unknown node simulate average jumping apart from the distribution function curve to its jumping figure and positional information, filter out qualified anchor node again, draw the unknown node coordinate in view of the above; Defining this method is the AFDV-Hop algorithm, and concrete steps are following:

1) a random scatter n unknown node and m anchor node in the zone, wherein, 20≤m≤50,50≤n≤100, the through-put power of the embedded node module of each unknown node and anchor node is identical;

2) anchor node is with the control mode broadcast message in network that floods, and the information content comprises that anchor node sign, coordinate initial value are 0 jumping figure.Neighbor node receives after the information, and jumping figure is added 1, broadcasts again.In the process that floods, ordinary node may receive the different jumping figure information from same anchor node, chooses the minimum information of jumping figure this moment.After the end that floods, ordinary node has all been preserved the coordinate of all anchor nodes and has been arrived their fewest number of hops.

3) coordinate of unknown node u temporarily is defined as the barycenter (x of its neighbours' anchor node _Uc, y _Uc), note this (x _Uc, y _Uc) be not the estimation coordinate of unknown node u, just use in order to screen qualified anchor node.The reason that can so do is: this barycenter (x _Uc, y _Uc) and the physical location (x of unknown node _u, y _u) be identical basically to the path of certain particular anchor node.According to following formula, distance is on average jumped in the estimation that calculates all anchor node i and unknown node u.

${\mathrm{ADPH}}_{\mathrm{iu}}=\frac{\sqrt{{(x_i-x_{\mathrm{uc}})}^2+{(y_i-y_{\mathrm{uc}})}^2}}{{\mathrm{Hop}}_{\mathrm{iu}}}$

In the formula, Hop _IuExpression anchor node i is to the jumping figure of unknown node u.

4) in network, observe ADPH _IuDistribution situation.Calculate ADPH _IuProbability-distribution function following:

$P\left({\mathrm{ADPH}}_{\mathrm{iu}}\right)=\frac{\mathrm{Num}\left[{\mathrm{ADPH}}_{\mathrm{iu}}\right]}{\underset{k=0}{\overset{R}{\mathrm{\Σ}}}\mathrm{Num}[{\mathrm{ADPH}}_{\mathrm{iu}}=k]}$

In the formula, R representes the effective propagation path of ordinary node, the span of k be (0, R), Num [ADPH _Iu=k]

Expression is average jumps apart from being the number of k, and whole denominator is represented the sum of all jumping distances, molecule Num [ADPH _Iu]

The average jumping distance of expression is ADPH _IuNumber, P (ADPH _Iu) represent each and jump the probability-distribution function of distance.

Can all anchor nodes and the probability-distribution function of the average jumping distance of unknown node u according to following formula.According to this distribution function figure the anchor node of participating in calculating is screened, this distribution function is normal distribution:

$P\left({\mathrm{ADPH}}_{\mathrm{iu}}\right)=\frac{1}{\sqrt{2\mathrm{\π}}\mathrm{\σ}}{e}^{-\frac{{({\mathrm{ADPH}}_{\mathrm{iu}}-\mathrm{\μ})}^2}{{2\mathrm{\σ}}^2}}$

Wherein σ and μ are according to the constant value that simulates the normal distribution curve function, ADPH _IuValue be (0, R), can know P (ADPH according to the characteristic of normal distyribution function _Iu) most of valid data all concentrate on (ADPH _Iu) in ∈ [μ-σ, the μ+σ] zone, therefore when calculating the position of unknown node, only need average the jumping apart from the anchor node in μ-σ ~ μ+σ to get final product; And the data outside μ-σ ~ μ+σ can cause than mistake location Calculation, especially in network pockety.

5), utilize following formula can calculate anchor node and arrive unknown node apart from d for qualified anchor node _Iu

d _iu＝ADPH _iu×Hop _iu

In the formula, Hop _IuExpression anchor node i is to the jumping figure of unknown node u;

6) anchor node i to unknown node u apart from d _IuCan be write as again

(x _u-x _i) ²+(y _u-y _i) ²＝d _iu ²，i=1,2…N

Utilize least square method again, the coordinate of unknown node is carried out computing.

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

In the formula $A=-2\&times;\left[\begin{array}{ccc}{x}_1-x_n& & y_1-y_n\\ x_2-x_n& & y_2-y_n\\ & \&CenterDot;& \\ & \&CenterDot;& \\ & \&CenterDot;& \\ x_{n-1}-x_n& & y_{n-1}-y_n\end{array}\right];$ $B=\left[\begin{array}{c}{{\mathrm{<figure-callout\; id="1"\; label="d"\; filenames="HDA00001655626600012.png,HDA00001655626600031.png"\; state="\{\{state\}\}">d</figure-callout>}}_1}^2-{d_N}^2-{x_1}^2+{x_N}^2-{{\mathrm{<figure-callout\; id="1"\; label="y"\; filenames="HDA00001655626600012.png,HDA00001655626600031.png"\; state="\{\{state\}\}">y</figure-callout>}}_1}^2+{y_N}^2\\ {d_2}^2-{d_N}^2-{x_2}^2+{x_N}^2-{y_2}^2+{y_N}^2\\ \&CenterDot;\\ \&CenterDot;\\ \&CenterDot;\\ {d_{N-1}}^2-{d_N}^2-{x_{N-1}}^2+{x_N}^2-{y_{N-1}}^2+{y_n}^2\end{array}\right]$

Step 2 adopts classical non-location algorithm to calculate the unknown node coordinate;

Step 3, the coordinate of all unknown node is transferred to terminal through communication node, and computer quantizes comparison to step 1 unknown node coordinate that draws and the unknown node coordinate that classical non-location algorithm obtains:

If the positioning result of two kinds of methods of step 1 and step 2 differs less, then the node distributing homogeneity of sensor network is relatively good;

If the positioning result of two kinds of methods of step 1 and step 2 differs bigger, then the node distributing homogeneity in the sensor network is relatively poor;

Step 4 according to the screening parameter in the comparative result set-up procedure 2 in the step 3, is adjusted the screening ratio of qualified node, promptly screen effective anchor node zone and be μ-σ-Δ, μ+σ+Δ, wherein the Δ span be (0.5 σ, σ), the Δ value is 0 in the time of initially;

If the uniformity is relatively poor, then reduce the ratio that the qualified anchor node of participating in calculating accounts for total anchor node, promptly reduce the value of Δ, thereby reduce error;

If the uniformity is better, then increase the qualified anchor node of participating in calculating and account for total node ratio, promptly increase the value of Δ, thereby the calculating sample space is increased, reduce error.

Wherein, in the said step 3, the process that computer quantizes comparison is following:

(1) Integer N of initialization at first, expression is to two kinds of methods clearing Error Calculation number of times relatively;

(2) generating the network node skewness from computer is the node coordinate group of η, and in the network area distribution node;

(3) respectively with the non-location algorithm of the classics in the step 2 for example the AFDV-Hop algorithm computation in DV-Hop algorithm and the step 1 go out the coordinate E1 and the E2 of unknown node, the error of E1 and E2 is compared, and deposits comparative result in database;

(4) Integer N is subtracted 1,, then increase the value of η if N is not 0, and repeating step (2) and (3); If N is 0, forward step (5) to;

(5) draw out the situation of change of two kinds of Algorithm Error according to different unevenness η, reach a conclusion with η.

(3) terminal evaluation software system

Made up regional detection platform at terminal; The information stores that the zone is interior is in database; The coordinates computed (coordinate that calculates with DV-Hop algorithm and DV-Hop) that it can not only obtain unknown node in the zone can also compare estimation the unknown of multiple algorithm, can estimate the network characteristic of this wireless sensor network then; Thereby the parameter of adjustment location algorithm is optimized systematic function.Platform is constructed as follows:

The a.GIS function: integrated All Ranges GIS function is called with the form of API.Can draw out the reality and the estimated position of node.

B. communication function: the information from whole network need be accepted in the terminal, through the communication node in the network, accepts the information such as numbering estimated position of the node in the network.

C. data operation function: in the process of simulation screening, need analyze the threshold values condition of decision screening to the distribution situation of average jumping distance.

D. comparative evaluation function: the actual coordinate and the algorithms of different estimation coordinate of unknown node accepted at the terminal; Estimated position to various algorithms compares; Draw the error comparison diagram of estimated position; Network characteristic to wireless sensor network is estimated, thereby adjustment network algorithm or parameter reach performance optimization.

E. data storage function: be used for the positional information of storage administration node, database access interface is provided.

Beneficial effect: network anchor node screening plant and method thereof based on non-range finding of the present invention, compare with traditional non-location algorithm with conventional device, have the following advantages:

1, the network anchor node screening plant based on non-range finding of the present invention, it is skillfully constructed, and is simple in structure, be easy to build, and has advantages such as low-power consumption, low price, fault freedom be strong;

2, method of the present invention through adopting the AFDV-Hop algorithm, has effectively improved non-location algorithm positioning accuracy, has overcome the not good problem of non-location algorithm positioning accuracy.Especially in ordinary node network pockety, the AFDV-Hop algorithm more can increase substantially the positioning accuracy of traditional non-location algorithm.The estimation comparative result shows that the AFDV-Hop algorithm can effectively improve the positioning accuracy of traditional non-location algorithm.And unevenness is high more, the AFDV-Hop algorithm to improve effect obvious more.Otherwise, if known the estimated position of two kinds of algorithms, can infer the uniformity characteristic of this sensor network in view of the above to unknown node, that is, gap is big more as a result in two kinds of algorithm estimations, shows then that the distributing homogeneity of this sensor network nodes is poor more.Thereby can adjust the positional parameter of AFDV-Hop,, make that the location is more accurate location algorithm optimization.

Description of drawings:

Fig. 1 is an embedded node modular structure sketch map;

Fig. 2 is for calculating and terminal evaluation software flow chart;

Fig. 3 is that the average jumping of all anchor nodes in the unknown node is apart from distribution function figure;

Fig. 4 a is in the equally distributed network node, adopts the position error comparison diagram of DV-Hop algorithm and AFDV-Hop algorithm;

Fig. 4 b is in the network node of non-uniform Distribution, adopts the position error comparison diagram of DV-Hop algorithm and AFDV-Hop algorithm;

Fig. 5 is under the uneven distribution situation, the positioning accuracy comparison diagram of two kinds of algorithms.

Mark is following in the accompanying drawing:

The 1-power supply

The 2-NRF2401 wireless communication chips

The 3-radio antenna

4-MSP430F2XX series programming chip

5-Flash pin-saving chip unit

The 6-GPS module

Embodiment

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

In the present embodiment, classical non-location algorithm is an example with traditional DV-Hop algorithm all.

A kind of network anchor node screening plant based on non-range finding comprises the ordinary node of random distribution, and ordinary node comprises anchor node and unknown node, and anchor node and unknown node include the embedded node module and the power supply of actuating force is provided for the embedded node module.As shown in Figure 1; The embedded node module comprises programming chip 4, Flash storage chip 5, communication chip 2 and antenna 3; Each anchor node that imports in 4 pairs of networks of programming chip is analyzed, is screened and calculate to the jumping figure and the position data of this node, and through jumping figure and the position data of each anchor node of order control communication chip broadcasting to this node; Communication chip 2 and antenna 3 receive and transmission the jumping figure of wireless transmission and position data under the control of programming chip 4, and can the control transmission coverage; Position and jumping figure information that Flash storage chip 5 is preserved anchor nodes all in the network; Also be equipped with GPS module 6 in the anchor node, 6 pairs of anchor nodes of GPS module position, and accurately obtain the anchor node positional information, and positional information will be stored in the Flash storage chip of anchor node, and in general red broadcasting, outwards transmit; Communication node be used for unknown node unknown calculate finish after, all node coordinates of network are sent to terminal, communication node is made up of wireless communication module.Wherein, programming chip 4 is MSP430 programming chip, and communication chip 2 is the NRF2401 communication chip; Power supply 1 is a lithium battery, also comprises commissioning device 7.

For instance, a random distribution m=50 anchor node and n=100 unknown node in the regional extent of 200m*200m, the communication radius of all ordinary nodes is R=15m.After the embedded node module powered on, MSP430 programming chip at first carried out initialization to the hardware effort parameter, gets into workflow according to predetermined working method then.In running, anchor node can obtain its physical location in advance through GPS module 6, then its position is broadcast to neighbor node.Unknown node receives after the positional information and jumping figure of anchor node, can at first information be kept at local Flash storage chip 5, again jumping figure is added 1 and transmits.In this flow process,, then choose the minimum information of jumping figure if node receives the information from the different jumping figures of same anchor node.The data of inter-node communication are carried out verification through CRC.

Fig. 3 be the average jumping of all anchor nodes in the unknown node apart from distribution function figure, unknown node has been obtained after all anchor node information, and their average jumping apart from analyzing, is on average jumped apart from being normal distribution:

$P\left({\mathrm{ADPH}}_{\mathrm{iu}}\right)=\frac{1}{\sqrt{2\mathrm{\&pi;}}\mathrm{\&sigma;}}{e}^{-\frac{({\mathrm{ADPH}}_{\mathrm{iu}}-\mathrm{\&mu;})}{{2\mathrm{\&sigma;}}^2}}$

Characteristic according to normal distyribution function can be known, P (ADPH _Iu) most of valid data all concentrate on (ADPH _Iu) in ∈ [μ-σ, the μ+σ] zone.Therefore when calculating the position of unknown node, only need select average the jumping to get final product apart from the anchor node in μ-σ～μ+σ.

Fig. 4 a is in the equally distributed network node, adopts the position error comparison diagram of DV-Hop algorithm and AFDV-Hop algorithm.In the network of node distribution uniform, utilize the DV-Hop algorithm computation to go out the position of unknown node.Make (x _u, y _u) be the position of unknown node u, (x _i, y _i) be the position of anchor node i, d _IuBe the distance of anchor node i, following equation group arranged for the network that N anchor node arranged to unknown node u:

(x _u-x _i) ²+(y _u-y _i) ²＝d _iu ²，i=1,2…N

The employing least square method is estimated, obtains the position of u through computes:

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

In the formula $A=-2\&times;\left[\begin{array}{ccc}{x}_1-x_n& & y_1-y_n\\ x_2-x_n& & y_2-y_n\\ & \&CenterDot;& \\ & \&CenterDot;& \\ & \&CenterDot;& \\ x_{n-1}-x_n& & y_{n-1}-y_n\end{array}\right];$ $B=\left[\begin{array}{c}{{\mathrm{<figure-callout\; id="1"\; label="d"\; filenames="HDA00001655626600012.png,HDA00001655626600031.png"\; state="\{\{state\}\}">d</figure-callout>}}_1}^2-{d_N}^2-{x_1}^2+{x_N}^2-{{\mathrm{<figure-callout\; id="1"\; label="y"\; filenames="HDA00001655626600012.png,HDA00001655626600031.png"\; state="\{\{state\}\}">y</figure-callout>}}_1}^2+{y_N}^2\\ {d_2}^2-{d_N}^2-{x_2}^2+{x_N}^2-{y_2}^2+{y_N}^2\\ \&CenterDot;\\ \&CenterDot;\\ \&CenterDot;\\ {d_{N-1}}^2-{d_N}^2-{x_{N-1}}^2+{x_N}^2-{y_{N-1}}^2+{y_n}^2\end{array}\right]$

Utilize anchor node screening AFDV-Hop algorithm again, jump apart from distribution situation, filter out qualified anchor node, be suitable for least square equally and do computing according to above-mentioned anchor node.The calculating of two kinds of algorithms recorded a demerit to be compared, and can obtain Fig. 4 a.

Fig. 4 b is in the network node of non-uniform Distribution, adopts the position error comparison diagram of DV-Hop algorithm and AFDV-Hop algorithm.In node network pockety, utilize AFDV-Hop algorithm and traditional non-location algorithm (like DV-Hop algorithm, APIT algorithm) that unknown node is estimated, can obtain Fig. 4 b.

As shown in Figure 2, after the unknown node positioned, communication node can be transferred to computer software evaluation end through communication node with the result of calculation of multiple kind of algorithm under the different uniformity situation.The process that computer quantizes comparison is following: (1) is Integer N of initialization at first, and expression is to two kinds of method clearing Error Calculation number of times relatively; (2) generating the network node skewness from computer is the node coordinate group of η, and in the network area distribution node; (3) respectively with the non-location algorithm of the classics in the step 2 for example the AFDV-Hop algorithm computation in DV-Hop algorithm and the step 1 go out the coordinate E1 and the E2 of unknown node, the error of E1 and E2 is compared, and deposits comparative result in database; (4) Integer N is subtracted 1,, then increase the value of η if N is not 0, and repeating step (2) and (3); If N is 0, forward step (5) to; (5) draw out the situation of change of two kinds of Algorithm Error according to different unevenness η, reach a conclusion with η.According to the different uniformitys,, draw out the tendency chart (with traditional DV-Hop algorithm is example, compares) of the error of algorithms of different with the increase of η value.

As shown in Figure 5, under the uneven distribution situation, the positioning accuracy of two kinds of algorithms relatively based on the comparison of AFDV-Hop algorithm and traditional non-location algorithm estimated position, also can instead be released the uniform properties of this wireless sensor network.If the positioning result of two kinds of algorithms differs less, then the node distributing homogeneity of sensor network is relatively good; If the positioning result of two kinds of algorithms differs bigger, then the node distributing homogeneity in the sensor network is relatively poor.Thereby the screening parameter Δ of adjustment AFDV-Hop algorithm, Δ span are that (0.5 σ, σ), the Δ value is 0 in the time of initial.Adjust the screening ratio of qualified node, even the relatively poor qualified anchor node that then participate in to calculate of the uniformity ratio that accounts for total anchor node should reduce, thereby reduces error; If the uniformity is better, the qualified anchor node of then participating in calculating accounts for total node ratio and should increase, and makes even all thereby accomplish to enlarge computer memory, improves positioning accuracy.Adopting said method feedback, thereby the optimization of realization positioning performance.

The above only is a preferred implementation of the present invention; Be noted that for those skilled in the art; Under the prerequisite that does not break away from the principle of the invention, can also make some improvement and retouching, these improvement and retouching also should be regarded as protection scope of the present invention.

Claims (4)

1. A network anchor node screening device based on non-ranging, characterized in that: comprise randomly distributed common nodes, common nodes comprise anchor nodes and unknown nodes, and said anchor nodes and unknown nodes all comprise embedded node modules and are The embedded node module provides the power supply of the driving force. The embedded node module includes a programming chip, a Flash memory chip, a communication chip and an antenna. The data is analyzed, screened and calculated, and the communication chip broadcasts the hops and location data from each anchor node to the node through commands; the communication chip and the antenna are under the control of the programming chip, and the wireless transmission of the hops and location data Receive and send, and can control the effective distance of transmission; the Flash memory chip saves the position and hop number information of all anchor nodes in the network; the anchor node is also equipped with a GPS module, which can locate the anchor node accurately Obtain the location information of the anchor node, the location information will be stored in the Flash memory chip of the anchor node, and will be transmitted outward during the red broadcast; the communication node is used to send the coordinates of all nodes in the network after the unknown node is calculated. From computer terminals, communication nodes are composed of wireless communication modules. 2. The non-range-based network anchor node screening device according to claim 1, characterized in that: the programming chip is an MSP430 programming chip, the communication chip is an NRF2401 communication chip; the power supply is a lithium battery. 3. An anchor node screening method based on the device according to claim 1 or 2, characterized in that it comprises the following steps: Step 1: Fit the average hop distance distribution function curve according to the number of hops and location information received by the unknown node from all the anchor nodes to it, and then screen out qualified anchor nodes, and then obtain the coordinates of the unknown node, and define this method as AFDV -Hop algorithm, the specific steps are as follows: 1) Randomly scatter n unknown nodes and m anchor nodes in the area, among them, 20≤m≤50, 50≤n≤100, the transmission power of the embedded node modules of each unknown node and anchor node is the same; 2) Anchor nodes broadcast information in the network in a controlled flooding manner. The content of the information includes anchor node identifiers, coordinates, and hops with an initial value of 0. After the flooding ends, ordinary nodes save the coordinates of all anchor nodes and each The minimum number of hops from the anchor node to the common node; 3) Temporarily define the coordinates of unknown node u as the centroid (x _uc , y _uc ) of its neighbor anchor nodes, and calculate the estimated average hop distance of all anchor nodes i and unknown node u according to the following formula ${\mathrm{<span\; class="notranslate">\; ADPH</span>}}_{\mathrm{<span\; class="notranslate">\; iu</span>}}<span\; class="notranslate">\; =</span>\frac{\sqrt{{<span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; x</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}<span\; class="notranslate">\; -</span>{\mathrm{<span\; class="notranslate">\; x</span>}}_{\mathrm{<span\; class="notranslate">\; uc</span>}}<span\; class="notranslate">\; )</span>}^{\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; +</span>{<span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; the\; y</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}<span\; class="notranslate">\; -</span>{\mathrm{<span\; class="notranslate">\; the\; y</span>}}_{\mathrm{<span\; class="notranslate">\; uc</span>}}<span\; class="notranslate">\; )</span>}^{\mathrm{<span\; class="notranslate">\; 2</span>}}}}{{\mathrm{<span\; class="notranslate">\; Hop</span>}}_{\mathrm{<span\; class="notranslate">\; iu</span>}}}$ In the formula, Hop _iu represents the number of hops from anchor node i to unknown node u; 4) In the network, observe the distribution of ADPH _iu and calculate the probability of ADPH _iu . The calculation function is as follows: $\mathrm{<span\; class="notranslate">\; P</span>}<span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; ADPH</span>}}_{\mathrm{<span\; class="notranslate">\; iu</span>}}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; =</span>\frac{\mathrm{<span\; class="notranslate">\; Num</span>}<span\; class="notranslate">\; [</span>{\mathrm{<span\; class="notranslate">\; ADPH</span>}}_{\mathrm{<span\; class="notranslate">\; iu</span>}}<span\; class="notranslate">\; ]</span>}{\underset{\mathrm{<span\; class="notranslate">\; k</span>}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; 0</span>}}{\overset{\mathrm{<span\; class="notranslate">\; R</span>}}{\mathrm{<span\; class="notranslate">\; \&Sigma;</span>}}}\mathrm{<span\; class="notranslate">\; Num</span>}<span\; class="notranslate">\; [</span>{\mathrm{<span\; class="notranslate">\; ADPH</span>}}_{\mathrm{<span\; class="notranslate">\; iu</span>}}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; k</span>}<span\; class="notranslate">\; ]</span>}$ In the formula, R represents the effective transmission distance of common nodes, the value range of k is (0, R), Num[ADPH _iu =k] Indicates the number of average hop distances of k, the entire denominator represents the total number of all hop distances, and the numerator Num[ADPH _iu ] Indicates the number of ADPH _iu whose average jump distance is, and P(ADPH _iu ) indicates the probability distribution function of each jump distance; According to the above formula, the probability distribution function of the average hop distance of all anchor nodes and unknown nodes u can be obtained, and the anchor nodes participating in the calculation are screened according to the probability distribution function graph, and the distribution function is normally distributed: $\mathrm{<span\; class="notranslate">\; P</span>}<span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; ADPH</span>}}_{\mathrm{<span\; class="notranslate">\; iu</span>}}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; =</span>\frac{\mathrm{<span\; class="notranslate">\; 1</span>}}{\sqrt{\mathrm{<span\; class="notranslate">\; 2</span>}\mathrm{<span\; class="notranslate">\; \&pi;</span>}}\mathrm{<span\; class="notranslate">\; \&sigma;</span>}}{\mathrm{<span\; class="notranslate">\; e</span>}}^{<span\; class="notranslate">\; -</span>\frac{{<span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; ADPH</span>}}_{\mathrm{<span\; class="notranslate">\; iu</span>}}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; \&mu;</span>}<span\; class="notranslate">\; )</span>}^{\mathrm{<span\; class="notranslate">\; 2</span>}}}{{\mathrm{<span\; class="notranslate">\; 2</span>}\mathrm{<span\; class="notranslate">\; \&sigma;</span>}}^{\mathrm{<span\; class="notranslate">\; 2</span>}}}}$ Among them, σ and μ are constant values according to the function of the normal distribution curve, and the value of ADPH _iu is (0, R). According to the characteristics of the normal distribution function, most of the effective data of P(ADPH _iu ) are concentrated in (ADPH _iu )∈[μ-σ, μ+σ] area, so when calculating the position of unknown nodes, only the anchor nodes whose average hop distance is within μ-σ~μ+σ are needed; while in μ-σ Data other than ~μ+σ will cause large errors in positioning calculations, especially in networks with uneven distribution; 5) For a qualified anchor node, the distance d _iu from the anchor node to the unknown node can be calculated using the following formula, d _iu =ADPH _iu ×Hop _iu In the formula, Hop _iu represents the number of hops from anchor node i to unknown node u; 6) The distance d _iu from anchor node i to unknown node u can be written as (x _u -x _i ) ² +(y _u -y _i ) ² ＝d _iu ² ，i=1,2…N Then use the least squares method to calculate the coordinates of the unknown nodes: $\mathrm{<span\; class="notranslate">\; x</span>}<span\; class="notranslate">\; =</span>\left[\begin{array}{c}{\mathrm{<span\; class="notranslate">\; x</span>}}_{\mathrm{<span\; class="notranslate">\; u</span>}}\\ {\mathrm{<span\; class="notranslate">\; the\; y</span>}}_{\mathrm{<span\; class="notranslate">\; u</span>}}\end{array}\right]<span\; class="notranslate">\; =</span>{<span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; A</span>}}^{\mathrm{<span\; class="notranslate">\; T</span>}}\mathrm{<span\; class="notranslate">\; A</span>}<span\; class="notranslate">\; )</span>}^{<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; 1</span>}}{\mathrm{<span\; class="notranslate">\; A</span>}}^{\mathrm{<span\; class="notranslate">\; T</span>}}\mathrm{<span\; class="notranslate">\; B</span>}$ In the formula $A=-2\&times;\left[\begin{array}{ccc}{x}_1-x_{\mathrm{no}}& & {\mathrm{the\; y}}_1-{\mathrm{the\; y}}_{\mathrm{no}}\\ x_2-x_{\mathrm{no}}& & {\mathrm{the\; y}}_2-{\mathrm{the\; y}}_{\mathrm{no}}\\ & \&Center\; Dot;& \\ & \&Center\; Dot;& \\ & \&CenterDot;& \\ x_{\mathrm{no}-1}-x_{\mathrm{no}}& & {\mathrm{the\; y}}_{\mathrm{no}-1}-{\mathrm{the\; y}}_{\mathrm{no}}\end{array}\right];$ $B=\left[\begin{array}{c}{d_1}^2-{d_N}^2-{x_1}^2+{x_N}^2-{{\mathrm{the\; y}}_1}^2+{{\mathrm{the\; y}}_N}^2\\ {d_2}^2-{d_N}^2-{x_2}^2+{x_N}^2-{{\mathrm{the\; y}}_2}^2+{{\mathrm{the\; y}}_N}^2\\ \&Center\; Dot;\\ \&Center\; Dot;\\ \&Center\; Dot;\\ {d_{N-1}}^2-{d_N}^2-{x_{N-1}}^2+{x_N}^2-{{\mathrm{the\; y}}_{N-1}}^2+{{\mathrm{the\; y}}_{\mathrm{no}}}^2\end{array}\right]$ Step 2, using the classic non-ranging algorithm to calculate the coordinates of unknown nodes; Step 3, the coordinates of all unknown nodes are transmitted to the computer terminal through the communication node, and the computer quantitatively compares the unknown node coordinates obtained in step 1 with the unknown node coordinates obtained by the classical non-ranging algorithm: If the difference between the positioning results of the two methods in step 1 and step 2 is small, the node distribution uniformity of the sensor network is relatively good; If the positioning results of the two methods in step 1 and step 2 differ greatly, the distribution of nodes in the sensor network is poorly uniform; Step 4, adjust the screening parameters in step 2 according to the comparison results in step 3, and adjust the screening ratio of qualified nodes, that is, the area of effective anchor nodes to be screened is μ-σ-Δ, μ+σ+Δ, where the value range of Δ is (-0.5σ, σ), the initial value of Δ is 0; If the uniformity is poor, reduce the proportion of qualified anchor nodes participating in the calculation to the total anchor nodes, that is, reduce the value of Δ, thereby reducing the error; If the uniformity is better, increase the proportion of qualified anchor nodes participating in the calculation to the total nodes, that is, increase the value of Δ, so as to increase the calculation sample space and reduce errors. 4. The anchor node screening method according to claim 3, characterized in that: in the step 3, the computer performs quantitative comparison as follows: (1) First initialize an integer N, which represents the number of calculations and comparisons of the settlement errors of the two methods; (2) Generate a node coordinate group with a network node distribution unevenness of η from the computer, and distribute nodes in the network area; (3) Use the classic non-ranging algorithm in step 2 and the AFDV-Hop algorithm in step 1 to calculate the coordinates E1 and E2 of the unknown nodes, compare the errors of E1 and E2, and store the comparison results in the database ; (4) Subtract 1 from the integer N, if N is not 0, increase the value of η, and repeat steps (2) and (3); if N is 0, go to step (5); (5) According to different unevenness η, draw the variation of the error of the two algorithms with η, and draw a conclusion.

Images