CN106658643A - RSSI based valid anchor node selecting method - Google Patents

Abstract

The invention discloses an RSSI (Received Signal Strength Indication) based valid anchor node selecting method and belongs to a field of coordinative locating. The method includes a step 1 of broadcasting the location and the RSSI signal by each anchor node; a step 2 of calculating an area path loss coefficient by each anchor node after receiving signals of other anchor nodes in the communication radius of said anchor node; a step 3 of determining constraint conditions and establishing a valid anchor node selecting model; a step 4 of screening anchor nodes according to the constraint conditions and picking up a set of valid anchor nodes with no barriers in the surrounding; a step 5 of selecting four valid anchor nodes the most adjacent to a target node from the valid anchor node set for four-side locating. According to the invention, considering the real environment of sensor nodes, the valid anchor nodes are picked up for target locating, so that influence on positioning results due to elements such as environment and enemy interference. At the same time, locating precision and locating coverage are improved effectively.

Description

Effective anchor node choosing method based on RSSI

Technical field

The present invention relates to co-located field, and in particular to based on RSSI (received signal strength Indicator effective anchor node choosing method).

Background technology

Sensor positioning is an important component part in electronic warfare system, resource allocation, task scheduling, target with The highly difficult task such as track is required for based on accurate positioning.Organize multiple different classes of reconnaissance sensors to perform collaboration to detect Examine the developing direction that location tasks are current electronic warfares.Traditional single-sensor independently scouts positioning due to by environment, enemy Many restrictions such as square situation, self performance, it is difficult to meet the demand of modern electronic warfare.Because sensor node itself has nothing Line communication capacity, it is easy to obtain signal strength values, so RSSI is a kind of convenience, cheap ranging technology.

For different co-located models, for multi-sensor cooperation orientation problem, some correlations are had abroad and is ground Study carefully achievement.Jeongsu Lee of Information And Communication university et al. devise a kind of military mobile base station alignment system, and the system is same Shi Liyong RSSI and AOA signal carries out mixed positioning, can faster orient the small range region at target place；But be not suitable for It is accurately positioned.Tian He of University of Virginia et al. propose a kind of APIT algorithms, using communication information and section between node The relative distance that the RSSI value that the neighbor node that point is received sends is come between decision node, with the simple calculation using communication information Method is compared, and can effectively improve positioning precision；But the algorithm needs each node to have good information processing capability, is positioned to This is higher.

The country also has some achievements in research in terms of multi-sensor cooperation positioning.The such as Wang Shanshan of the National University of Defense technology People proposes a kind of RSSI-NLP algorithms, according to the RSSI value between node calculate can between communication node distance relativeness, and Positioned with reference to the method for linear programming, there is preferable locating effect；But with the increase of unknown node number, positioning precision meeting Drastically decline.Xiong Zhiguang of University Of Chongqing et al. proposes the RSSI positioning strategies based on intermediate value principle with space compensation model, real Show compromise on algorithm complex and positioning performance；But its positioning result is overly dependent upon the quality of localizing environment.

Existing location algorithm has both sides not enough mostly：One side is that the complexity of location algorithm is too high, in node In the more network of number, the computing capability that often seems is not enough；On the other hand it is the accumulation of error, sentences due to lacking effective error Other method, even the error of very little all can constantly be accumulated in position fixing process in initial alignment environment, ultimately results in The decline of positioning precision.

The content of the invention

It is an object of the invention to provide a kind of effective anchor node choosing method based on RSSI.

The technical solution for realizing the object of the invention is：A kind of effective anchor node choosing method based on RSSI, specifically Comprise the following steps：

Each anchor node broadcasts its own position and its RSSI signal in step 1, positioning scene；

Step 2, for each anchor node B_i, receiving in its communication radius R after the signal of other anchor nodes, calculate B_i Zone routing loss index n (B_i)；

Step 3, determine constraints, set up effective anchor node Selection Model；

Step 4, anchor node is screened according to constraints, select the effective anchor node set without barrier around S。

Compared with prior art, its remarkable advantage is the present invention：1) present invention has according to sensor local environment dynamic select It is positioned after effect anchor node is carried out, and effectively reduces the impacts of the factor to positioning result such as environmental disturbances.2) present invention is for classics Path loss index takes the mode of empirical value to be improved, the path being dynamically determined according to sensor local environment between node Loss index.3) present invention will be applied to wireless senser co-located field based on effective anchor node Selection Model of RSSI, Co-located precision and co-located coverage rate can be significantly improved, error accumulation is prevented.

Description of the drawings

Fig. 1 is based on effective anchor node Selection Model flow chart of RSSI.

Fig. 2 is example experiment figure.

Fig. 3 is the node localization coverage figure of error threshold α.

Fig. 4 is the node locating Error Graph of error threshold α.

Fig. 5 is position error comparison diagram.

Specific embodiment

The present invention is in depth analyzed traditional RSSI localization methods, for the problems referred to above, it is proposed that a kind of effective Anchor node Selection Model EAS (Efficient Anchor-Node Selection), and entered with other models by emulation experiment Row contrast.The simulation experiment result shows that EAS models can effectively eliminate the error of initial network environment generation, improves positioning Precision, effectively prevent the accumulation of error.

A kind of effective anchor node choosing method based on RSSI of the present invention, comprises the following steps：

Step 1, each anchor node broadcasts its own position and its RSSI signal in positioning scene；

Step 2, for each anchor node B_i, receiving in its communication radius R after the signal of other anchor nodes, calculate anchor Node B_iZone routing loss index n (B_i)；The zone routing loss index n (B_i) computational methods are：

Step 2-1, determine any two anchor node B_i,B_jThe distance between, formula used is：

Wherein, (x_i,y_i,z_i),(x_j,y_j,z_j) it is respectively anchor node B_i,B_jCoordinate；

Step 2-2, determine any two anchor node B_i,B_jBetween path loss index n (B_i,B_j), the formula is：

Wherein, B_jFor sending node, B_iFor receiving node；D is the distance between two nodes；RSSI(B_i,B_j) it is B_i,B_jIt Between signal strength signal intensity；X₀It is defined as the Gaussian random variable that average is that 0, variance is 5；P_TFor transmission power, PL (d₀) it is to receive work( Rate；

Step 2-3, determine Gauss path loss impacts coefficient f_i(B_j), the formula is：

Wherein, (x_i,y_i,z_i),(x_j,y_j,z_j) it is respectively anchor node B_i,B_jCoordinate, f_i(B_j) it is that average is for 0, variance 1 gauss of distribution function；

Step 2-4, determine zone routing loss index n_R(B_i), the formula is：

Wherein, zone routing loss index n_R(B_i) be made up of N number of local path loss index, N is B_iIn communication radius R The quantity of other anchor nodes, wherein j-th local path loss index is determined by jth (1≤j≤N) individual anchor node, each office Domain path loss index calculates gained, B by the product of Gauss path loss impacts coefficient sum related to path loss index_jRoad Footpath loss index is related and represents B_jWith other anchor nodes B_kPath loss index sum between (1≤k≤N, k ≠ j).

Step 3, the constraints for determining effective anchor node, set up effective anchor node Selection Model；Effective anchor node Constraints be：

Wherein, wherein n (B_i,B_j) it is anchor node B_iCommunicate with other anchor nodes B in radius R_jBetween (1≤j≤N) Path loss index, n_R(B_i) it is anchor node B_iZone routing loss index, (x_i,y_i,z_i),(x_j,y_j,z_j) it is respectively anchor section Point B_i,B_jCoordinate, α is error threshold.

Step 4, the constraints determined according to step 3 are screened to anchor node, select having without barrier around Effect anchor node set S, completes the selection of effective anchor node.Effective anchor node screens principle：

As anchor node B_iCommunicate with other anchor nodes B in radius R_jPath loss index n (B between (1≤j≤N)_i, B_j) and B_iZone routing loss index n_R(B_i) difference when being respectively less than error threshold α, represent anchor node B_iIt is an effective anchor Node；As anchor node B_iWhen there is barrier in surrounding, B_iEffective anchor node constraints must be not content with, so as to B_iIt is not one Individual effective anchor node.

In order that those skilled in the art more fully understand technical problem in the application, technical scheme and technique effect, The effective anchor node Selection Model based on RSSI of the present invention is made further in detail with reference to the accompanying drawings and detailed description Explanation.

The present invention provides a kind of effective anchor node Selection Model based on RSSI, and basic procedure is as shown in Figure 1.Concrete steps It is as follows：

Step 1：Each anchor node broadcasts its own position and its RSSI signal；

Step 2：For each anchor node B_i, receiving in its communication radius R after the signal of other anchor nodes, calculate B_i Zone routing loss index n (B_i)；

Wherein it is determined that two anchor node B_i,B_jThe distance between, formula used is：

Wherein, (x_i,y_i,z_i),(x_j,y_j,z_j) it is respectively anchor node B_i,B_jCoordinate.

Determine two anchor node B_i,B_jBetween path loss index n (B_i,B_j), the formula is：

Wherein, B_jFor sending node, B_iFor receiving node；D is the distance between two nodes；RSSI(B_i,B_j) it is B_i,B_jIt Between signal strength signal intensity；X₀It is defined as the Gaussian random variable that average is that 0, variance is 5；P_TFor transmission power, PL (d₀) it is to receive work( Rate.

Determine Gauss path loss impacts coefficient f_i(B_j), the formula is：

Wherein, (x_i,y_i,z_i),(x_j,y_j,z_j) it is respectively anchor node B_i,B_jCoordinate.f_i(B_j) it is that average is for 0, variance 1

Gauss of distribution function.The advantages of there is high precision, broad covered area due to Gaussian Profile, therefore we are by anchor node The distance between obtain Gaussian Profile path loss impacts coefficient f as the stochastic variable of Gaussian Profile_i(B_j), and amount is come with this Change B_iOther anchor nodes are to n in communication radius R_R(B_i) impact.

Determine zone routing loss index n_R(B_i), the formula is：

Wherein, zone routing loss index n_R(B_i) be made up of N number of local path loss index, N is B_iIn communication radius R The quantity of other anchor nodes is not (including B_i).Wherein j-th local path loss index is by the individual anchor node institute of jth (1≤j≤N) It is determined that.Each local path loss index is calculated by the product of Gauss path loss impacts coefficient sum related to path loss index Gained, embodies different anchor nodes to B_iZone routing loss index Different Effects.B_jPath loss index is related and generation Table B_jWith other anchor nodes B_kPath loss index sum between (1≤k≤N, k ≠ j).The each two anchor section in calculating process Path loss index between point is all calculated respectively once, so needing to be multiplied by coefficient 1/2.

Step 3：Determine constraints, set up effective anchor node Selection Model；

Wherein, wherein n (B_i,B_j) it is anchor node B_iCommunicate with other anchor nodes B in radius R_jBetween (1≤j≤N) Path loss index, n_R(B_i) it is anchor node B_iZone routing loss index, (x_i,y_i,z_i),(x_j,y_j,z_j) it is respectively anchor section Point B_i,B_jCoordinate, α is error threshold.

Step 4：Anchor node is screened according to constraints, selects the effective anchor node set without barrier around S；

As anchor node B_iWith other anchor nodes B in its its communication radius R_jPath loss index n between (1≤j≤N) (B_i,B_j) and B_iZone routing loss index n_R(B_i) difference when being respectively less than error threshold α, represent anchor node B_iIt is one effective Anchor node.As anchor node B_iWhen there is barrier in surrounding, B_iEffective anchor node constraints must be not content with, so as to B_iNo It is an effective anchor node.

Step 5：4 nearest effective anchor nodes of distance objective node T are selected to carry out four sides from effective anchor node set S Positioning, so as to verify the effect of effective anchor node Selection Model.

It is positioned after the present invention is carried out according to the effective anchor node of sensor local environment dynamic select, effectively reduces environmental disturbances Etc. impact of the factor to positioning result.

Further detailed description is done to the present invention with reference to embodiment.

Embodiment

200 unknown nodes of random distribution in the three dimensions of 300m × 300m × 300m, as shown in Figure 2；Wherein, Δ represents anchor node, and O represents unknown node, and * is represented and oriented come by the effective anchor node Selection Model based on RSSI Positioning anchor node, black region represents the random barrier region for generating；

Define error quantization formula as follows：

Wherein, N represents unknown node sum, and R represents a unknown node communication radius,Represent unknown node m Actual coordinate, (x_m,y_m,z_m) represent the coordinate that point m is oriented through algorithm.For cannot position or unsuccessful positioning Node, position error takes the half of its communication radius.

Experiment have chosen 4 kinds of different threshold values and be tested, and Fig. 3 is the node localization coverage figure of error threshold α.From figure In it can be seen that when α=0.05, screening of the EAS algorithms to anchor node is very strict, and little anchor node participates in positioning, unknown The Signal Coverage Percentage of node is relatively low；It is stricter for the screening of anchor node when threshold alpha=0.1, participate in the anchor node of positioning Although becoming many compared to α=0.05, compared to the quantity of unknown node or less, the Signal Coverage Percentage of unknown node is not It is high；When threshold alpha=0.15, the screening to anchor node is moderate, and the Signal Coverage Percentage of unknown node is higher；When threshold alpha=0.2 When, although the anchor node for participating in positioning increases, but for the screening effect of effective anchor node dies down, causes the positioning of unknown node Coverage rate and α=0.15 be more or less the same.

Fig. 4 is the node locating Error Graph of error threshold α, figure 4, it is seen that with the increase of anchor density, α Positioning mean error when=0.15 will be significantly less than position error during α=0.2；From for the angle of Signal Coverage Percentage, α= 0.15 is with the Signal Coverage Percentage difference of α=0.2 and little, so α=0.15 is a relatively reasonable threshold value.

In the case of threshold value 0.15, compare fixed route loss index model with the standard based on EAS model orientations Exactness.

From fig. 5, it can be seen that when anchor node number is less, the advantage of EAS models is not obvious, with the increasing of anchor node number Many, the Selection effect of EAS models gradually manifests, and the locating effect of EAS models is substantially better than the positioning of fixed route loss model Effect.

Claims (4)

1. a kind of effective anchor node choosing method based on RSSI, it is characterised in that comprise the following steps：

Step 1, each anchor node broadcasts its own position and its RSSI signal in positioning scene；

Step 2, for each anchor node B_i, receiving in its communication radius R after the signal of other anchor nodes, calculate anchor node B_iZone routing loss index n (B_i)；

Step 3, the constraints for determining effective anchor node, set up effective anchor node Selection Model；

Step 4, the constraints determined according to step 3 are screened to anchor node, select the effective anchor without barrier around Node set S, completes the selection of effective anchor node.

2. effective anchor node choosing method of RSSI is based on as claimed in claim 1, it is characterised in that：In step 2, the area Domain path loss index n (B_i) computational methods are：

Step 2-1, determine any two anchor node B_i,B_jThe distance between, formula used is：

$d(B_i,B_j)=\sqrt{{(x_i-x_j)}^2+{(y_i-y_j)}^2+{(z_i-z_j)}^2}$

Wherein, (x_i,y_i,z_i),(x_j,y_j,z_j) it is respectively anchor node B_i,B_jCoordinate；

Step 2-2, determine any two anchor node B_i,B_jBetween path loss index n (B_i,B_j), the formula is：

$n(B_i,B_j)=\frac{P_T-PL\left(d_0\right)-RSSI(B_i,B_j)-X_0}{10\×\lg \[d(B_i,B_j)/d_0\]}$

Wherein, B_jFor sending node, B_iFor receiving node；D is the distance between two nodes；RSSI(B_i,B_j) it is B_i,B_jBetween believe Number intensity；X₀It is defined as the Gaussian random variable that average is that 0, variance is 5；P_TFor transmission power, PL (d₀) it is receiving power；

Step 2-3, determine Gauss path loss impacts coefficient f_i(B_j), the formula is：

$f_i\left(B_j\right)=\frac{1}{\sqrt{2\mathrm{\π}}}\exp \[-\frac{{(x_j-x_i)}^2+{(y_j-y_i)}^2+{(z_j-z_i)}^2}{2}\]$

Wherein, (x_i,y_i,z_i),(x_j,y_j,z_j) it is respectively anchor node B_i,B_jCoordinate, f_i(B_j) be average be height that 0, variance is 1 This distribution function；

Step 2-4, determine zone routing loss index n_R(B_i), the formula is：

$n_R\left(B_i\right)=\frac{1}{2}\underset{j=1}{\overset{N}{\Σ}}\left(f_i\right(B_j)\×\underset{k=1,k\≠j}{\overset{N}{\Σ}}n(B_k,B_j\left)\right)$

Wherein, zone routing loss index n_R(B_i) be made up of N number of local path loss index, N is B_iIn communication radius R other The quantity of anchor node, wherein j-th local path loss index is determined by jth (1≤j≤N) individual anchor node, each local road Footpath loss index calculates gained, B by the product of Gauss path loss impacts coefficient sum related to path loss index_jPath damage Consume correlation of indices and represent B_jWith other anchor nodes B_kPath loss index sum between (1≤k≤N, k ≠ j).

3. effective anchor node choosing method of RSSI is based on as claimed in claim 1, it is characterised in that described to have in step 3 Effect anchor node constraints be：

$\left\{\begin{array}{c}|n(B_i,B_j)-n_R\left(B_i\right)|\≤\mathrm{\α}\\ d(B_i,B_j)=\sqrt{{(x_i-x_j)}^2+{(y_i-y_j)}^2+{(z_i-z_j)}^2}\≤R\end{array}\right.$

Wherein, wherein n (B_i,B_j) it is anchor node B_iCommunicate with other anchor nodes B in radius R_jDamage in path between (1≤j≤N) Consumption index, n_R(B_i) it is anchor node B_iZone routing loss index, (x_i,y_i,z_i),(x_j,y_j,z_j) it is respectively anchor node B_i,B_j Coordinate, α is error threshold.

4. effective anchor node choosing method of RSSI is based on as claimed in claim 1, it is characterised in that described to have in step 4 Effect anchor node screens principle：

As anchor node B_iCommunicate with other anchor nodes B in radius R_jPath loss index n (B between (1≤j≤N)_i,B_j) with B_iZone routing loss index n_R(B_i) difference when being respectively less than error threshold α, represent anchor node B_iIt is an effective anchor section Point；As anchor node B_iWhen there is barrier in surrounding, B_iEffective anchor node constraints must be not content with, so as to B_iIt is not one Effective anchor node.