INDOOR WIRELESS LOCALISATION NETWORK USING A

DYNAMIC POSITION TRACKING MODEL

Montserrat Ros, Joshua Boom

School of Electrical, Computer and Telecommunications Engineering, University of Wollongong, Wollongong, Australia

Matthew D’Souza

Autonomous Systems Laboratory, CSIRO ICT Centre, Brisbane, Australia

Keywords: Wireless Sensor Network, Localisation, Zigbee, Propagation Channel Modelling.

Abstract: There has been great interest in using position location systems for indoor environments. Indoor

environments present many challenges to using wireless localisation systems, due to the interference caused

by metal beams and walls present. Current position localisation systems use wireless channel propagation

characteristics, such as receive signal strength to localise a user’s position. We present an inexpensive and

robust wireless localisation network that can track the location of users in an indoor environment. Our

localisation network uses a dynamic position tracking model to improve the real-time tracking of mobile

nodes. The localisation network uses the Zigbee/802.15.4 wireless communications protocol. Reference

nodes are placed at known positions in a building. The reference nodes are used by mobile nodes, carried by

users to localise their position. Further work involves improving the dynamic position tracking model by

incorporating the use of motion sensors to aid tracking and to investigate how large numbers of active users

can be supported.

1 INTRODUCTION

The widespread usage of position localisation

systems such as Global Positioning (GPS) has lead

to a variety of location based services applications

such as street map guide or asset tracking. Recently

there has been great interest in position locating for

indoor navigation applications. Indoor environments

tend to cause interference for wireless devices due to

the presence of obstacles such as metal beams or

walls. This causes outdoor localisation technologies

such as GPS to not function indoors. A variety of

radio frequency localisation techniques have been

developed such as Received Signal Strength or Time

of Arrival. An example is the cellular phone network

towers used by the GSM communications network

(Otsason et al., 2005) for indoor positioning. The

received signal strength indicator allows most RF

wireless transceivers to be used as localisers.

Our paper presents a wireless localisation

network that tracks the position of users in an indoor

environment. Our localisation network was designed

to track the position of users, in particular visually

impaired people, within a building. A user carries a

mobile node that tracks their current position. The

mobile node allows the user to approximate view

their location. The localisation network consists of

reference nodes placed at predetermined coordinates

in a building level. The reference nodes are used to

determine the coordinates of the user within the

region covered by the localisation network.

Using received signal strength or other wireless

channel propagation properties may not be suitable

for tracking users in real-time due to the lengthy

time taken to accurately measure channel

propagation parameters. This paper also looks at

using a motion odometry model for tracking people

with in a building. Odometry information includes

directional-heading and speed which can be used to

predict the next position of user.

The ZigBee/802.15.4 wireless communications

protocol is used by the localisation network. Zigbee

is a low data rate wireless communications protocol

that can operate on devices with limited computing

or power resources and cater for large networks of

active devices (ZigBee Alliance, 2006).

Ros M., Boom J. and D’Souza M. (2009).

INDOOR WIRELESS LOCALISATION NETWORK USING A DYNAMIC POSITION TRACKING MODEL.

In Proceedings of the International Conference on Wireless Information Networks and Systems, pages 21-26

DOI: 10.5220/0002226200210026

 SciTePress

Figure 1: Localisation Network Overview.

The contributions of this paper can be

summarised as:

• Deployment of a sensor network for indoor

localisation and tracking.

• Analysis of a wireless channel propagation

model for indoor environments.

• Development and testing of a dynamic position

tracking model.

This paper is organised into 5 sections. Section 2

presents a review of related work. Section 3 discuss

the localisation network implementation. Section 4

presents the findings of testing conducted on the

localisation network. Conclusions and futher areas

of investigation are discussed in Section 5.

2 RELATED WORK

Different types of wireless technologies, such as

GPS have been investigated for outdoor and indoor

location systems. Unfortunately, GPS is not suitable

for indoor use and this has led to research into the

use of other wireless technologies including UWB

(Schwarz et al., 2005), ultrasonic and GSM (Otsason

et al., 2005) platforms. Regulations are not clear for

the use of UWB, and ultrasonic location detection

still requires RF transceivers. GSM uses existing

infrastructure, however accurate position resolution

indoors is difficult.

Lamarca et al (Hightower et al., 2006, LaMarca

et al., 2005) describe the Placelab geophysical

location system which users can determine their

position in an urban environment. Placelab uses the

Received Signal Strength Indicators (RSSI) from

Wireless LAN hotspots and GSM broadcast towers

to determine a user's position. The Placelab software

uses a database of known wireless LAN hotspots

and GSM broadcast towers. The Placelab software

can be used with a PDA or laptop with wireless

LAN or GSM connectivity. Localisation accuracy is

stated as being less then GPS, with 20-25m using

wireless LAN and 100 to 150m for GSM broadcast

towers. A similar technique of using RSSI is

employed by the reference node network.

Klingbeil and Wark (Klingbeil and Wark, 2008)

developed a wireless sensor network for monitoring

human motion and position in an indoor

environment. Mobile nodes with inertial and heading

sensors were worn by a person inside a building. A

Monte Carlo based localisation algorithm that used a

person’s heading, indoor map information and static

node positions was developed and tested.

3 LOCALISATION NETWORK

The localisation network as seen in Figure 1 consists

of three different nodes, coordinator, reference and

mobile. Mobile nodes are carried by users to

determine their current location. The reference nodes

are used to determine a mobile node’s position via

trilateration. The coordinator node displays the

current position of the mobile nodes in use, on a

building floor-plan.

WINSYS 2009 - International Conference on Wireless Information Networks and Systems

Figure 2: Mobile Node Position Localising Process.

3.1 Coordinator Node

The coordinator node tracks the location coordinates

of each mobile node. The coordinator node

communicates with the mobile node via the Zigbee

mesh routing connection using the reference nodes.

The coordinator is accessed by the Z-Location

graphical user interface (Texas Instruments). It

displays the current locations of users on a building

floor-plan. The coordinator node was implemented

using a CC2430 Zigbee/802.15.4 module on a

SmartRF development board, (Texas Instruments).

The coordinator node is connected via a serial

connection to a Personal Computer and is also

3.2 Reference Node

Each reference node communicates to the

coordinator node via a Zigbee network connection.

The position of each reference node has to be

initialised by the coordinator node. The reference

nodes are used by the mobile nodes for triangulation.

The reference node was implemented using the

CC2430 Zigbee/802.15.4 wireless transceiver

module from Texas Instruments (Texas

Instruments). Each CC2430 module has a unique 64-

bit network address which is used as the reference

node’s identifier. The reference nodes are powered

by rechargeable batteries.

3.3 Mobile Node

The function of the mobile node is to determine a

user’s position. The mobile node detects reference

nodes in near proximity, to determine a current

user’s position. The mobile node uses the RSSI

values of nearby reference nodes to calculate its

position. More details are given in Section 3.4. The

mobile node was implemented using the CC2431

Zigbee/802.15.4 Location Engine Module from

Texas Instruments (Texas Instruments).

3.4 Position Localisation

As mentioned earlier, the mobile node uses the

coordinates of nearby reference nodes and RSSI

values to localise the user’s current position. Figure

2 shows the process in which the mobile nodes

interact with the reference nodes. The mobile node

periodically transmits “RSSI Measure” messages to

all reference nodes within range (Figure 2a). The

reference nodes use the RSSI Measure messages to

estimate the received signal strength. Typically 5

messages are needed to calculate an averaged RSSI

value. As seen in Figure 2b, once a cycle of RSSI

Measure messages has been transmitted, the mobile

node will then broadcast an “RSSI and Position

request” message to all reference nodes in range.

Each reference node will then respond with its

calculated RSSI value and coordinates.

The RSSI value from the reference node is used

to calculate the distance between the mobile and

reference nodes using the channel propagation

equation as shown in equation (1).

(

)

nARSSI

−

= 10

(1)

Where:

• RSSI = Received Signal Strength Indicator of

Reference Node (dBm),

• d = Separation Distance between Reference and

Mobile Nodes ,

• n = the pathloss coefficient of the channel,

• A = Absolute Power received at a distance of 1m

from the transmitter (dBm) ,

Once the separation distances between all

detected reference nodes and the mobile node has

been calculated, a trilateration algorithm (Texas

Instruments) is used to compute the position of the

mobile node. The trilateration algorithm uses each

reference node’s coordinates and separation

distance. The mobile node’s CC2431 Zigbee

transceiver has an onboard location engine module

that computes the position when given the RSSI and

coordinates of detected reference nodes (Texas

Instruments). The CC2431 Zigbee transceiver has

special registers that allow the propagation channel

model to be configured (Texas Instruments).

INDOOR WIRELESS LOCALISATION NETWORK USING A DYNAMIC POSITION TRACKING MODEL

Figure 3: Test Deployment of Localisation Network.

3.5 Dynamic Position Tracking

One of the key aims of the localisation network is to

track a person in real-time. One of the drawbacks of

using RSSI as a localisation mechanism is that the

receive signal strength must be averaged over a

period of time. Averaged RSSI values are needed to

provide suitable accuracy. However lengthy

averaging periods can lead to localisation

inaccuracies. For example a moving mobile node

will distort the average received signal strength

values calculated by reference nodes.

One way to improve the accuracy of real-time

tracking is to use other odometry information such

as a person’s directional-heading and speed to

predict the next position of user. Odometry motion

models have been used for localisation tracking by

Klingbeil et al, (Klingbeil and Wark, 2008) and E.

Lau et al (Lau and Chung, 2007). We implemented

the Dynamic Position Tracking model used in (Lau

and Chung, 2007), to localise a person’s position.

The dynamic position tracking model used the speed

and current position to predict the person’s next

position. The dynamic position tracking model was

implemented using the following equations of

motion:

est(i)

pred(i)

+ a(

prev(i)

−

pred(i)

)

(2)

)

ˆˆ

(

ˆˆ

)()()()( ipredipreviprediest

VV −+=

(3)

pred(i+1)

est(i)

(4)

)()(

ˆˆ

iestipred

VV = (5)

Where:

est(i)

= The ith estimated Range,

pred(i)

= The ith predicted Range,

prev(i)

= The ith measured Range,

est(i)

= The ith estimated Velocity,

pred(i)

= The ith predicted Velocity,

a, b = Gain Constants,

= Time Update Period

Equation (5) is an iterative process, which uses

Equation (3) to adjust the position calculated by the

mobile node until the position error has been

sufficiently reduced. The speed value was initially

WINSYS 2009 - International Conference on Wireless Information Networks and Systems

set to the average human walking speed of 1.3m/s

(Murray et al., 1964) for Equation (4). The next

section will examine the accuracy of the dynamic

position tracking model. A prototype of the dynamic

position tracking model was implemented on the

Personal Computer connected to the Coordinator

node. The current prototype of the dynamic position

tracking model does not yet perform tracking in real-

time. However we intend to incorporate the model

into the coordinator node’s graphical user interface

as part of our ongoing future work.

4 EVALUATION

An initial trial of the localisation network used six

reference nodes, a mobile and coordinator nodes.

The localisation network was deployed in a building

floor corridor and covered a space of 72m

. The aim

of the trial was to evaluate the channel propagation

and the dynamic position tracking accuracy. The test

deployment can be seen in Figure 3.

4.1 Channel Propagation Estimation

We estimated the channel propagation parameters

for Equation (1): pathloss coefficient (n) and

absolute transmitted power (A), by conducing range

tests. This involved moving the mobile node in a

straight direction-heading from a reference node. At

specific separation distances, the RSSI was

measured and recorded. Maximum range testing of

the mobile node showed that it had a range of at

least 20m indoors.

Using the measured RSSI values and calculated

positions, we were able to estimate suitable values

for n and A. We found that that n=3.375 and A=40

from the linear relationship between the RSSI and

separation distance. Our testing also showed that the

Figure 4: Distance vs Time with a Constant Velocity.

channel propagation parameters are dependent on

the surrounding indoor environment and the

positions of the reference nodes. Walls and other

metal fittings will cause multipath fading to occur.

Incorrect calibration of n and A values will cause

large errors in estimating position.

4.2 Dynamic Position Accuracy

We conducted a series of experiments to test the

accuracy of the dynamic position tracking model

used by the localisation network. The first

experiment involved moving the mobile node in a

straight line in the test deployment area at a constant

walking speed. Figure 4 shows the actual, measured

and estimated travelled distances of the mobile over

time. The distance is the displacement of the mobile

node from a starting point. The measure distances

are calculated directly from the mobile node’s

coordinates (using RSSI) and the estimated distance

is computed by using the dynamic position tracking

model. The maximum error in position is 75% using

the mobile node’s coordinates. Using the dynamic

tracking model, the maximum position error was

55%. Using the dynamic position tracking model

produced a smaller error in the position estimation

the mobile node.

The second experiment was similar to the first

experiment except that a delay of 15s was

introduced midway during the test. This was done to

test how the localisation network responds to

changes in movement. Figures 4 & 5 show the

actual, measured and estimated distances with

constant velocity and changing velocity.

Two estimated distances were calculated by

varying the gain constants: a and b, in dynamic

position tracking equations’ (1) and (3). As seen in

Figure 5, the dynamic tracking model is better at

tracking the position of the mobile node than by only

relying on the mobile node’s coordinates. Also

Figure 5: Distance vs Time with a Changing Velocity.

INDOOR WIRELESS LOCALISATION NETWORK USING A DYNAMIC POSITION TRACKING MODEL

is the large position inaccuracy (50%), if the gain

constants a and b are not calibrated correctly. Using

a=0/06 and b=0.01 was found to produce the best

results.

5 CONCLUSIONS AND FURTHER

WORK

In this paper we presented a localisation network

system that tracked users in an indoor environment.

The localisation network consisted of reference

nodes placed at known positions throughout a

building. A user carried mobile node that tracked

their current position.

A dynamic position tracking model for real-time

tracking of users was also developed. We found that

using received signal strength or other wireless

channel propagation properties was not suitable for

tracking users in real-time due to the lengthy time

required to sample the channel propagation

parameters.

An initial trial of the localisation network was

conducted using six reference nodes, a mobile and

coordinator nodes. We deployed the localisation

network along a building floor corridor and covered

a space of 72m

. We measured the channel

propagation parameters and the dynamic position

tracking model accuracy. We found that by using the

dynamic position tracking model, the maximum

error in position location was reduced from 75% to

50%.

Further work involves developing a multi-

hypothesis testing model to accurately predict and

track user position. We will also look at

incorporating human motion sensors such as

accelerometers to accurately determine walking

speed. We will also be looking at a larger scale

deployment over multiple building levels and

investigating how the network’s capacity to facilitate

large numbers of active users.

ACKNOWLEDGEMENTS

The authors acknowledge the support provided to

this project by the Urban Interfaces Project,

Australasian CRC for Interaction Design (ACID).

REFERENCES

Hightower, J., LaMarca, A. & Smith, I. E. (2006) Practical

Lessons from Place Lab. Pervasive Computing, IEEE,

5, 32-39.

Klingbeil, L. & Wark, T. (2008) A Wireless Sensor

Network for Real-Time Indoor Localisation and

Motion Monitoring. Proceedings of the 7th

international conference on Information processing in

sensor networks. IEEE Computer Society.

LaMarca, A., Chawathe, Y., Consolvo, S., Hightower, J.,

Smith, I., Scott, J., Sohn, T., Howard, J., Hughes, J.,

Potter, F., Tabert, J., Powledge, P., Borriello, G. &

Schilit, B. (2005) Place Lab: device positioning using

radio beacons in the wild. Pervasive Computing. Third

International Conference, PERVASIVE 2005.

Proceedings (Lecture Notes in Computer Science Vol.

3468). Springer-Verlag. 2005, 116-33.

Lau, E.-E.-L. & Chung, W.-Y. (2007) Enhanced RSSI-

Based Real-Time User Location Tracking System for

Indoor and Outdoor Environments. Convergence

Information Technology, 2007. International

Conference on.

Murray, M. P., Drought, A. B. & Kory, R. C. (1964)

Walking Patterns of Normal Men. J Bone Joint Surg

Am, 46, 335-360.

Otsason, V., Varshavsky, A., LaMarca, A. & de Lara, E.

(2005) Accurate GSM indoor localization. UbiComp

2005: Ubiquitous Computing 7th International

Conference, UbiComp 2005. Proceedings (Lecture

Notes in Computer Science Vol. 3660) . Springer-

Verlag. 2005, 141-58.

Schwarz, V., Huber, A. & Tuchler, M. (2005) Accuracy of

a commercial UWB 3D location/tracking system and

its impact on LT application scenarios.

Texas Instruments CC2430 System-on-Chip Solution for

2.4 GHz IEEE 802.15.4 / ZigBee™, accessed 2009,

http://focus.ti.com/docs/prod/folders/print/cc2430.html.

Texas Instruments CC2431 ZigBee Development Kit,

accessed 2009, http://focus.ti.com/docs/toolsw/folders/

print/cc2431zdk.html.

Texas Instruments CC2431System-on-Chip (SoC) Solution

for ZigBee/IEEE 802.15.4 Wireless Sensor Network,

http://focus.ti.com/docs/prod/folders/print/cc2431.html.

ZigBee Alliance (2006) ZigBee Specification,

http://www.zigbee.org/en/spec_download/.

WINSYS 2009 - International Conference on Wireless Information Networks and Systems