CN113316084A - RSSI-based gray wolf optimization differential correction centroid positioning algorithm - Google Patents

Abstract

The invention relates to the technical field of wireless sensor network positioning. In particular to an RSSI-based gray wolf optimization differential correction centroid positioning algorithm. Mainly aiming at the problems of low speed and low precision of a traditional wireless sensor network positioning method, the grey wolf optimization differential correction centroid positioning algorithm based on RSSI is provided, the algorithm calculates the distance between the algorithm and a beacon node through the received signal strength and determines the centroid, a differential correction factor is calculated according to the position relation between the centroid and the beacon node to correct the centroid coordinate, the grey wolf optimization algorithm is adopted to obtain the unknown node position based on the centroid coordinate, and the positioning precision is effectively improved.

Description

An RSSI-based Optimal Differential Correction Centroid Localization Algorithm for Grey Wolf

technical field

The invention relates to the technical field of wireless sensor network positioning. Specifically, it relates to an RSSI-based grey wolf optimal differential correction centroid location algorithm. Calculate the distance to the beacon node and determine the centroid according to the received signal strength, calculate the differential correction factor according to the positional relationship between the centroid and the beacon node to correct the centroid coordinates, and use the gray wolf optimization algorithm to obtain the unknown node position based on the centroid coordinates, which effectively improves the positioning accuracy.

Background technique

The methods of sensor network node location are roughly divided into two categories, one is based on ranging, and the other does not require ranging. The advantage of the positioning algorithm that does not require ranging is that the power consumption is small, but the accuracy is relatively low. The node localization algorithm based on ranging has high accuracy, but the disadvantage is that the computational complexity is high. The RSSI-based positioning algorithm is one of the most widely used wireless sensor network positioning algorithms. It is a positioning algorithm based on the connectivity of nodes and does not need to measure the distance between nodes. It is also one of the research hotspots.

Compared with the traditional mathematical method, the intelligent optimization algorithm provides another idea for solving the coordinates of unknown nodes based on the calculation of the centroid and through the idea of group optimization, which can reduce the positioning error to a certain extent. Common swarm intelligence optimization algorithms include genetic algorithm, particle swarm algorithm, and ant colony algorithm. These swarm intelligence optimization algorithms can improve the positioning accuracy. In comparison, the Grey Wolf Optimization algorithm (GWO, Grey Wolf Optimization) has better optimization stability and accuracy.

SUMMARY OF THE INVENTION

The purpose of the present invention is to solve the problems of slow speed and low precision of traditional wireless sensor network positioning methods, to provide a gray wolf optimization differential correction centroid positioning algorithm based on RSSI, and to introduce gray wolf intelligent optimization algorithm (GWO) to optimize the centroid to achieve accurate The positioning of unknown nodes improves the positioning accuracy of RSSI-based positioning methods.

The technical scheme of the present invention is: an RSSI-based optimal differential correction centroid positioning algorithm for gray wolf, which calculates the distance from the beacon node and determines the centroid according to the received signal strength, and calculates the differential correction factor according to the positional relationship between the centroid and the beacon node. The center coordinate is corrected, and the gray wolf optimization algorithm is used to obtain the unknown node position based on the center of mass coordinate, which effectively improves the positioning accuracy. The specific steps include the following.

Step 1: Initialization of the wireless sensor network: The sensor nodes are randomly arranged in the target area, and the beacon node periodically broadcasts a signal, and the signal contains its own ID, coordinates and other information.

Step 2: The unknown node receives multiple beacon node signals and preprocesses each group of RSSI values. Then, the distance to the beacon node is calculated by the propagation loss model, the distance is sorted from small to large, and the mapping between the distance and the beacon node is established. Take three distance values and the corresponding beacon nodes one after another in order to construct an overlapping area, and calculate the intersection of the overlapping area and the centroid of the intersection.

Step 3: Calculate the difference correction factor through the centroid M ₁ of the first three distances, and then use the difference correction factor to perform differential correction on other centroids.

Step 4: Take the obtained k (k=n-2) centroids as the initial value of the gray wolf optimization algorithm, and assign its coordinates to the gray wolf individual. Step 5: Obtain the unknown node position through the gray wolf optimization algorithm, and iterate to the preset number of times threshold. Further, in step 2, the unknown node receives multiple beacon node signals, and preprocesses each group of RSSI values. The specific content is as follows: Preprocess a set of RSSI values of each beacon node. The formula is as follows:

为第i个信标节点的RSSI的均值，

为第i个信标节点的RSSI的中值，计算公式如下：where RSSI _i is the RSSI preprocessing result of the i-th beacon node,

is the mean value of RSSI of the i-th beacon node,

is the median value of RSSI of the i-th beacon node, and the calculation formula is as follows:

where m is the number of RSSI value samples received by the unknown node for each beacon node.

Further, in step 3, the difference correction factor is calculated by the centroid M ₁ of the first three distances, and then the first centroid is used as a reference node to calculate the difference correction factor to correct the centroid in the differential correction of other centroids by using the difference correction factor. Further, in step 4, the obtained k (k=n-2) centroids are used as the initial value of the gray wolf optimization algorithm, and their coordinates are assigned to the gray wolf individuals to compare the size of k and N, which are divided into two cases, so that The algorithm can be initialized normally.

Further, in step 5, the gray wolf algorithm is used to optimize the centroid to achieve precise positioning of the unknown node, thereby improving its positioning accuracy. The beneficial effects of the present invention are: based on the RSSI technology, the present invention calculates the distance to the beacon node and determines the centroid according to the received signal strength, calculates the differential correction factor according to the positional relationship between the centroid and the beacon node, and corrects the coordinates of the centroid, based on the centroid The coordinates use the gray wolf optimization algorithm to obtain the position of the unknown node, which effectively improves the positioning accuracy.

Description of drawings

Fig. 1 is the algorithm flow chart of the present invention;

Figure 2 is a comparison chart of the average error under different beacon node densities;

Fig. 3 is the average error comparison graph under the condition of different node communication radius.

Detailed ways

In order to illustrate the technical solution of the present invention more clearly, the technical solution of the present invention is further described in detail below in conjunction with the accompanying drawings: As shown in Figure 1, the present invention is an RSSI-based gray wolf optimization differential correction centroid positioning algorithm. Calculate the distance from the beacon node and determine the centroid based on the received signal strength, calculate the differential correction factor according to the positional relationship between the centroid and the beacon node to correct the centroid coordinates, and use the gray wolf optimization algorithm to obtain the unknown node position based on the centroid coordinates, effectively improving positioning precision. The specific steps include the following.

Step 1: Initialization of the wireless sensor network: The sensor nodes are randomly arranged in the target area, and the beacon node periodically broadcasts a signal, and the signal contains its own ID, coordinates and other information.

Step 2: The unknown node receives multiple beacon node signals and preprocesses each group of RSSI values. Then, the distance to the beacon node is calculated by the propagation loss model, the distance is sorted from small to large, and the mapping between the distance and the beacon node is established. Take three distance values and the corresponding beacon nodes one after another in order to construct an overlapping area, and calculate the intersection of the overlapping area and the centroid of the intersection.

Step 3: Calculate the difference correction factor through the centroid M ₁ of the first three distances, and then apply the difference correction factor to differentially correct other centroids.

Step 4: Take the obtained k (k=n-2) centroids as the initial value of the gray wolf optimization algorithm, and assign its coordinates to the gray wolf individual.

Step 5: Obtain the unknown node position through the gray wolf optimization algorithm, and iterate to the preset number of times threshold. Further, in step 2, the unknown node receives multiple beacon node signals, and preprocesses each group of RSSI values. The specific content is as follows: Preprocess a set of RSSI values of each beacon node. The formula is as follows:

为第i个信标节点的RSSI的均值，

为第i个信标节点的RSSI的中值，计算公式为：Among them, RSSI _i is the RSSI preprocessing result of the i-th beacon node,

is the mean value of RSSI of the i-th beacon node,

is the median value of RSSI of the i-th beacon node, and the calculation formula is:

Further, in step 3, the differential correction factor is calculated by the centroid M ₁ of the first three distances, and then the differential correction factor is used to perform differential correction on other centroids. The differential correction factor is:

The centroid is corrected by:

为修正后的质心坐标。通过对质心坐标的差分修正，减小了质心与未知节点的偏差。in,

are the corrected centroid coordinates. The deviation between the centroid and the unknown node is reduced by the differential correction of the centroid coordinates.

Further, in step 4, the obtained k (k=n-2) centroids are used as the initial value of the gray wolf optimization algorithm, and their coordinates are assigned to the gray wolf individual. The specific steps are as follows: set the initial iteration number t=0, set The population size is N. If k≥N, then according to the order of the triangles corresponding to the centroids, the first N will be taken as the initial value of the gray wolf population, and M ₁ (x _m1 , y _m1 ), M ₂ (x _m2 , y _m2 ) , ..., M _i (x _mi , y _mi ), ..., M _N (x _mN , y _mN ) are assigned to the gray wolf individuals (x _m1(t) , y _m1(t) ), (x _m2(t ) ₎ , y _m2(t) ), ..., (x _mi(t) , y _mi(t) ), ..., (x _mN(t) , y _mN(t) ), i is [1, N]; if k<N, you need to use the nearest centroid M ₁ (x _m1 , y _m1 ) to the unknown node to generate Nk centroid coordinates, and assign them to (x _m1(t) , y _m1(t) ) , the value of i is [k+1,N], (x _mi(t) , y _mi(t) ) is the coordinate value of the i-th gray wolf in the t-th iteration.

In step 5, the unknown node position is obtained through the gray wolf optimization algorithm, and the specific steps are as follows:

(1) Calculate the fitness function value of each individual gray wolf:

Among them, (x _mj(t) , y _mj(t) ) is the coordinate of the jth gray wolf individual, and d _ij is the distance between gray wolf i and j. The fitness values of individual gray wolves are sorted in ascending order, the first one is defined as alpha wolf, the second one is defined as beta wolf, and the third one is defined as delta wolf. (2) Update the location of the gray wolf. During the gray wolf attack phase, the position of the gray wolf is updated through a formula. (3) Recalculate the fitness function of each individual gray wolf, and update the positions of α, β, and δ according to the fitness function, and the number of iterations is t=t+1. (4) t _max is the threshold of the number of iterations, if t≤t _max , jump to (1); if t>t _max , stop the search.

The following compares the method of the present invention with the GWO positioning algorithm and the weighted centroid positioning algorithm under the condition of different beacon node densities and different node communication radii.

In order to verify the performance of the algorithm, 100 sensor nodes are randomly arranged in an area of 50m*50m, including 30% unknown nodes. The population size of the gray wolf algorithm is 30, and the maximum number of iterations of the algorithm is 300. The algorithm of this chapter, the three-sided centroid localization algorithm and the weighted centroid localization algorithm are compared and analyzed from the two aspects of node communication radius and beacon node density.

Figure 2 shows the relationship between the beacon node density and the average positioning accuracy. It can be seen from the figure that as the beacon node density increases, more beacon nodes participate in the positioning, so the average positioning error is gradually decreasing. The average positioning error of the algorithm in this paper is always lower than the other two algorithms, which shows the effectiveness of the method of improving the difference correction centroid through gray wolf optimization. When the beacon node density is greater than 35%, the change of the average positioning error is small. When the beacon node density is 50%, the average positioning error is 0.096.

Figure 3 shows the influence on the average positioning error with the change of the communication radius when the beacon node density is 35%. It can be seen from Figure 3 that when the communication radius is small, there are fewer unknown nodes within the coverage of the beacon node, and the average positioning error is large. It is obviously better than the other two positioning algorithms. When the communication radius is 30, the average positioning accuracy of this algorithm is 0.092, the GWO positioning algorithm is 0.124, and the weighted centroid positioning algorithm is 0.16.

The above-mentioned embodiments are only intended to illustrate the technical concept and features of the present invention, and the purpose is to enable those who are familiar with the art to understand the content of the present invention and implement it accordingly, and cannot limit the protection scope of the present invention. All equivalent transformations or modifications made according to the spirit of the present invention should be covered within the protection scope of the present invention.

Claims (5)

1. An RSSI-based gray wolf optimization differential correction centroid positioning algorithm is characterized in that the distance between the RSSI-based gray wolf optimization differential correction centroid positioning algorithm and a beacon node is calculated through the strength of received signals, the centroid is determined, a differential correction factor is calculated according to the position relation between the centroid and the beacon node to correct a centroid coordinate, an unknown node position is obtained through the gray wolf optimization algorithm based on the centroid coordinate, the positioning accuracy is effectively improved,

the method comprises the following specific steps:

step 1, initializing a wireless sensor network: the method comprises the following steps that sensor nodes are randomly arranged in a target area, and beacon nodes periodically broadcast signals which comprise information such as ID and coordinates of the beacon nodes;

step 2, the unknown node receives a plurality of beacon node signals, each group of RSSI values are preprocessed, then the distance to the beacon nodes is calculated through a propagation loss model, the distances are sorted from small to large, the mapping between the distances and the beacon nodes is established, three distance values and the corresponding beacon nodes are sequentially taken, an overlapping area is constructed, and the intersection point of the overlapping area and the mass center of the intersection point are calculated;

step 3, passing the center of mass M of the first three distances₁Calculating a difference correction factor, and then performing difference correction on other centroids by using the difference correction factor;

step 4, taking the obtained k (k = n-2) centroids as initial values of a gray wolf optimization algorithm, and assigning coordinates of the k (k = n-2) centroids to gray wolf individuals;

and 5, obtaining the position of the unknown node through a wolf optimization algorithm, and iterating to a preset time threshold value.

2. The RSSI-based grayish wolf optimized differential correction centroid localization algorithm as claimed in claim 1, wherein in step 2, the unknown node receives a plurality of beacon node signals and preprocesses each set of RSSI values, the specific steps are as follows:

preprocessing a set of RSSI values for each beacon, as follows:

wherein RSSIi is the result of RSSI preprocessing of the ith beacon node,

is the mean value of the RSSI of the ith beacon,

for the median of the RSSI of the ith beacon, the calculation formula is as follows:

where m is the number of RSSI value samples each beacon node is received by the unknown node.

3. The RSSI-based grayish wolf optimized differential modified centroid localization algorithm as claimed in claim 1, wherein: centroid M through the first three distances in step 3₁And calculating a difference correction factor, and then performing difference correction on other centroids by using the difference correction factor, namely calculating the difference correction factor by using the first centroid as a reference node to correct the centroids.

4. The RSSI-based grayling optimization differential correction centroid localization algorithm as claimed in claim 1, wherein k (k = n-2) centroids obtained in step 4 are used as initial values of the grayling optimization algorithm, and coordinates thereof are assigned to grayling individuals, specifically comprising the following steps:

setting initial iteration times t =0, setting a population size N, if k is larger than or equal to N, taking the first N as initial values of the gray wolf population according to the sorting of triangles corresponding to centroids, and assigning M1(xm1, ym1), M2(xm2, ym2), … …, Mi (xmi, ymi), … …, MN (xmN, ymN) to gray wolf individuals (xm1(t), ym1(t)), (xm2(t), ym2(t)), … …, (xmi (t), ymi (t)), … …, (xmn t), ymN (t)), and i takes the value of [1, N ]; if k < N, then N-k centroid coordinates need to be generated using centroid M1(xm1, ym1) nearest to the unknown node and assigned to (xm1(t), ym1(t)), i takes the value of [ k +1, N ], (xmi (t), ymi (t)) is the coordinate value of the ith grey wolf in the t-th iteration.

5. The RSSI-based grayling optimization differential correction centroid localization algorithm as claimed in claim 1, wherein in step 5, the unknown node location is obtained by grayling optimization algorithm, and the fitness function value formula of each individual grayling is calculated as follows:

wherein (x)_mj(t),y_mj(t)) Is the coordinate of the jth individual wolf, d_ijFor the distance between the gray wolves i and j, the positioning error formula is calculated as follows:

。

Images