9 » IC Electronic information » Category: R

IC Electronic information

RSSI-based distance correction for wireless sensor network localization algorithm

In Electronic Infomation Category: R | on April 19,2011

Abstract: The node itself is a wireless sensor network localization based targeting. Wireless sensor network localization algorithms, including two based on distance and TPS73625DBVR datasheet and distance independent. RSSI-based localization algorithm in which the simple and TPS73625DBVR price and widely used, but the RSSI method of ranging error large, thus affecting the localization accuracy. Proposed based on RSSI distance correction for wireless sensor networks localization algorithm. The RSSI ranging algorithm to calculate approximate location of the center of mass, as a reference point for distance correction, and TPS73625DBVR suppliers and then determine the location of the node. Simulation results show that the algorithm can improve the localization accuracy.

0 Introduction

For most wireless sensor network applications, there is no location information of the data is meaningless. Targeting wireless sensor network nodes tracking the premise that its own position. Wireless sensor network node localization algorithm based on distance and distance can be divided into two major categories of independent, distance-based location algorithm are RSSI, TOA, TDOA, AOA, etc. independent of the localization algorithm from the main centroid algorithm, DV-hop algorithm, convex Planning, MDS-MAP and so on.

RSSI distance without additional hardware, simple, low power consumption, low cost, is widely used. The technical principle is known RSSI anchor node transmitting signal strength, according to the unknown node received signal strength, signal propagation model using the distance between two points. Because of multipath, interference, occlusion and other factors, RSSI ranging less precise, various algorithms must be used to reduce the location error of positioning accuracy, which was proposed based on RSSI ranging from wireless sensor networks Fixed positioning algorithm, which can effectively reduce the RSSI ranging error of localization accuracy.

1 algorithm model

1.1 wireless signal propagation model

RSSI ranging radio signal propagation model used, including empirical model and theoretical model, theoretical model is the model in a lot of experience on the basis of data extracted from the summary.

For the empirical model, the density of first selected according to certain reference points, the establishment of signal strength and the punctuation to a letter from the mapping matrix, the actual positioning based on measured signal strength compared with the mapping matrix, and using mathematical node under test to determine fitting way to anchor the distance between nodes.

Wireless signal propagation model are free-space propagation model, the log distance path loss model, the log - normal distribution models, in which the log - normal distribution model most widely used.

Log - normal distribution models such as equation (1) below:

Where n is the path loss index, indicating that the path loss with distance growth rate in the range between 2 to 6. d0 reference for the near-Earth distance from the test decision. Type (1) can predict when the distance d, the average received energy. As the case of the same distance d, between different locations surrounding the introduction of very large and therefore X, X is a mean 0 Gaussian distribution variable.

Distance correction in order to better describe the location algorithm presented here are two reasonable assumptions:

the impact of a variety of obstacles, most of the actual situation, the equation (1) to predict the PL (d) [dB] larger than the actual signal energy; When the distance d increases, PL ( d) [dB] and the relative deviation of the actual energy consumption will increase.

1.2 to determine the centroid of the intersection model

Known three nodes A, B, C of coordinates (xa, ya), (xb, yb) and (xc, yc), node O to their distance ra, rb and rc, assuming the node O coordinates (xo, yo), then (xo, yo) values ??can be type (2) obtained, that is to A, B and C point as the center, to ra, rb, and rc is the radius for the circle, the three circles will intersect with the point O, as shown in Figure 1 (a) below.

three circles intersect the situation in Figure 1.

But in reality, since the RSSI ranging errors exist, and because the actual path loss larger than the value of the theoretical model, which means measuring the unknown point out the distance to the anchor node is always greater than the actual distance d r. To A, B and C as the center point to da1, db1 and dc1 radius for the circle, three circles will not intersect at the point O, but there is an intersection, as shown in Figure 1 (b) below.

Three circle intersection point of intersection of three boundaries, center of mass for the three points D. The coordinates of which point you can type D (3) solution.

But quadratic equation, the solution process large amount of calculation, which used the text as shown in Figure 1 (b) shows the coordinates of the point D1 D approximate coordinates of the center of mass. Pairwise intersection of three circles, the three intersecting lines will intersect at the point D1. Eq (2) pairwise subtraction equations, the intersection line, respectively, are each linear equation, D1 coordinates can be solved by the linear equation, such as the type (4).

1.3 distance correction

In some articles to point O D1 of the coordinates as an approximation, although its accuracy positioning than the trilateral approach should be high, but large errors may still exist, especially when da1, db1, dc1 and ra, rb, and rc the relative error is not particularly evident in the same, therefore the need for the distance measured RSSI methods da1, db1 and dc1 corrected, and then repeat to find out the new three-lane intersection of the coordinates of D2, you can use the points D2, the approximate coordinates as coordinates of points O.

Set point A, B and C to the distance D1 la1, la2 and la3, the overall correction factor, such as formula (5) below.

Assumed under Section 1.1 , the farther the distance from the relative error is larger, the greater the degree of its amendment, the amendment is da1 systems such as equation (6) shows, db1 and dc correction factor similar.

Amended from da2 by equation (7) obtained, db2 and dc2 similar.

2 algorithm process

Algorithm process is as follows:

the same power of the anchor node to periodically broadcast to the surrounding location information message including the node ID and coordinates. After an ordinary node receives location information to calculate the average RSSI the same anchor node;

When a certain number of ordinary nodes to collect the information anchor node, it will not receive new information. According to the common node RSSI from strong to weak sort of anchor, from (1) Find the distance from the node to the anchor node;

Select the nearest three anchor nodes;

by equation (4) Calculate the intersection of three lines D1 coordinates;

3 anchor nodes were calculated with the intersection distance;

by equation (5) calculation of the overall correction factor;

by equation (6), the respective correction factors were calculated, then equation (7) Calculate the corrected distance;

again by equation (4) Calculate the modified three-lane intersection of the coordinates of D2, D2 is the point of coordinates O approximation.

3 Simulation and Analysis The algorithm using MATLAB simulation

basic initial conditions of wireless sensor networks in 100 m 100 m in the region, the lower left corner of the region is (0,0), upper right corner is (100,100). 4,9,16,25 a homogeneous region of anchor nodes deployed, including 16 anchor nodes deployed position as shown in Table 1.

Unknown node randomly distributed in the region, the path loss coefficient is set to 2.4, each simulation experiment 500 times, the simulation results take the average of 500 times, various simulation results shown in Table 2.

anchor node 16 in Table 1 the coordinates

Table 2 Simulation results

Can be seen from the simulation results, when the small number of anchor nodes, an increase in the number of anchor nodes can significantly improve the positioning accuracy. Path loss system also have an impact on the positioning accuracy, the greater the path loss system, the higher positioning accuracy. As can be seen from Table 2, from the increased frequency of the positioning accuracy of correction had no significant effect, meaning that under normal circumstances only a distance correction can be carried out by using the distance from the revision and amendment is not compared to positioning accuracy is significantly improved .

4 Conclusion

Wireless sensor network localization algorithm based on the RSSI ranging is simple and widely used. However, RSSI ranging accuracy is not high, reducing the node positioning accuracy. RSSI-based distance correction for wireless sensor networks localization algorithm using RSSI distance, by determining the intersection near the center of mass, as a reference point for distance correction, and then determine the unknown node location. Simulation results show that the algorithm has higher ranging error tolerance, and have a high positioning accuracy. Shown in Figure 2, the three circles intersect there was no case of three intersections, the next step will study in detail the circumstances shown in Figure 2 of the algorithm to improve and perfect the algorithm.

Figure 2 three circle intersection point of intersection of the situation without 3

TPS73625DBVR datasheetTPS73625DBVR suppliersTPS73625DBVR Price

All right © 2010-2016 Certificate