Distributed Localization Using a Moving Beacon in Wireless Sensor Networks

IEEE Transactions on Parallel and Distributed System, Vol. 19, No. 5, May 2008

+Bin Xiao, +Hekang Chen and *Shuigeng Zhou+ Department of Computing, Hong Kong Polytechnic University, Hong Kong.* Department of Computer Science and Engineering, Fudan University.

Outline

Introduction Distributed Localization using a Moving Beacon Movement Patterns of the Beacon Localization in the Real Environment Performance Evaluation Conclusions

Introduction

The techniques used to identify the position of each sensor node are central to such location-aware operations. where events take place tracking moving targets assisting traffic routing providing the network geographic coverage

Introduction

It is not infeasible to equip each node in a network with a Global Positioning System. Constraints of cost and power consumption

Sensor nodes may be used to identify the position of other nodes in a sensor network. Range-based Range-free

Introduction

In this paper, the authors propose distributed method to localization of sensor nodes low hardware cost use moving beacon feasible and accurate

Distributed Localization – using a Moving Beacon

Static sensor

Moving beacon

Distributed Localization – using a Moving Beacon

Static sensor

Moving beacon

arrival position

prearrival position

departure position

post departure position

Arrival and Departure Overlap (ADO)

Distributed Localization – using a Moving Beacon

Static sensor

Moving beacon

Movement Patterns of the Beacon

Sparse-Straight-Line (SSL) Dense-Straight-Line (DSL) Random movement pattern

SSL Movement Patterns

SSL Movement Patterns

Arrival and Departure Overlap (ADO)

upper HADO

lower HADO

Distributed Localization – using a Moving Beacon

Static sensor

Moving beaconupper HADO

lower HADO

SSL Movement Patterns Moving beacon

B

SSL Movement Patterns

Line 1

Line 2

Line 3

B

Rule 1. If B is in the row immediately above the (i-1)th line, the position of A is below the ith line.

Moving beacon

Rule 2. If HADO(B) cannot contact HADOupper(A), the position of A is below the ith line.

SSL Movement Patterns

Line 1

Line 2

Line 3

A

A’

Moving beacon

SSL Movement Patterns

Line 1

Line 2

Line 3

C

Rule 3. If HADO(C) can contact HADOupper(A), the position of A is below the ith line.

A

A’

Moving beacon

DSL Movement Patterns Moving beacon

Uncovered Space Moving beacon

Random movement pattern

A

Static sensor

Moving beacon

(previous position, current position, next position)

(prearrival, arrival, departure, postdeparture)

Random movement pattern The overlap of ADOs creates a single kernel overlap area (KO

A)

),( 11

n

y

n

xn

i

n

i i i

Suppose that the KOA consists of n vertices. Let the coordinates of those n vertices (from v1 to vn) be (x1, y1), (x2, y2), . . . , (xn, yn)

Random movement pattern If node G can acquire the information of k ADOs

k prearrival positions k arrival positions k departure positions k postdeparture positions

OutPoints as a set to encompass both the prearrival and postdeparture positions

InPoints as a set to encompass both the arrival and departure positions

Point set P contain all the intersection points of these 4k circles.

Random movement pattern

Given a point u in P, its initial degree is set to 0. If the distance between u and an InPoint is not more than r, its de

gree is increased by 1 If the distance between u and an OutPoint is not less than r, its de

gree is increased by 1

Localization in the Real Environment

Out of range

Arrival position

In range

Performance Evaluation

300 nodes 500 m x 500 m Transmission range r ： 30~50 m Beacon interval s ： 1~9 m

Comparisons of the Three Movement Patterns

SSL DSL Random

General Performance Comparison

Detailed Estimate ErrorInformation

s=0.3 m s=0.6 m

Percentage of accurately localized nodes

Communication overhead

Communication overhead

Conclusions

The authors present a distributed range-free localization method use only one moving beacon feasible and accurate

The basic idea is to narrow down the possible location of a node use the arrival and departure constraint SSL pattern DSL pattern Random movement pattern

Thank You ~