WI

{
1
I

1

140
541
THS

 

firmafilfi

252 lJRDARY
Mlorugan State

University

 

 

 

 

This is to certify that the
thesis entitled

MECHANISMS FOR COMBINING INFRARED AND
ULTRASOUND SIGNALS FOR INDOOR WIRELESS
LOCALIZATION

presented by

DAVID RURANGIRWA

has been accepted towards fulfillment
of the requirements for the

MS. degree in ELECTRICAL AND
COMPUTER ENGINEERING

 

 

95M. QM

 

Major Professor’s Signature
ti / 3 0/0;
, - .

Date

 

MSU is an affirmative-action, equal-opportunity employer

.I-O-I-n-c-o-l-o-‘-o-l-.-l-0-l-l-u-u-l-o-o-I-I-I-o-I-l-l-l-0—I-g-.-u—-—.—.-.-.—.. .. .

 

 

PLACE IN RETURN BOX to remove this checkout from your record.
To AVOID FINES return on or before date due.
MAY BE RECALLED with earlier due date if requested.

 

DATE DUE DATE DUE DATE DUE

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

6/07 p:IClRC/DateDue.indd-p.1

 

 

MECHANISMS FOR COMBINING INFRARED AND ULTRASOUND SIGNALS
FOR
INDOOR WIRELESS LOCALIZATION
By

David Rurangirwa

A THESIS
Submitted to
Michigan State University

In partial fulfillment of the requirements
For the degree of

MASTER OF SCIENCE
Electrical and Computer Engineering

2007

ABSTRACT
MECHANISMS FOR COMBINING INFRARED AND ULTRASOUND SIGNALS
FOR
INDOOR WIRELESS LOCALIZATION
By

David Rurangirwa

This thesis presents a new mechanism for secure indoor localization through a
combination of infrared and ultrasound Signals. While a number of existing systems use
ultrasound and radio frequency (RF) based localization, the use of RF gives rise to a
series Of operational difficulties including lack of localization privacy and collisions
among the localization beacons. In this thesis, infrared is used to mitigate these
limitations of RF. Collisions among the localization beacons, placed in different rooms,
are avoided by leveraging the attenuation of infiared signals through walls and other
indoor partition materials. Privacy is ensured by the complete isolation of the infiared
Signal across different rooms and hallways. Also, the unlicensed usage of infrared can
provide a significant Operational advantage compared to the RF based solutions.

We implement a time dijfirence of arrival (T DOA) mechanism in which the
localization beacons send simultaneous infrared and ultrasound pulses which are received
at localization modules, which compute the distance to a beacon by measuring the TDOA
between the IR and the US signals. Applications of such indoor localization systems
include robot navigation, location-aware sensor network protocols, equipment

localization, and various location-based wireless services.

T 0 My Mother
For All Her

Love, Prayers, and Support

iii

Acknowledgements

I am grateful to God for good health, without which I could not have
accomplished much. I am also grateful to Fulbright for Sponsoring my education at
Michigan State University.

At the start of the second year of my master’s program, I asked Dr. Biswas to
direct my master’s thesis research. I explained to him that I wanted to develop a network
system. He then suggested to me the indoor location system based on infrared and
ultrasound. Throughout this work, Dr. Biswas has been of great help, always finding time
to brainstorm, troubleshoot, and make recommendations, without which, this work could
not have been possible. I am especially thankful for his systematic approach to problem
solving which enabled me to complete this project. I would like to Sincerely thank him
for not only supporting me in my research, but also for his advice, motivation, and his
willingness to let me decide which way I Should go.

Fan Yu worked with me from the start of the project until the final day when the
whole system was working. Thanks to him, I have acquired more skills of system
debugging. I would like to sincerely thank him for his tireless efforts to see that this
master’s work was complete. All along, he kept an easy outlook of things, which I am
also very thankful for.

I would also like to thank Jayanthi Rao for her help in debugging the system’s
code. She was always eager to help and was happy when things seemed to fall into place.

My fellow students in the NEEWS lab created an environment conducive to
research and I would therefore like to thank them. Their questions and suggestions during

the demo of this project were of great help.

iv

I would also like to thank the ECE Shop staff for their great help in acquisition Of
system components and experimental apparatus. I made numerous visits to the shop and

every time they were eager to help me.

TABLE OF CONTENTS

 

 

 

 

List of Tables ..................................................................................... .viii
List of Figures ....................................................................................... ix
CHAPTER 1:INTRODUCTION -- 1
1.1 BACKGROUND ...................................................................................................... 1
1.1.1 Indoor Location Systems ................................................................................. 2
1.1.2 Indoor Location System Application Example ................................................ 3
1.2 MOTIVATION ........................................................................................................ 4
1.3 CONTRIBUTIONS OF THE THESIS ........................................................................... 5
1.4 THESIS ORGANIZATION ........................................................................................ 5
CHAPTER 2:RELATED WORK- 6
2.1 THE MIT CRICKET ................................................................................................ 6
2.2 BROADBAND ULTRASONIC SYSTEM ..................................................................... 7
2.3 THE ACTIVE BADGE LOCATION SYSTEM .............................................................. 7
2.4 THE BAT ULTRASONIC LOCATION SYSTEM .......................................................... 8
2.5 IN-BUILDING RADAR .......................................................................................... 8
2.6 THE HIBALL TRACKING SYSTEM .......................................................................... 9
2.7 LOW COST INDOOR POSITIONING SYSTEM ............................................................ 9
2.8 THE HORUS WLAN LOCATION DETERMINATION SYSTEM ................................. 10
CHAPTER 3:CONCEPT 1 1
3.1 RANGING ............................................................................................................ 1 l
3.2 RANGING BASED ON TDOA OF IR AND US SIGNALS .......................................... 13
3.3 FACTORS AFFECTING DISTANCE COMPUTATION BASED ON TDOA OF IR AND US
SIGNALS .............................................................................................................. I 5
3.4 NODE LOCALIZATION MECHANISMS .................................................................. 16
CHAPTER 4:SYSTEM ARCHITECTURE 17
4.1 HARDWARE CONFIGURATION ............................................................................. 18
4.1.1 Overview of System ........................................................................................ 19
4.1.2 1R transceivers ............................................................................................... 21
4.1.3 US transducers and Supporting Circuits ....................................................... 22
4.1.4 Microcontroller ............................................................................................. 24
4.1.5 System parameters ......................................................................................... 26
4.1.5.1 Beacon ................................................................................................... 26
4.1.5.2 Listener ................................................................................................. 27
4.2 SOFTWARE CONFIGURATION .............................................................................. 27
4.2.1 Beacon Program ............................................................................................ 28
4. 2.2 Listener Program ........................................................................................... 29

vi

 

 

 

 

 

CHAPTER 5:SYSTEM CHARACTERIZATION 32
5.1 EXPERIMENTAL SETUP ....................................................................................... 32
5.2 PERFORMANCE RESULTS .................................................................................... 32

5. 2. 1 Eflect of distance and angle of rotation ......................................................... 33
5.2.2 Effect of ambient light on computed distance ................................................ 34
5.2.3 Eflect of temperature on system accuracy ..................................................... 35
5.2.4 Comparisons of errors at short distances between the Cricket and the
5.2.5 Comparisons of errors at long distances between the cricket and the
IR US ............................................................................................................. 3 7
5.2.6 Error in computed distance observed at the oscilloscope ............................. 3 7
5.3 CHALLENGES ...................................................................................................... 39

CHAPTER 6:CONCLUSION S 47

CHAPTER 7 :FUTURE WORK 48

APPENDIX A: SYSTEM SCHEMATICS 49
Al BEACON ............................................................................................................. 49
A.2 LISTENER ............................................................................................................ 50
A.3 ULTRASOUND SECTION ...................................................................................... 51

A. 3. 1 Transmitter .................................................................................................... 51
A.3.2 Receiver ......................................................................................................... 52

APPENDIX B: SYSTEM ASSEMBLY PROGRAM 53
B.I BEACON .............................................................................................................. 53
B.2 LISTENER ............................................................................................................ 58

BIBLIOGRAPHY 75

 

vii

LIST OF TABLES

Table 1: Beacon parameters .............................................................................................. 26

Table 2: Listener parameters ............................................................................................. 27

viii

LIST OF FIGURES

Figure 1: Example of an indoor localization application .................................................... 3
Figure 2: Ultrasonic Ranging ............................................................................................ 12
Figure 3: Node coordinate assignment from coordinate axes ........................................... 16
Figure 4: Node coordinate assignment from inter-node distances .................................... 16
Figure 5: Modules of the IRUS Indoor Location System .................................................. 17
Figure 6: Components of the IRUS indoor location system .............................................. 18
Figure 7: Block diagram of listener .................................................................................. 20
Figure 8: Block diagram of beacon .................................................................................. 21
Figure 9: Infrared Section of the beacon ........................................................................... 23
Figure 10: System Initializations ....................................................................................... 29
Figure 11: Experimental Setup ......................................................................................... 33

Figure 12: Directivity of US transducers used in the system; directly taken fi'om the
datasheet ........................................................................................................................... 34

Figure 13: Percentage error in computed distance with increase in distance and angle of

rotation ............................................................................................................................... 35
Figure 14: Error in computed distance in different light conditions ................................. 36
Figure 15: Error with distance increase shown with and without temperature
compensation ..................................................................................................................... 38
Figure 16: Error with computed distance at Short distances .............................................. 39
Figure 17: Error in computed distance at long distances ................................................... 40
Figure 18: TDOA determination at the scope ................................................................... 40
Figure 19: Error in computed distance observed at the oscilloscope ................................ 41
Figure 20: Beacon generated Signals ................................................................................. 43

ix

Figure 21: Infrared signal structure ................................................................................... 43

Figure 22: loomed-in structure of the infrared signal ...................................................... 44
Figure 23: loomed-in structure of an infrared signal (2) .................................................. 44
Figure 24: Ultrasound Signal structure .............................................................................. 45
Figure 25: Zoomed-in structure of an US signal ............................................................... 45
Figure 26: Recursive Error caused by delay routines at the beacon .................................. 46

Chapter 1: Introduction

This chapter gives the background and motivation for this thesis work. An effort is
made to lay the foundation for consequent topics. It also describes the contributions that

this thesis work has made.

1.1 Background

Localization is a mechanism that is used to remotely find the location of objects,
either indoors or outdoors. It has been a challenge for researchers to devise means of
telling the exact location of men and their belongings. Up to date, there continues to be
tremendous breakthroughs into this field. Currently, through the Global Positioning
System (GPS), it is possible to locate a person within a few meters of accuracy. This
technology has found many applications such as vehicle navigation and tracking of
people and animals.

The GPS and RADAR systems have been employed to solve user location and
tracking in outdoor scenarios. However, it is becoming more and more evident that
indoor location and tracking systems are needed for numerous applications. The
conventional means of outdoor location cannot be employed for indoor purposes due to
obstacles found inside buildings, which reflect and attenuate data signals [1]. It is
therefore necessary that systems that are Specifically suited for indoor localization be
developed. Some of the applications of indoor location systems are:

0 Localizing visually impaired navigators

l

0 Navigation tools for humans and robots

0 Finding critical personnel or resources faster

0 Asset tracking

0 Location-aware sensor networking

0 Simplifying the user interface for mobile voice and data functions
0 Improving the security and effectiveness of Wi-Fi networks

0 Restricting online shopping to certain people in a certain room

0 Improvement of roaming capabilities.

There have been many solutions suggested for indoor localization, some of which
are: In-building RADAR [5], Active Bat Location system [10], Active Badge Location
System [3], HiBall Head Tracking system [6], Ubisense Location system [11], Broadband
Ultrasonic location system [2], MIT Cricket [1], Bristol Indoor Position system [7], and

Nibble location system [9].

1.1.1 Indoor Location Systems

Indoor localization involves objects finding their position in reference to other
objects in an indoor scenario. The reference Objects may be stationary or mobile. The
Object seeking to know its position and the reference Object may be active or passive. In
case they both have to be active, some synchronization and scheduling mechanisms are
needed. In many cases, the Object seeking to know its position is passively listening to
messages sent from the reference point. In this case, the synchronization step is

eliminated, but scheduling has to be implemented, especially if there has to be multiple

transmitters.

1.1.2 Indoor Location System Application Example

Figure 1 shows a possible application of the IRUS indoor location system. The
robot can be programmed to move within the boundaries demarcated by beacons 1 & 2.
For instance, the beacons can be used to represent edges of a table or any elevated surface

from which the robot Should not topple and fall. Simply put, the robot sees its boundaries.

 

 

 

 

 

 

 

I I | IR‘I'X Beacon 1|

 

 

 

 

 

 

 

Host Controller I
I us 1x
3 I9——-Boundary
RObOt IR RX |
<1 I
03 RX I
O O “mam,“
l>

 

 

 

—————————JUSTX

 

 

 

Figure 1: Example of an indoor localization application

The indoor localization devices can be distinguished by the media used for data
exchange and the algorithms used to achieve localization. The media that have been
explored so far are Radio Frequency, Infrared, and Ultrasound. Some of the algorithms

used are: Angle of Arrival (AOA) or Direction of Arrival (DOA) [12], Time of Arrival

(TOA) [10], and Time Difference of Arrival (TDOA) [1].

The indoor localization solution presented in this thesis work is based on time
difference of arrival (TDOA) of infrared and ultrasound signals. There are two modules
in the system designed; a transmitter (beacon) and a receiver (listener). The beacon sends
an infrared Signal followed by an ultrasound signal. The listener detects these signals and
computes the difference in their times of arrival. This difference is then multiplied by the
speed of sound to obtain the distance separating the two modules.

Time difference Of arrival has been employed in the MIT Cricket system [1,2].
This system uses RF and US signals to achieve distance computation based on TDOA.
We introduce the use of infiared because of its inability to penetrate walls, therefore

reducing interferences to neighboring rooms.

1.2 Motivation

Indoor environments have different characteristics. To design an indoor location
system, this factor should be considered. We had this consideration in mind when
designing the system in this thesis. We wanted to explore an alternative medimn of data
transmission, other than RF. IR was opted for because of its inability to penetrate walls.
This feature would ensure protection of user privacy and avoid interferences to

neighboring rooms.

1.3 Contributions of the Thesis

In this thesis, a new method Of indoor localization is presented. This method
involves periodically sending an infi'ared signal followed by an ultrasound Signal. The
receiver of these Signals (listener) records their times of arrival and based on the time
difference of arrival (TDOA) [1] algorithm running on it, it calculates the distance
separating it and the signal sender (beacon).

We developed a prototype of the indoor location system using off-the shelf
components. We then carried out experiments to characterize and evaluate the
performance of the system under different indoor conditions including ambient lighting

and temperature.

1.4 Thesis Organization

This chapter of the thesis presented a background of indoor location systems.
The next chapter is a survey of related work in this area of networking. Chapter three
gives a detailed description of the concept behind indoor localization presented in this
thesis. The fourth chapter presents a detailed description of the hardware as well as the
software architecture of the indoor location system designed in this thesis. Chapter five
presents the characterization procedures, followed by the results. It also outlines the
challenges encountered and how they were overcome. The sixth chapter presents the
conclusion of this work and the seventh chapter presents recommendations for future

work.

Chapter 2: Related Work

This chapter describes some of the existing indoor location systems. Research on
indoor location systems is a fast growing area and this chapter is in no way exhaustive as
far as their coverage is concerned. However, an attempt is made to cover as much a wide

area as possible, in this dynamic research area.

2.1 The MIT Cricket

The MIT Cricket [1, 2] is an indoor location system based on time difference of
arrival of radio frequency and ultrasound signals. It is composed of beacons and listeners.
Beacons are strategically placed, either on ceilings or on walls. They periodically
transmit RF Signals with id information. At the same time, an ultrasonic signal is
transmitted. At the listener, the RF signal is received first and its time of arrival noted.
The ultrasound Signal lags behind the RF signal, and this lag period is determined at the
listener. It is then multiplied by the speed of sound to calculate the distance between the
beacon and the listener.

The Cricket system is distributed and scales well Since the listeners are passive.
However, there is a likelihood of RF Signals leaking to neighboring rooms. The system
suggested in this thesis is similar to the cricket system in terms of distance determination.
However, instead of RF, we use infrared. The advantages that arise from this change

were discussed in section 1.2.

2.2 Broadband Ultrasonic System

This system, developed at the University of Cambridge, can be seen as a way to
enhance the Cricket system by increasing the transmission rate in the Ultrasound channel.
Unlike the Cricket system and the ranging system in this thesis work, the Broadband
Ultrasonic system incorporates data into the ultrasonic signal.

The Broadband Ultrasonic location System is polled and centralized [2]. When a
system is termed as polled, it means that the transmission is coordinated. The
centralization property means that a centralized system exists to collect and analyze data
from the nodes.

To determine the location of a node, times of flight of messages between
transmitters and receivers are used to compute the distances between the transmitter and
receiver by multiplying times Of flight by the speed of sound in air.

The weakness of the Broadband Ultrasonic System can be seen as a lack of user
privacy due to the use of a centralized system to collect and analyze data; and the huge

amount of energy used to achieve the broadband data transmission.

2.3 The Active Badge Location System

The Active Badge Location System [3] is based on badges that transmit signals
with information about their location to a centralized system. A sensors network is put

around a host building, and it picks up signals that are periodically transmitted. These

badges are worn by people in an area, whose locations need to be determined. The signals
transmitted are infrared signals.
The drawback of the Active Badge is its user privacy compromise since it is based

on a centralized system.

2.4 The Bat Ultrasonic Location System

In order to improve the Active Badge system, the Bat Ultrasonic Location System
[4] was developed. Unlike the Active Badge System, this location system is able to
provide location and orientation information in 3D. To Obtain location information, times
of flight of an ultrasonic signal between a transmitter (Bat) and the object to be located;
are determined and these are multiplied by the speed of sound to calculate the distances
from the Bat to each receiver. With three or more bat-receiver distances, enough
information is available to determine the Bat’s 3D position. The orientation of an object
can be calculated by determining the relative positions of two or more Bats that are
attached to that object.

The disadvantages of relying on ultrasound to transmit data still apply to the Bat

Ultrasonic Location System.

2.5 In-Building RADAR

This indoor location and tracking system is RF-based [5]. Information describing

signal strength is theoretically computed and empirically-determined at multiple receiver
8

locations. This information is then used to triangulate the coordinates of the user. This

system suffers from effects of radio channel interferences, which reduce its accuracy.

2.6 The HiBall Tracking System

This electro-optical system was built for high precision head tracking for virtual
reality applications [6]. It consists of panels of LED’s that are flashed sequentially, head-
mounted cameras that determine the position of the flashing LED’S, and a computer
system that uses the knowledge about the cameras’ coordinates to Obtain the location

information. The drawback of this system is its expensive implementation.

2.7 Low Cost Indoor Positioning System

This system, developed at the University of Bristol-UK, is based on radio
frequency and ultrasonic Signals [7]. The RF signal is used to synchronize the
transmitters and the receiver. Four ultrasonic transmitters are placed on the ceiling of the
experimental room. The Signals sent are detected and their times of flight are recorded at
the ultrasound receiver. The times of flight are factored with the Speed of sound to
determine the distances between the transmitters and the receiver. Four times of flight are
used to ensure that the system’s range is increased and that signals that were lost are

compensated.

The synchronization mechanism used in the low cost indoor positioning system
increases the cost of the system. Since both the transmitter and receiver have to be active,

this system fails to scale very well.

2.8 The Horus WLAN Location Determination System

The Horus location system [8] is based on RF. Signal strengths of frames
transmitted by the access points are used to provide user location. The system is currently
implemented in the 802.11 wireless LANS context. Since it is based on the already
existing wireless LANS, it is seen as a software solution built on top of the wireless
infi‘astructure. Just like other WLAN technology location systems, Horus works in two
phases. The first phase is referred to as oflline phase. In this phase, data representing
signal strength is collected from points of access and tabulated into a radio-map. The
second phase is called the location determination (online) phase. Here, the system
searches the radio-map based on newly received signals from the access points to
estimate the user location.

The drawback of the Horus WLAN system is in the fact that there can be
interferences from neighboring rooms or other RF based devices, which would result to

incorrect samples in the oflline phase.

10

Chapter 3: Concept

This chapter presents the concept behind the system designed in this thesis. It
explains the principle behind ranging and distance computation based on time difference

of arrival (TDOA) of infi'ared and ultrasound signals.

3.1 Ranging

Indoor location systems have ranging as their core function. Ranging involves
determining the distance separating one point from another. To achieve this, various
techniques can be employed.

Most organisms are cognizant of their surroundings through eyesight. Eyesight
can judge distances and tell the organism how far it should fly, run, fall, and so on.
Eyesight is in essence a ranging mechanism based on reflection of light off of objects to
the retina of the eye. The brain then processes this information and the organism is able
to interpret it and is therefore able to navigate its surroundings. Some indoor localization
systems mimic the eyesight mechanism. They send a light signal e. g. infi'ared, which
bounces off of an object. The reflection is then interpreted by a microprocessor of the
receiving device. The information extracted can tell a robot, for instance, how far or near
a boundary it should navigate.

The bat is a blind, flying mammal and its ranging mechanism has inspired many
commendable discoveries. It sends out ultrasonic Signals that are reflected Off of its

surroundings and fall on its ears. Its brain then processes this information and it is able to

11

judge its surroundings at amazing precisions. This idea has been employed in ranging
systems. For instance, ultrasound based ranging systems use echoes to calculate
distances. A node seeking to know its position sends out an ultrasound signal which is
reflected off a reference point. The transmitting node has to record the time that the signal
was sent. When the signal reaches the reference point, it is reflected back to the
transmitter, which records the time of arrival of the reflected signal. The difference in the
departure and arrival times corresponds to the distance separating the transmitter and the
reference point. This difference is then multiplied by the Speed of sound to obtain the
distance separating the node and the reference point. Figure 2 represents ultrasonic

ranging.

 

 

 

 

 

Q. Signal
Echo

Transmitter G— Distance _____._, Reference

 

 

 

 

 

 

 

 

 

Figure 2: Ultrasonic Ranging

If there are two systems that have a synchronized clock, either of them can tell the
distance from the other by noting the time of departure and arrival, and then multiplying
the difference by the speed of the medium of propagation. The transmitter has to send the
packet with time of departure information. The receiver will then extract this information

and read its clock for the time of arrival.

12

The preceding two methods have their strong as well as weak points. The
echo/reflection based ranging is simple and inexpensive but suffers from interferences.
For instance, any object that can reflect the signal used can be confused for the reference
point, thus leading to erroneous computations of distances. The second method that
involves synchronization ensures that distances are computed only at the correct
reference points, but it is costly and has a limitation as far as scaling is concerned.

Suppose two signals moving at different speeds left a transmitter at the same time.
They would reach a reference point at different times. The reference point would then
take note of this difference in times of arrival and use them to calculate the distance from
the transmitter. This method of ranging is based on a technique called “Time Difference
of Arrival”-TDOA [1]. TDOA avoids the necessity to synchronize the system clocks,
and can therefore be less costly than the synchronized system. It also scales well since the
transmitter is active but the listener is passive. Section 3.2 gives a detailed description of

this method of ranging.

3.2 Ranging based on TDOA of IR and US Signals

The solution suggested in this thesis is based on ranging using the TDOA
algorithm [1], based on infrared and ultrasound signals. The TDOA technique has been
used in the MIT cricket [1, 2], where radio-frequency was used together with ultrasound.

We propose the use of IR due to its following advantages:

13

0 Minimal interference to/from other rooms, since IR transmissions cannot
penetrate walls. This feature also provides user privacy.
0 Unlicensed data transmission, allowing for flexibility of experimentation
and prototyping
o No requirement for an antenna which is a cost and Size issue in RF
technologies
In the system described in this thesis, ranging is achieved using TDOA of infrared
and ultrasound Signals. The beacon periodically transmits an infrared signal followed by
an ultrasound Signal. These signals are detected at the listener. The speed of light and that
of sound are known. These values can then be used, together with the times of arrival, to
determine the distance between the beacon and the listener.
Let the Speed of IR be Vi, and that of sound be Vus. Since Vi, is greater than Vus,
the ultrasound signal lags behind the infi'ared signal as they move from beacon to listener.

The listener determines the time lag, ST, and can calculate the distance D, from:

D D

_1_ __1_(1)I11

Vus Vzr

Under normal conditions, the speed of sound is approximately 344m/s, and that of light is

5T=

3x1 Osm/s. Since Vir>>Vus,

Dz 5T. V“, (2) [1]

l4

3.3 Factors Affecting Distance Computation based on TDOA of IR and
US signals

AS shown in equation 2, distance computation depends on the accurate
determination of OT as well as the speed of sound. The accurate determination of ST in
turn depends on the accuracy of the time recording algorithms running on the listener, as
well as the distance computing algorithms.

On the other hand, the speed of sound depends on environmental factors such as
relative humidity, temperature, and atmospheric pressure. For example, At 25 0C and
101.325 kPa (atmospheric pressure at sea level), the speed of sound changes by only
about 0.5% as relative humidity changes fi'om 0% to 100% [1]. At 25 OC & 50% relative
humidity, the speed of sound changes by only about 0.6% as the atmospheric pressure
changes from 101.325 kPa to 30 kPa (atmospheric pressure at the top of Mount Everest)
[1]. Therefore, atmospheric pressure and relative humidity have limited effect on the
speed of sound.

In air, the speed of sound changes by 0.18% for every 1 0C change at 25°C [1].
This change is approximately equal to 60cm/S change in the speed Of sound for every 1
0C change in temperature. This property necessitates compensation. To monitor
temperature change and adjust the Speed of sound appropriately, we used a temperature
sensor. For experimental purposes only, the temperature sensor is only included in the

listener circuitry.

15

3.4 Node Localization Mechanisms

To find the exact location of Objects, we would need several nodes to communicate. For
example, figure 3 shows node localization based on coordinate axes. The axes can
represent walls, floor, or ceiling of a room. Figure 4 Shows localization based on inter-
node distances. All the nodes can have information about distances to neighboring nodes.

Distances from three nodes can give user location as well as orientation information.

 

Y
/\
O
O
O
6% 0x1 ,y‘l
jun

 

>X

Figure 3: Node coordinate assignment fi'om coordinate axes

Y
/\

Vt

 

 

>x

Figure 4: Node coordinate assignment from inter-node distances

l6

Chapter 4: System Architecture

This chapter outlines the hardware as well as the software architectures of the
IRUS wireless indoor location system. It also gives a detailed analysis and description of
the system’s configuration and parameters.

Figure 5 shows the modules of the IRUS indoor location system designed in this
thesis and figure 6 is a depiction of its organization. The Beacon [1] which is the
transmitting node can be located on the ceiling, elevated points in a room or on the wall

while the listener [1], which is the receiving node is attached to a host PC or device that

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

needs to be localized.
———I l
Host PC
0E]
LiStOIIOI' m RX IR TX Beacon
<1
us RX [j
us Dr

 

 

 

Figure 5: Modules of the IRUS Indoor Location System

Since the listener is passive, this system can scale very well and also ensures that

user privacy is protected. Both infrared and ultrasound do not penetrate walls. Therefore,

interferences to/ from neighboring rooms are avoided.

17

 

US (IRUS)
l

L 1
Hardware Software
system system
I I
f l I l
[ Listener ] [ Beacon ] [ Listener ] [ Beacon ]
IR Receiver IR Transceiver TDOA & Signal
connected as Range generation and
TX computation transmission
US Transducer US Transducer Output program
(RX section) (TX section)

Figure 6: Components of the IRUS indoor location system

Location system
based on IR and

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

4.1 Hardware Configuration

This thesis work was hardware intensive. Therefore, a significant amount of time
was spent on the hardware design. The system was assembled on a proto-board, using

off-the Shelf components.

18

4.1.1 Overview of System

Figures 7 and 8 Show block diagrams of the listener and beacon used in the indoor
location system based on Infrared and Ultrasound signals. Each module consists of a
microcontroller, ultrasound transducer and supporting circuits, and infrared devices.

The beacon has the TFDU4300 IR transceiver configured as an IR transmitter, while the
listener has the TSOP381 for IR signal reception. The infrared transceiver is connected to
the microcontroller via an external UART. The external UART used is MAX3100.

The microcontroller used is PIC16F874, and runs at 20 MHZ. For lower power
consumption, a slower clock can be used. This microcontroller was picked due to its
Simplicity and inexpensiveness. It is based on the RISC architecture and has only 35
single word instructions.

The Ultrasound transducer used is similar to the one used in the MIT Cricket [1,
2]. Modifications were made to the supporting circuitry to suit our application. See
schematic in appendix A.

To communicate with the host PC, the listener runs the R8232 protocol, taken
care of by the R8232 chip. The listener sends data to the host at a baud rate of 9600bps.
Only the listener has a temperature sensor. This option was taken so as to simplify the
design of the experimental system. The temperature sensor will be incorporated in the
beacon architecture once the beacon has been programmed to send an IR message. The
temperature sensor used is LM34DZ, which is a precision Fahrenheit temperature sensor.
Its output voltage is linearly proportional to the temperature in Fahrenheit. This

temperature sensor was configured as a basic Fahrenheit temperature sensor and

19

calibrated to measure temperatures in indoor scenarios as low as 50°F and as high as

89°F.

 

 

Microcontroller

(PIC16F874)
IR Receiver
RS232 Connector

Vcc o—/
Amplitude I +

> a Detector 5V "

 

 

 

 

 

 

 

 

 

 

US Receiver

 

‘2‘ Temperature Sensor

 

 

 

Figure 7: Block diagram of listener [1]

For experimental purposes, only two modules, one beacon and one listener, were
built. Therefore, since there was no risk of sending a signal fi'om the wrong beacon, the
listener was not programmed to process the message sent by the beacon. Every signal

from the beacon is Simply taken as a pulse that triggers an interrupt at the listener.

20

 

 

 

mil:
IR

Transceiver

(TFDU4300)

PD

 

 

 

 

Microcontroller
(PIC16F874)

 

 

 

 

Vcc o—-/-—

 

 

 

Voltage
multiplier

 

 

 

 

 

 

Ultrasonic
driver

 

 

US transmitter

 

Figure 8: Block diagram of beacon [1]

4.1.2 IR transceivers

The TFDU4300 Infrared transceiver was used as an IR transmitter at the beacon.
Figure 9 shows the beacon’s infrared hardware section. The microcontroller generates
signals to configure the external UART to transmit infiared data according to IrDA
standard at a data rate of 9600 bps. The external UART (MAX3100) receives data fiom
the Serial Peripheral Interface (SPI), formats it to IrDA, and sends it to the Infrared

transceiver, which in turn transmits the infrared signal to the infrared receiver on the

listener side. MAX3100 was designed to directly drive optocouplers, whereas IrDA

modules have inverting buffers. This feature calls for inversion of the TX and RX signals.

A NAND gate was used for this purpose.

TFDU4300 is compliant to the IrDA standard and can also be configured to

transmit data in the remote control mode. The transmitter has an output radiant intensity

21

 

equal tO 65 mW/sr and a peak emission wavelength Of 880-900 nm. Its spectral
bandwidth is 45 nm. It has an optical rise and fall time of 10-100 ns. For an input pulse
width of 1.63 us, the optical output pulse duration is 16-1 .8 us. For an input pulse width
greater than or equal to 20 us, the optical output pulse duration is 20-300 us. In all these
cases, a data rate of 115.2 kbps is assumed [33].

For short distance transmissions, TFDU4300 can also be used as a receiver. In
SIR mode the receiver has a minimum detection threshold irradiance of 40-80 mW/m2.
Its maximum detection threshold irradiance is 5 kW/mz. The rise and fall times of the
output signal are 10-100 ns. The Rxd pulse width of the output signal for an input pulse
length greater than 1.2 us is 1.65-3.0 us. Its stochastic jitter at the leading edge is 250 m
at a data rate less than or equal to 115.2 kbps [33].

For long distance transmissions, TSOP341 IR receiver module was used. Its
typical transmission distance is 45 m. Its minimum irradiance is 0.1-0.25 mW/mz, while

its maximum irradiance is 30 W/mz. Its directivity is :450 [32].

4. 1.3 US transducers and Supporting Circuits

At the beacon, the microcontroller produces a periodic wave, at 40 KHz through
its pulse width modulation mode (PWM). This signal iS then amplified so as to drive the
ultrasound transducer. The ultrasound transducer used is the 255-400 series where the
transmitter is 255-4OOST12 and the receiver is 255-4OOSR12. It has a center frequency of
40 KHZ 11.0 KHZ. The sensitivity of the receiver at the center frequency is -67dB. The

minimum driving voltage is 1V, while the maximum is 20V. The sensitivity of the

22

receiver at the center frequency is -67dB. The minimum driving voltage is 1V, while the

maximum is 20V. The transducer’s bandwidth is 2 kHz. [30].

Typically, the ultrasound signal that triggers an interrupt at the listener is

approximately 5V. The amplification of the ultrasound signal to this level is done at the

listener. The amplifier circuit has a potentiometer that can be varied manually to adjust

the sensitivity level of the circuit to the incoming ultrasound signal. This potentiometer

can be adjusted for short distance range computations. See schematic in appendix A.

 

Plc1 6F874
R132 35

RBI 3‘

RED/INT 33

vss 31
RD7/PSP? 1
RD6/PSP6 29
RDS/PSPS 2"
RD4/PSP4 27

RC7/RX/Drr 2“
RC6fI'X/CK 25
RCS/SDO 2‘

 

 

RC4/SDL’SDA 23 H

 

 

 

 

 

MAX31 00

 

 

 

c1=c2=22 pF

 

V00

D—RXD

 

F533 51':

 

TXD

 

 

IREDC

 

TFDU4300

I: 1.8432Mhz:

 

c1

 

 

Figure 9: Infrared Section of the beacon

23

4. 1.4 Microcontroller

PIC16F874, the microcontroller used in the design Of the IRUS location system, is
an 8-bit microcontroller. It has up to 8K x 14 words of FLASH Program Memory, up to
368 x 8 bytes of Data Memory (RAM), and up to 256 x 8 bytes of EEPROM Data
Memory. It is able to handle up to 14 sources of interrupts. Its hardware stack is 8 levels
deep. It supports direct, indirect and relative addressing modes. It features a power saving
SLEEP mode. Its operating voltage range is 2.0 V-5.5 V. Power consumption can be as
low as <0.6mA, operating at 3 V and 4 MHZ. Operating at 3 V and 32 kHz, a 20 [IA low
power consumption is feasible. The typical standby current is <1 uA. [34, 35]

The peripheral features of PIC16F 874 were exploited to achieve the system
functionalities of the indoor location system described in this thesis.

On the beacon side, timer2, an 8-bit timer with an 8-bit period register, prescaler
and postscaler; was used to control the PWM module. The resolution of PWM can be as
high as 10-bits. The PWM mode enables the generation of pulses of various periods and
duty cycles.

The Synchronous Serial Port (SSP) with Serial Peripheral Interface (SPI) was
used in the master mode to generate serial data that was converted to IrDA by an external
UART, MAX3100, thus making transmission of infrared Signals possible.

On the listener side, timerl, a 16-bit timer/counter with a prescaler, was used
together with an external crystal; to keep time of the system. This timer is the one that

enables recording of the times of arrival of signals to be done. Reading timerl takes place

24

in two steps: reading the high byte and reading the low byte. In order for the time
information to be correct, reading must take place before or after rollovers.

Readings between rollovers lead to erroneous computations. At 20 MHZ, a 16-bit
counter will roll over every 13ms. This property calls for the implementation of a rollover
counter that Should also be read every time timerl is read. A 32.768 kHz crystal
oscillator was used to control the timerl as a timer in asynchronous mode. In this
configuration, the microcontroller can actually keep real time and the need of a rollover
counter is avoided. However, since the timer is read in two stages, it is still possible to
read a value between rollovers of the lower byte of the timer. This error is checked in
software. Section 3.2.2 gives a detailed description of how this is achieved.

The microcontroller also features an analog-to-digital converter (ADC) which is
lO-bit and is multi-channel. This feature is the one that was employed for temperature
sensing. The output of the temperature sensor is connected to one of the ADC input pins
and the voltage value present on the ADC pin is converted to a digital value, which is
then used to represent the ambient temperature in the lookup table implemented in
software.

The Universal Synchronous Asynchronous Receiver Transmitter (U SART) was
used to send data to the host PC. It was configured as an asynchronous, high speed
transmitter, transmitting data at a baud rate of 9600 bps.

PIC16F 874 supports In-Circuit Serial Programming (ICSP). This mode of
programming was used to download programs into the microcontroller. The programmer

used was the MPLAB In-Circuit Debugger (ICD 2).

25

4. I. 5 System parameters

This section outlines the parameter values of the various components of the IRUS

wireless indoor location system.

4.1.5.1 Beacon

Table 1 shows the parameters of the transmitting module (beacon). Here, beacon
frequency refers to how often the beacon transmits signals to the listener and it is shown
to be 1 second. The maximum clock speed was used in the prototype, but a lower Speed
can be used for the clock and this can result to lower power consumption. The IrDA baud
rate used is the rate that showed best compliance with the external UART (MAX3100).
The ultrasound pulse duration used was obtained through experimentation and it gave the

best performance for most distance values.

 

 

 

 

 

 

 

 

Parameter Value Description

Beacon Frequency 1 Hz Beacon sends signals every second

Microcontroller clock Speed 20 MHZ Maximum clock speed, resulting to 200nS
instruction cycle

IrDA baud rate 9600 bps Standard IrDA data rate

IR wavelength 880-900ns Typical IrDA wavelength

US frequency 40 KHZ Optimum response frequency of the
transducer used

US Pulse duration 1500 us Optimum duration for proper US Signal
detection at the listener

 

 

 

Table l: Beacon parameters

26

 

4.1.5.2 Listener

The listener parameters are shown in table 2. The clock speed and the US

transducer frequency are similar to those of the beacon. The listener features a precision

temperature sensor that is able to measure temperatures to _+_1_°F. Temperatures likely to

be measured in an average indoor location were considered when calibrating the

temperature sensing mechanism of the listener. This feature can be readily modified to

cover the full range of temperatures.

 

 

 

 

 

 

 

 

 

Parameter Value Description

Microcontroller clock Speed 20 MHZ Maximum clock Speed, resulting to 200ns
instruction cycle

US frequency 40 kHz Optimum response frequency of the
transducer used

US RX gain 70-78 db Detection of the US Signal at approx. 8 m
when the modules are directly opposite each
other [1].

Temperature sensor range 50—89 0F Range of temperatures that are likely to be
found in indoor environments

R8232 data rate 9600 bps Standard data rate

 

Table 2: Listener parameters

4.2

Software Configuration

The programs that run on the beacon and listener were written in assembly

language, and tested using the MPLAB IDE. Codes were written, and compiled using the

MPASM feature of the MPLAB IDE. Simulation followed using the MPLAB simulator.

For supported functions, external stimuli were introduced to simulate system

27

 

performance. However, some features Of the microcontroller cannot be simulated. For
instance, currently, the real-time clock mechanism, serial-peripheral interface, cannot be
simulated, thus the code for those functions had to be tested on the physical system. After
the necessary simulations and debugging, the resulting hex files were downloaded into
the microcontrollers.

Figure 10 represents a summary of the initializations done in software for both the
beacon and listener. Sections 4.2.1 and 4.2.2 give a more detailed description of what

goes on inside the microcontrollers of the beacon and the listener.

4. 2. I Beacon Program

The beacon runs a program that periodically sends an infrared signal followed by
an ultrasound signal. The initialization part of the code involves configuring the
microcontroller to send serial data via the Serial Peripheral Interface port to the external
UART. The microcontroller also generates the configuration word for the external
UART. The external UART converts the serial data to IrDA data, which is then passed on
to the IR transmitter.

Initialization also involves enabling the pulse width modulation mode and
produces a wave of approximately 40 kHz for 1500 us, which is the Ultrasound signal.
To control the duration of the ultrasound signal, the duty cycle is defined as 50% for
1500 us and then it is defined as 0% for the duration within which the infrared signal is

generated.

28

The repetitive loop of the beacon program involves calling for the serial data
signal to be transmitted as infrared, followed by the ultrasound signal. This sequence is
repeated every second.

The initialization routine takes approximately 17.6 ps. The infiared signal
generation routine takes 9.4 ps, after which the ultrasound Signal is generated. This delay
is very negligible and it is therefore assumed that the signals leave the beacon at the same

time.

 

 

Beacon Listener

 

 

Enabling of R8232
interrupts Protocol

 

Figure 10: System Initializations

4. 2.2 Listener Program

The listener runs a program that detects the signals transmitted {Tom the beacon;
records their times of arrival and determines the range based on the ranging protocol. The

initialization part of the code configures the timerl for asynchronous, real time keeping.
29

It also enables the necessary interrupts and it is here that error correction mechanisms are
implemented.

The initialization part of the code forms the repetitive loop Of the listener
program. When a signal is detected, an interrupt is generated and the execution of the
program now goes to the Interrupt Service Routine (ISR), to service the interrupt
generated.

The infrared signal is detected at the interrupt-on-change pin. As soon as a valid
rising edge appears on this pin, the point of execution in the repetitive loop is recorded
and the working and status register values are saved. The interrupt service routine is then
invoked. Here, timerl values are read and recorded as the times of arrival of the infrared
signal. A mechanism exists to check that if the low byte of timerl is read between its
rollovers, timerl is read again to ensure a correct time value. When timerl has been read,
the microcontroller sets a bit that will be used to check for erroneous computations. This
bit is referred to as free and it is in the error_check file register. It then lights a diagnostic
LED to Show that the arrival of infrared has successfully been reported. The working and
Status register values are then restored and the program counter resumes at the point
where the interrupt occurred.

The ultrasound signal is detected at the external interrupt pin of the
microcontroller. As soon as there is a valid rising edge at the external interrupt pin, the
working and status register values are saved and the program counter jumps to the start
point of the interrupt service routine. Timer] values are read, checked for rollover of

TMRI L, and recorded as times of arrival of the ultrasound signal.

30

The ultrasound Signal used is 1500 us long, while the infrared signal is 750 us
long. This means that there can be multiple ultrasound interrupts just before another
infi'ared of another valid sequence of signals is noted. This is where the flee bit that was
set after reading the time Of arrival of the infrared Signal comes in handy. After the timerl
values are read at the arrival of the ultrasound signal, the flee bit is checked. If it is found
set, the program goes ahead with distance computations.

Distance computations involve finding the time difference of arrival (TDOA) and
multiplying it by the speed of sound. It also involves reading the value of the temperature
sensor and carrying out the necessary compensations. After the distance values have been
sent to the host PC, an ultrasound diagnostic LED iS lit. Then, the normal procedure to
exit the interrupt service routine is executed.

If the flee bit is found clear, the distance computation routine is skipped. The
ultrasound diagnostic LED is lit and the interrupt service routine is exited. The checking
Of the flee bit ensures that no erroneous distance computations are returned. Just by
looking at the diagnostic LED’S, we cannot tell whether a valid sequence of IR and US
signals has been detected. However, the error checking mechanism ensures that only
those ultrasound Signals that come after infrared Signals contribute to distance
computations. In other words, distances are computed when and only when an infrared
signal followed by an ultrasound signal event is detected at the listener.

Appendix B Shows the source code of the programs used in the beacon and

listener.

31

Chapter 5: System Characterization

This chapter describes the experimental setup and gives results of the
experimental system characterization as well as the analysis of the performance results. It
also describes the challenges encountered and suggests possible solutions to these

challenges.

5.1 Experimental Setup

For characterization, two modules were used. The beacon and listener were
placed as shown in figure 11. The angle 0 was varied and then distance was computed.
The distances computed were recorded at the listener over a period of two minutes. With
the beacon sending signals every second, this amounts to a total of 120 samples of
distances, from which errors were computed. This amount of samples is a best case
scenario where all signals sent contribute to distance computation. However, as will be
seen in the challenges section, it is not all signals from the beacon that contribute to
distance computations. The measured distances were then used to obtain the percentage

CITOI‘S.

5.2 Performance Results

This section shows graphs that were Obtained from the various system
performances. The graphs show percentage errors resulting from the measured distances

under different conditions.

32

 

\
d\e

/

Listener

 

 

Beacon

 

Figure 11: Experimental Setup [1]

5. 2.1 Effect of distance and angle of rotation

Figure 13 shows the percentage error in computed distance with increase in
distance and change in the angle of rotation. The graph Shows that error in distance
computation is minimized when the beacon and listener are directly Opposite each other
(angle of rotation=0 degrees). As the angle of rotation increases, the ultrasound signal
strength at the listener decreases, and it takes the listener a long time to detect it, thus
contributing to the error increase. This property can be attributed to the nature of
ultrasound transducers used in the system. As can be seen in figure in figure 15, the
ultrasound Signal transmitted can only be detected at a limited range of angles of rotation.

AS the nodes move away from each other, the ultrasound Signal is attenuated, and
it takes the listener a long time to detect it, thus contributing to the error increase in

computed distances [1].

33

 

 

 

 

 

 

 

Figure 12: Directivity of US transducers used in the system; directly taken fiom the
datasheet [30]

5. 2.2 Effect of ambient light on computed distance

The beacon and listener were placed directly Opposite each other (0 degrees) and
the distance separating them was varied. Three such experiments were performed under
different ambient light conditions: a lit lab during the day, a dark lab, and outdoor. The
graph shown in figure 14 below was then plotted from the results.

Outdoors, the system performed poorly and the results of the experiments were
not consistent. This poor performance can be attributed to more interference in outdoor
environments such as other sources of ultrasound. When we compare the errors observed
in the lit and dark labs, we can conclude that ambient light has negligible effect on

system performance. The infiared transceivers used are properly Shielded from other light

34

source interferences, hence the consistent performance in different ambient light

 

 

 

 

 

 

 

 

 

conditions.
14
+50cm +100cm
+200cm +gggem
12 . +400cm + cm
.0. ‘\
E 8 ‘
m
=3 6 .
4 _
‘\‘\I’,’
2 - v e e e e H e e e e v
o I I I I I I T I I T T T
-90 -75 -60 -45 -30 -15 0 15 30 45 60 75 90
Angle of rotation

Figure 13: Percentage error in computed distance with increase in distance and angle of
rotation

5. 2.3 Effect of temperature on system accuracy

The speed of sound changes by approximately 60cm/s for every one degree
Celsius increment in ambient temperature [29]. This property can introduce errors in
distance computations. For this reason, a temperature sensor was incorporated in the

system to help the system adjust the Speed of sound according to the ambient

temperature.

35

TO show the effect that temperature change can have on the system accuracy,
distances were computed at various temperatures, first without the temperature sensor
(with a fixed speed of sound=344m/s), and then with the temperature sensor. The graph
shown in figure 15 was then plotted. The graph shows that compensations with
temperature change help to reduce errors in computed distance, by an average of

approximately 7cm.

 

3.5

 

+ Lit Lab
+ Dark Lab

 

 

 

3.0 —

1.0 -

0.5 -

 

 

 

50 100 200 300 400 500
Distance In cm

Figure 14: Error in computed distance in different light conditions

5. 2.4 Comparisons of errors at short distances between the Cricket and the IR US
The IRUS indoor location system can be adjusted for short distance computations by
adjusting the ultrasound Signal strength at the beacon and listener. After this adjustment

36

was made, the IRUS accuracy with measured short distances was compared with that of
the Cricket system. This comparison is shown in figure 16.
The graph shows that for most Short distances measured, the IRUS system has a

percentage error less than 10.

5. 2. 5 Comparisons of errors at long distances between the cricket and the IR US

The IRUS system was adjusted for long distance computation and the resulting
errors were compared to those claimed by the cricket team. The resulting graph (figure
17) shows an error difference of about 1.2% between the two systems. The ultrasound
signal is greatly attenuated beyond 8m and therefore no distances were computed beyond

this value.

From the two preceding graphs (figures 16 & 17), we conclude that the IRUS
system performs within the expected ranges of an indoor location system such as the MIT

Cricket.

5. 2. 6 Error in computed distance observed at the oscilloscope

With the beacon and listener directly opposite each other, signals sent were observed
at the oscilloscope, as shown in figure 18. The time difference of arrival was obtained

and multiplied by 344 (speed of sound at room temperature).

37

The resulting errors in computations of various distances were then plotted against the
measured distances, as shown in figure 19. The graph Shows that at long distances, there
was an error increase. The distance computations were affected by the scope’s resolution,
hence the Observed increase in error in computed distances. Also, by looking at the
signals arriving at the listener, it is impossible to tell exactly what edge causes an
interrupt, hence the possibility of errors. The results from this experiment also

demonstrate that the routine running on the listener can be trusted.

 

4.0

 

+ wlsensor
3.5 . +wlo sensor

   

 

 

 

 

 

 

0.0 r . .
200 400 600 800
Distance In cm

Figure 15: Error with distance increase Shown with and Without temperature
compensation

38

7.0

 

 

+ Cricket
6.0 - + IRUS

 

 

 

5.0 ~

4.0 ~

% error

3.0 -

2.0 .

1.0 r

 

 

 

50 100 150 200
Distance in cm

Figure 16: Error with computed distance at short distances

5.3 Challenges

Figure 20 is a representation of both Signals as they leave the beacon. The figure
shows a delay before the production of the ultrasound signal. This delay is negligible and
therefore it is assumed that the signals leave the beacon at the same time.

Figure 21 shows the structure of the infiared signal that is sent from the beacon.
The signal is a narrow beam, which is IrDA compliant. Figure 22 and 23show a zoomed

in structure of the infrared Signal.

39

 

2.5

2.0 ‘

°/o Error
."'
OI

d
O
1

0.5 «

 

 

 

 

 

+ IRUS
+Cricm//\.

 

 

0.0

200

I

400
Distance In cm

600

800

Figure 17: Error in computed distance at long distances

 

Infrared pulse

 

 

obthdnhéaoamtoam<

muasomd pulse

 

 

‘OANQJ‘AUI<

L

 

Figure 18: TDOA determination at the scope

4O

 

 

6.0

 

+ Scope observed
5'0 _ error

 

 

 

4.0 ~

3.0 ~

% Error

2.0 .

1.0 ~

 

 

 

0.0 I I I 7 r I I I
10 30 50 70 90 200 300 400 500
Distance In cm

Figure 19: Error in computed distance observed at the oscilloscope

The arrival of an inflated signal is detected when there is a rising edge at the
interrupt on-change pin of the microcontroller. Since there is a possibility that two arrival
events of infrared happen consecutively, before the arrival of an ultrasound signal, it is
likely to have two values for the same distance estimate. However, such occurrences are
very rare and would be very likely in cases involving long distances where ultrasound
signals delay. In case they happen, the difference in the values computed is not

significant since the infrared signal is very narrow.

41

Figure 24 shows a sequence of US signals as they leave the beacon. Figure 25 is a
zoomed in representation of the US Signal. Note that the pulse length is approximately
1500 us. It is slightly greater than 1500 due to errors introduced by delay routines and
clock precision.

The arrival of an ultrasound signal at the listener is detected when there is a valid
rising edge at the external interrupt pin of the microcontroller. TO calculate the distance
between the beacon and the listener, the listener has to detect both the infrared and
ultrasound signals within an allowable time frame. False distance computations due to
multiple arrivals of US Signals before IR signals arrive are avoided in the algorithm
running on the listener. Here, it is ensured that no distance computation is done before the
IR arrival event followed by the US arrival event happen.

Signals at the beacon are generated through delay routines. Delay routines in themselves
have errors, which can be reflected in the values of distances computed. For instance,
figure 24 shows a recurring error caused by software delay routines at the beacon. This
error can be corrected by filtering out values corresponding to this TDOA.

AS was described in the software architecture, the listener runs a repetitive
loop until a Si gnal arrives, after which an interrupt is generated. There are several
interrupts competing for system resources and this feature can lead to erroneous distance

computations.

42

 

 

 

 

 

 

 

 

 

 

 

IR signal US signal
w \
4
3
2
1
0 a
-1
-2
-3
4 .
5
O 5 10 15 20 ED
Figure 20: Beacon generated signals
V
5
4
i 3
2
1
~ w - o
-1
-2
.3
I 2 3 4 5 6 7 8 9 13

 

 

 

Figure 21: Infrared signal structure

43

 

 

 

 

8

I

éagbbgoANwrsm<

 

Figure 22: Zoomed-in structure of the infi'ared signal

 

 

 

4 6 8 10 12 14 16

18

B¢1L¢g,;.o-smwr>m<

 

 

Figure 23: Zoomed-in structure of an infrared signal (2)

 

 

 

 

 

 

 

 

 

odtkdorblsoamwrsm<

U)

 

Figure 24: Ultrasound signal structure

 

 

 

ombamhmamwhm<

 

 

25: loomed-in structure of an US signal

45

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

l
A

 

 

 

Outlier TDOA value

Figure 26: Recursive Error caused by delay routines at the beacon

46

Chapter 6: Conclusions

In this thesis, the design, implementation, and characterization of a new indoor
location system has been presented. It has been shown that the time difference of arrival
(TDOA) algorithm, developed by the MIT Cricket team; can be applied to a new
combination of Si gnals: Infrared and Ultrasound. This new combination ensures that
interferences between nodes in neighboring rooms is avoided Since both ultrasound and
infrared signals do not penetrate walls.

A novel system of indoor localization has been designed. This system is based on
range computation using the time difference of arrival of infrared Si gnals and ultrasound
signals. The system architecture has been presented, followed by its characterization in
different ambient conditions and experimental results.

It has been Shown that a location system based on TDOA of infi'ared and
ultrasound signals can be implemented in readily available off-the—shelf components.
Possible advantages Of such a system have been explored and presented.

The system developed and presented in this thesis can be modified to compute
Short distances e.g. from 5-200 cm. At the same time, it can operate at longer distances:
from 2-8 m. Characterization results have shown that the system’s performance at short

as well as long distances is commendable.

47

Chapter 7: Future Work

The system presented in this thesis work is intended for experimentation. In
future, multiple units will be assembled and other aspects of localization will be
investigated. For instance, by building two more beacons, we will enable the listener to
infer its position fiom the messages sent by the three beacons. This feature will call for
the inclusion of a message send and receive mechanism. The listener Should therefore be
able to tell which beacon sent what message.

The use of multiple modules will call for miniaturization. The components used in
the system architecture are available in surface mount technology, and therefore the
system can be miniaturized and industrially produced.

For commercialization, both functions of beacon and listener will be combined in
one module and the option of configuring the module as either beacon or listener will be
left to the user.

Once the characterization of multiple modules has been done, applications will be
developed based on the system. Some of the applications that can be experimented with

are asset tracking and robot navigation.

48

Appendix A: System Schematics

5V

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

RB7 & RB6 are
reserved for
5V programming and ICSP
A.1 Beacon —_ gliffefema for low
voltage programming
U fir—.2 uF
1 ‘M—CLR/VPP RB7/PCB 4° _
2 RAO/ANO RB6/PGC 3’
3 RAl/ANl RBS 3'
4 RAZ/ANZ/VREF— RB4 37
s RA3/AN3/VREF+ RIB/perv] 3‘
a RA4/TOCKI RB2 35
7 RAS/AN4788 " RBI 3‘
s REOIRD/ANS h RBOIINT ’3
9 RBI/Emma a VDD 3’ d
10 REzES/Am O vss 3' '—
11 VDD " RD7/PSPT 3"
12 vss 2 RD6/PSP6 2’
13 OSCl/CLKIN D RDSIPSPS 2”
14 OSC2/CLKOUT RD4/Psp4 27
15 RC0/I‘lOSO/T1CKRC7/Rx/DT 2‘ >>IR_OUT
16 RC1/TlOSI/CCP2 Rme/cx 25
17 RC2/CCP1 RCS/SDO 2“
rs RC3/SCK/SCL Rc4/SDI/SDA 23
19 RDO/PSPO RD3/PSP3 ’2
20 RDl/PSPI RD2/PSP2 2‘

 

 

 

 

 

 

 

 

 

 

 

49

 

A.2 Listener

5V

 

rave

Temp. Sensor

 

 

 

 

100K

5V

 

 

low-[z

 

 

 

“l"l"|"l"|"l"|"l°l°°l“|“l“l‘l”i"l“1

 

 

"II-n o-I

U

WIT/VP? RB7/POD
RAD/AND ngc
RAl/ANI RB,
WAN WRET- R34
RABIANB/VRD'+ MGM
RAuTocm Rm
RM/ANZISS RE“

I
REU/TlD/ANS h Rail/INT
REI/‘WR/Auts E

VDD
RDTCS/AN’I . vss
V0” 5 RD7IPSP7-
VSS . RDIS/PSPfi

oscrlcmrnvl ”5,1,5”
oscz/CLKoUT mums“
RCWTIOSO/qulc-"Rx/DT

RCl/Tl OSI/CCPfiWCK
RC2/C CPI RCSISDO

RC3’5‘:K/SCLRC4/snr/smr

3904’s“) RDSIPSPS

RDWSPI RD2/PSPZ

 

 

 

1H:—

”_ =<IR_DETECI'

— <<US_DE'IECT

US diagnostic LED

 

1R diagnostic LED

 

I"|"l"|"l"l"|"l"l"l°|°l"I”l”|‘°|"|“l°l“l:l

 

 

 

 

 

50

_—

 

 

 

Timerl diagnostic LED

 

 

 

 

 

 

Ultrasound Section [1]

A3

Transmitter

A.3.1

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

.8."

8.03va
Nu;
__ I
__ ._.
Em l Em u 5
3. .5 fl A IL“? __ —
m \ 4.
.4. m 533—
m 2
.m E: w
A A I <w<F
w
a I: F
E... I Em Hi"
)\/\I
8x 8x

_
N

nmmhaa min

51

 

nuroomwfi

Receiver

A.3.2

 

 

 

 

 

I?)
Q‘l g
4 L

US AMP OUT

 

WM

 

 

 

 

 

 

 

 

o
.TIIs/Z’ll:
on :1
. #rNNx firs
a
_ q _ _
u
.3." 8x
5282:
5.533%
cm reserve» <8
W 8x
cm.>:e.<mmm
H H .Em 3x
AV

 

 

a\/

32m

Z

l

 

 

cmt>z_ut_u<<m

 

S

L
a

582"

 

Box

to
[a

 

 

\

 

 

iox

 

 

4+4

Em

5 a
. WIVV caroflmfl
. a-
— 5 Son;

awox

52

Appendix B: System Assembly Program

B.l Beacon

. W
2
m

;This program has parts taken from application notes provided by Microchip company

Filename: Beacon

Date: Aug 17, 2007

File Version: 1.0

Assembled using: MPLAB IDE 7.60.00.0

V. 00 00 v- v- we we we 90 no

Author: David Rurangirwa
CompanyzNEEWS LAB of MSU
; Files required: p16f874.inc

3

 

3
m
9

; Program Description:

;This program enables the PIC16F874 to periodically send a signal

;through the SP1 protocol which is converted into an IR signal by other
;peripherals; and Ultrasound Signal which is generated through the PWM
;mode of the PIC. The signals are sent with a delay Of approx. 9.4 us between
;them and this procedure is repeated every 1 S. The delay routines in this

;pro gram were taken from the delay code generator at
;http://www.piclist.com/techref/piclist/codegen/delayhtm

3

WW

LIST P=16F874

; 20MHZ crystal is being used, thus each instruction cycle = 200nS

53

;*******************RAM TegISter definitions**********************

outbytel
outbyte2
templ
temp2
nmsecO
nmsecl
nmsec2
nmsec_off0
nmsec_off1
nmsec_off2

equ 0x20
equ 0x21
equ 0x22
equ 0x23
equ 0x29
equ 0x24
equ 0x25
equ 0x26
equ 0x27
equ 0x28

o******** 4‘ **** III ' ' ' ********** *** ***************
, *** * * Brt defimtrons * **

CS

equ7

-******************* **********************************
, Include file

include "p 1 6f874.inc"
errorlevel -3 02

;include file for a PIC l 6F877
; suppress message 302 from list file

-It4i***************************************************************
,

org 0x000 ; set the reset vector
goto main ; go to the beginning of program
;!!!!!!!!!!!!llllllllllBegin Main Program!!!llllllllllllllllllllI!
main
IR_INI
bcf STATU8,RPO ; set to bank 0
clrf PORTC ; initialize portc to 0
bsf STATUS,RPO ; set to bank 1
movlw 0x10 ; all bits are outputs except SDI
movwf TRISC ; move the value to TRIS portc
bcf STATUS,RPO ; set to bankO
bsf PORTC,CS ; make the chip select high
clrf PIE] ; disable peripheral interrupts
clrf INTCON ; disables all interrupts
bcf STATU8,RPO ; set to bank 0
clrf SSPCON ; clear SSP control register
movlw 0x20 ; set up spi port, SPI master,
movwf SSPCON ; elk/16, ckp=l (mode 1,1)

54

bsf

STATUS,RPO

clrf
movlw
movwf

; set to bank 1
SSPSTAT ; clear SSP status register
0x80 ; set up spi port, SPI master,
SSPSTAT ; eke = 0 (mode 1,1)

;Send the read enable sequence (RDSR)

Loop

IR_Data

bcf STATUS,RPO ;bankl
bcf PORTC,CS ; clear the chip select line
movlw 0x40 ; load RDSR sequence
movwf outbytel ; store in RAM location outbyte
movlw 0x01 ;load teh LSB of RDSR
movwf outbyte2 ;store in RAM location outbyte
call output ;send the sequence
bsf PORTC,CS ; set the chip select line

;this part of the program repeats
call IR_Data ;send infrared pulse
call delay_l Ocm ;compensate for constant error of 10cm
call U8_Data ;send Ultrasonic pulse
call Delay_off ;wait for 1 S
goto Loop ;before sending again

;Send the write enable sequence (WREN)

bcf
movlw
movwf
movlw
movwf
call
bsf
bcf
movlw
movwf
movlw
movwf
call
bsf

return

PORTC,CS ; clear the chip select (active)
0xC0 ; load WREN sequence
outbytel ; store in RAM location outbyte
0x0A ;baud=9600bps

outbyte2 ;store in RAM location outbyte
output ;send the sequence

PORTC,CS ; set the chip select line
PORTC,CS ; clear the chip select line

0x80 ; load WRITE sequence
outbytel ; store in RAM location outbyte
OxAA ; load high address byte
outbyte2 ; store in RAM location outbyte
output ; call the SP1 output routine
PORTC,CS ; set the chip select line

; go to position where routine was last called

55

output

bcf STATUS,RPO ;bankO

movf outbyte1,w ; move outbytel into w

movwf SSPBUF ; place data in send buffer

call delay ;allow some time lapse

movf outbyte2,w ;move outbyte2 into w

movwf SSPBUF ;place data in send buffer

call delay ;allow some time lapse

return ; go to position where routine was last called
delay

movlw 0x32 ; move literal value into w

movwf templ ; move w value into templ

movlw OxF A ; move literal value into w

movwf temp2 ; move w value into ternp2

decfsz temp2,f ; decrement temp2, skip if zero

goto $-1 ; goto decrement 2 if not zero

decfsz temp1,f ; decrement templ , Skip if zero

goto $-1 ; goto decrement 1 if not zero

return ;retum both locations = 0
U8_Data

;routine to initialize production of US Signal

bsf STATUS,RPO ;move to bank 1

movlw 0x7C ;set the PWM period by writing to
PR2

movwf PR2

bcf STATUS,RPO ;BankO

movlw 0x3E ;duty cycle=50%

movwf CCPRl L ; ;set by writing to CPRl L

call Delay_on ;signal is on for 150us

movlw 0x00

movwf CCPRlL

56

;initiazile the timer and set the prescale of TMR2

bcf
clrf
clrf
movlw

PWM Operation

register

Delay_on

Delay_yus

Delay_off

Delay_xrns

movwf

movlw

movwf
return

movlw
movwf
movlw
movwf

decfsz
goto
decfsz
goto
goto
nop
return

movlw
movwf
movlw
movwf
movlw
movwf

decfsz
goto
decfsz
goto
decfsz

STATUS,RPO

TMR2
T2CON
0x2C

CCP l CON

0x04
T2CON

0x26
nmsecl
0x24
nmsec2

nmsecl ,F
3+2
nmse02,F

Delay _yus
$+1

0x2C
nmsecO
0xE7
nmsecl
0x0B
nmsecZ

nmsec0,F
$+2
nmsecl ,F
$+2
nmsec2,F

;bankO
;clear the TMR2 register

;clear the Timer2 control register
;Configure the CCP] module for

;by writing into the CCP control

;tum on the Timer2, prescale is 1:1

;delay values for 1500us

;delay of 1 s

57

goto Delay_xrns

 

goto 8+1
return
delay_1 0cm
movlw 0x2A
movwf nmsecO
movlw 0x02
movwf nmsecl
Delay_O
decfsz nmsecO, f
goto 8+2
decfsz nmsecl ,f
goto Delay_O
goto $+1
return
END
B.2 Listener
Filename: Listener.asm
Date: July 30th, 2007

“3"“. U. U. U. U. U. U. U. V. U. '0 V. VI '0 no '0

File Version: 4.0

Author: David Rurangirwa
Company: NEEWS Lab of MSU

 

Files required:
1. de26.a16
2. FXM66.a16
3. FP32.a16

 

58

;Notes: This program enables the listener microcontroller to

; detect infrared and ultrasound signals sent fiom the beacon,

, record their times of arrival and find the difference in the times

; of arrival, then calculates the distance between the listener and beacon.

; It features a compensation mechanism where the speed of sound is adjusted
; according to changes in ambient temperature.

 

list p=1 6f874 ; list directive to define processor

#include <p16f874.inc> ; processor specific variable definitions
#include <mathl 6.inc>

errorlevel -302 ;suppress "not in bank 0" message

_CONFIG _CP_OFF & _WDT_ON & _BODEN_ON & _PWRTE_ON &
_RC_OSC & _WRT_ENABLE_ON & _LVP_ON & _CPD_OFF

; ' CONFIG' directive is used to embed configuration data within .asm file.

; The lables following the directive are located in the respective .inc file.
; See respective data sheet for additional information on configuration word.

 

3

;***** Constants

SPBRG_VAL EQU .129 ;set baud rate 9600 for 20Mhz clock

SIG_FIG equ 8 ;set SIG_FIG equal to the number of significant figures in your decimal
number

;for example: ones, tenths,hundredths,thousandths, requires 4 si g figs

last_digit set ones

flag equ 10
TEMP_ad equ l 1
adover equ 0

adif equ 1

adgo equ 2

adie equ 6

gie equ 7

rpO equ 5

fi'ee equ 3

 

,

59

;Variables

CBLOCK 0x70
WREG_TEMP
STATUS_TEMP
PCLATH_TEMP
FSR_TEMP
temperature
error_check
ROC

d1

d2

d3

ENDC

CBLOCK
ACCbROC
ACCbHI
ACCbLO
ACCcHI
ACCcLO
EXPb
ACCdHI
ACCdLO
ACCeHI
ACCeLO
ENDC

0x50

CBLOCK
TMFROC_IR
TMPROC_US
TMPH_IR
TMPL__IR
TMPH_US
TMPL_US
ACCaROC
ACCaHI
ACCaLO
ROCH_IR
ROCH_US
ROCL_IR
ROCL_US

0x40

;storage for WREG during interrupt
;storage for STATUS during interrupt
;storage for PCLATH during interrupt

;storage for FSR during interrupt
;storage for ambient temperature

;storage for rollover count

;Vus, MSB and part of d
;Vus, LSB and part of (1
;part of d

;part of
;Vus, EXP and exp of (1

;IR arrival time, MSB
;IR arrival time, LSB
;US anival time, MSB
;US arrival time, LSB
;TDOA,MSB
;TDOA, MSB
;TDOA, LSB

60

vusl ;low byte Of vus
ENDC

CBLOCK 0x60
trnillions

millions
hundredthousands
tenthousands
thousands

hundreds

tens

ones

temp

digit_count ;counter used to cycle through each digit
ENDC

 

;Macros to select the register bank
; Many bank changes can be Optimized when only one STATUS bit changes

BankO MACRO ;macro to select data RAM bank 0
bcf STATUS,RPO
bcf STATUS,RP1
ENDM

Bankl MACRO ;macro to select data RAM bank 1
bsf STATUS,RPO
bcf STATUS,RP1
ENDM

Bank2 MACRO ;macro to select data RAM bank 2
bcf STATUS,RPO
bsf STATUS,RP1
ENDM

Bank3 MACRO ;macro to select data RAM bank 3
bsf STATUS,RPO
bsf STATUS,RP1
ENDM

61

 

ORG 0x000
clrf PCLATH
goto main

; processor reset vector
; ensure page bits are cleared
; go to beginning of program

 

9

; isr code can go here or be located as a call subroutine elsewhere

 

 

ISR
ORG 0x004 ;place code at interrupt vector
movwf WREG_TEMP ;save WREG
movf STATUS,W ;store STATUS in WREG
clrf STATUS ;select file register bankO
movwf STATUS_TEMP ;save STATUS value
movf PCLATH,W ;store PCLATH in WREG
movwf PCLATH_TEMP ;save PCLATH value
clrf PCLATH ;select program memory pageO
movf FSR,W ;store FSR in WREG
movwf F SR_TEMP ;save FSR value
INTR_POLL
BankO
btfsc PIRl , TMRl IF ;No timerl overflow interrupt?
goto T1_OVRFL
BTFSC INTCON,INTF ; No External interrupt?
goto U8_INT ;Service interrupt
BTFSC INTCON,RBIF;NO interrupt on change?
goto IR_INT
IR_INT
BankO
MOVF PORTB,1
call RDTMR1_IR
IR_rdend
bsf error_check,free ;set bit to avoid erraneous distance
computations
Bankl
bcf TRISD,6;light LED
BankO
BCF INTCON,RBIF
goto EXIT_INT

62

U8_INT
BankO
call
U8_rdend
btfsc
signal was received
call

RDTMRI_US

error_check,free ;check the free bit to determine if IR

TDOA ;if so, then calculate distance based on TDOA,

otherwise light LED and exit ISR

comp_done
Bank]
bcf
BankO
BCF
goto

T1_OVRFL
BankO
BCF
Bank]
bcf
goto

TRISD,7;light LED

INTCON,INTF;clear flag
EXIT _INT

PIRl, TMRlIF ; Clear Timer] Interrupt Flag

TRISD,5
EXIT_INT

 

;Time recordings of IR and US signals

RDTMRI_IR
Bank0
CLRF
MOVF
MOVWF
MOVF
MOVWF
MOVF
SUBWF
BTFSC
return;

INTCON ;All interrupts are disabled
TMRlH, W ; Read high byte

TMPH_IR ;store in temporary register
TMRl L, W ; Read low byte

TMPL_IR ;store in temporary register
TMRlH, W ; Read high byte

TMPH_IR, W ; Sub lst read with 2nd read
STATU8,Z ; IS result = 0

Good l6-bit read

; TMRl L may have rolled over between the read of the high and low bytes.
; Reading the high and low bytes now will read a good value.

63

 

MOVF TMRlH, W ; Read high byte
MOVWF TMPH_IR ;
MOVF TMRl L, W ; Read low byte
MOVWF TMPL_IR ;

; Re-enable the Interrupt (if required)
goto IR_rdend

RDTMRI_US
BankO
CLRF INTCON ;All interrupts are disabled
MOVF TMRlH, W ; Read high byte
MOVWF TMPH_US ;store in temporary register
MOVF TMRl L, W ; Read low byte
MOVWF TMPL_US ;store in temporary register
MOVF TMRlH, W ; Read high byte again
SUBWF TMPH_US, W ; Sub lst read with 2nd read
BTF SC STATU8,Z ; IS result = 0?
return

; TMRl L may have rolled over between the read of the high and low bytes.
; Reading the high and low bytes now will read a good value.

9

MOVF
MOVWF
MOVF
MOVWF

TMRlH, W ; Read high byte
TMPH_US ;

TMRlL, W ; Read low byte
TMPL_US ;

; Re-enable the Interrupt (if required)

goto

U8_rdend

 

9

TDOA

MOVF TMPL_IR,W

SUBWF
between high byte values
MOVWF
difference
MOVF
btfss
incfsz
subwf
MOVWF
difference

TMPL_US,0 ; find difference
ACCaLO ; store

TMPH_IR,W

STATUS,C ;add in carry

TMPH_IR,W

TMPH_US,0

ACCaHI ; store

64

call ReadADC ;read temperature sensor
call LookupTable ;compare with table
values to determine Vus

 

9

 

loadAB
movlw 0x01
movwf ACCbHI
movf vusl,w
movwf ACCbLO ;; loads ACCb = 344m/S, floating point
notation of PIC for 344
goto F_mpy
;Lookup table to vary Vus
LookupTable
movlw 0x23
subwf temperature,w
btfsc STATUS,C
goto NextLookup22
movlw 0x58
movwf vusl
goto loadAB
NextLookup22
movlw 0x24
subwf temperature,w
btfsc STATUS,C
goto NextLookup23
movlw 0x58
movwf vusl
goto loadAB
NextLookup23
movlw 0x25
subwf temperature,w
btfsc STATUS,C
goto NextLookup24
movlw 0x59
movwf vusl
goto loadAB

65

NextLookup24
movlw
subwf
btfsc
goto
movlw
movwf
goto

NextLookup25
movlw
subwf
btfsc
goto
movlw
movwf
goto

NextLookup26
movlw
subwf
btfsc
goto
movlw
movwf
goto

NextLookup27
movlw
subwf
btfsc
goto
movlw
movwf
goto

NextLookup28
movlw
subwf
btfsc
goto
movlw

0x26
temperature,w
STATUS,C
NextLookup25
0x59

vusl

loadAB

0x27
temperature,w
STATUS,C
NextLookup26
0x59

vusl

loadAB

0x28
temperature,w
STATUS,C
NextLookup27
0x5A

vusl

loadAB

0x29
temperature,w
STATUS,C
NextLookup28
0x5A

vusl

loadAB

0x2A
temperature,w
STATUS,C
NextLookup29
0x5A

66

movwf vusl
goto

NextLookup29
movlw
subwf
btfsc
goto
movlw
movwf
goto

NextLookup30
movlw
subwf
btfsc
goto
movlw
movwf
goto

NextLookup8

movlw
subwf
btfsc
goto
movlw
movwf
goto

NextLookup9

movlw
subwf
btfsc
goto
movlw
movwf
goto

loadAB

0x2B
temperature,w
STATUS,C
NextLookup30
0x5B

vusl

loadAB

0x2C
temperature,w
STATUS,C
NextLookup8
0x53

vusl

loadAB

0x16
hunpenuunaur
STFVILHLCI
thuLookup9
0x54

vuSl

loadAdB

0x17
temperature,w
STATUS,C
NextLookupl 0
0x54

vusl

loadAB

67

NextLookup] 0
movlw
subwf
btfsc
goto
movlw
movwf
goto

NextLookup] l
movlw
subwf
btfsc
goto
movlw
movwf
goto

NextLookup 1 2
movlw
subwf
btfsc
goto
movlw
movwf
goto

NextLookup l 3

movlw
subwf
btfsc
goto
movlw
movwf
goto

NextLookup] 4
movlw
subwf
btfsc
goto
movlw
movwf
goto

0x18
temperature,w
STATUS,C
NextLookupl 1
0x54

vusl

loadAB

0x19
temperature,w
STATUS,C
NextLookup] 2
0x55

vusl

loadAB

OxlA
temperature,w
STATUS,C
NextLookup] 3
0x55

vusl

loadAB

0x1 B
temperature,w
STATUS,C
NextLookup14
0x55

vusl

loadAB

0x1 C
temperature,w
STATUS,C
NextLookup] 5
0x56

vusl

loadAB

68

NextLookup] 5
movlw
subwf
btfsc
goto
movlw
movwf
goto

NextLookup] 6
movlw
subwf
btfsc
goto
movlw
movwf
goto

NextLookupl 7
movlw
subwf
btfsc
goto
movlw
movwf
goto

NextLookupl 8
movlw
subwf
btfsc
goto
movlw
movwf
goto

NextLookup] 9
movlw
subwf
btfsc
movlw
movlw
movwf
goto

0x1 D
temperature,w
STATUS,C
NextLookup] 6
0x56

vusl

loadAB

0x1 E
temperature,w
STATUS,C
NextLookupl 7
0x56

vusl

loadAB

OxlF
temperature,w
STATUS,C
NextLookupl 8
0x57

vusl

loadAB

0x20
temperature,w
STATUS,C
NextLookupl 9
0x57

vusl

loadAB

0x21
tenuxaahuenv
STUATTJSJZ
0x57

0x57

vusl

loadAdB

69

 

9

; Binary Floating Point Multiplication :
; ACCb(l 6 bits)EXP(b) * ACCa(16 bits)EXPa -> ACCb(16 bits)EXPb

F_mpy

mloop

9

setup movlw .

D_add

call
bcf
rrf
rrf
btfsc

movf
addwf
btfsc
incf
movf
addwf
btfsc
incf
return

setup

STATUS,C ; clear carry bit
ACCdHI, F ;rotate (I right, place result in W
ACCdLO, F

STATUS,C ;need to add?

D_add

ACCbHI, F

ACCbLO, F

ACCcHI, F

ACCcLO, F

temp, F ;loop until all bits checked

mloop

Convert_D

9999999???

; for 16 shifts

temp
ACCbHI,W
ACCdHI
ACCbLO,W
ACCdLO
ACCbHI
ACCbLO

;move ACCb to ACCd

; clear ACCb ( ACCbLO & ACCbHI )

ACCaLO,W ; Addition ( ACCb + ACCa -> ACCb )
ACCbLO, F ;add lsb
STATUS,C ;add in carry

ACCbHI, F

ACCaHI,W

ACCbHI,F ;add msb
STATUS,C;add in carry
ACCbHI,F

70

 

; ReadADC

 

ReadADC

 

 

 

 

 

BankO
bsf ADCON0,GO
btfsc ADCON0,GO ;loop until conversion is complete
goto 8-1
nop
movf ADRESH,W
movwf temperature
bsf ADCON0,GO
return
Convert_D
movf ACCbHI,W
movwf AARGBO
movf ACCbLO,W
movwf AARGBI
movf ACCcHI,W
movwf AARGB2
movf ACCcLO,W
movwf AARGB3
call int_ascii ;convert a 32-bit int to ASCII
goto Transmit
int_ascii
movlw last_digit
movwf FSR ;pointer = address of smallest digit
movlw SIG_FIG ;load counter with the number of
movwf digit_count ;Significant figures the decimal number
flo_asclp
clrf BARGBO ;Make the divisor 10.
movlw .10
movwf BARGBI
call FXD3216U ;divide (32-bit fixed) / 10 (to get remainder)
movf REMB1,w ;put remainder in w register

71

movwf

INDF ;put number into appropriate digit position

movlw
addwf
decf
decfsz
goto
return
include
include
include

0x30

INDF,f ;add 30h to decimal number to convert to ASCII
FSR,f ;move pointer to next digit

digit_count,f

flo_asclp

<fxd26.al6> ;fixed point 32/16 divide from AN617
<FXM66.al 6> ;32 bit float routines
<FP32.al6>

;we are using FPM32 for 32-bit multiply
;and INT3232 for 32-bit float to 32-bit int
;conversion. Routines are in AN575

 

 

return

Transmit ;send distance computation results to host PC
movf hundredthousands,W
call Send
movf tenthousands,W
call Send
movf thousandS,W
call Send
movf hundreds,W
call Send
movlw 0x0D
call Send
goto comp_done

Send

Bank]
btfss TXSTA,TRMT ; Wait until can send
goto
BankO

movwf TXREG ; Send data in W
return

 

72

 

EXIT_INT ; routine to end interrupt
BankO ;select bank 0
movf F SR_TEMP,W ; get saved FSR value
movwf F SR ;restore FSR
movf PCLATH_TEMP,W ;get saved PCLATH value
movwf PCLATH ;restore PCLATH
movf STATUS_TEMP,W ;get saved STATUS value
movwf STATUS ;restore STATUS
swapf WREG_TEMP,F ;prepare WREG to be restored
swapf WREG_TEMP,W ;restore WREG without affecting
STATUS
retfie ;retum from interrupt
main
Initializations
BankO
CLRF INTCON
CLRF PIRl ; Clear peripheral interrupts Flags
CLRF TlCON ; Stop Timerl, Internal Clock Source,
; Tl oscillator disabled, prescaler = 1:]
CLRF TMRIH ; Clear Tirnerl High byte register
CLRF TMRlL ; Clear Timer] Low byte register
CLRF INTCON ; Disable interrupts
Bank]
CLRF PIE] ; Disable peripheral interrupts
BSF TRISC,0
BankO
MOVLW 0x2A ; External Clock source with oscillator
MOVWF TlCON ; circuitry, 1:4 prescaler, Clock source
; is synchronous to device
;Tirnerl is stopped
BSF TlCON, TMRlON ; Tirnerl starts to increment
; The Timer] interrupt is disabled, do polling on the
overflow bit
bcf error_check,free ;clear the error_check bit

 

73

signal

Bank0
CLRF
CLRF
BSF
BSF

BSF
BSF
BSF
Bank]
movlw
movwf
movlw
movwf
BankO
movlw
movwf
Bank]
BSF
BSF

INTCON

PIRl

INTCON,GIE ; Enable Interrupts
INTCON,INTEDG;enable triggering at rising edge of

lNTCON,INTE;enable external interrupt at pin B0
PIE1,SSPIE;enable ssp interrupt

INTCON,RBIE
SPBRG_VAL ;set baud rate
SPBRG
0x24 ;enable transmission and high baud rate
TXSTA
;select bankO
0x90 ;enable serial port and reception, no errors
RCSTA

PIE1,RCIE ; Enable receive interrupts
PIE] ,TMRl IE ; Enable TMR] Interrupt

 

; InitADC : Analog-Digital Conversion initializations

 

9

InitADC

Bank]
clrf
bsf
BankO
movlw
movwf
bcf
goto

END

ADCON]
PIE1,ADIE

0x81 ;Analog channel is RAO
ADCONO ;ADC is on
PIR1,ADIF

Loop

; directive ‘end of program'

74

[1]

[2]

[3]

[5]
[6]
[7]

[8

[9]

[10]

[11]

[12]

[13]

[14]

Bibliography

Massachusetts Institute of Technology, "The Cricket Indoor Location System,"
Boston, MA, 2005.

Priyantlra B. Nissanka, "Te Cricket Location-Support System", in the 6th ACM
International Conference on Mobile Computing and Networking, Boston, MA,
2000.

Hazas, Mike & Ward, Andy "A Novel Broadband Ultrasonic Location System,"
in Proceedings of UbiComp 2002: Fourth International Conference on
Ubiquitous Computing, Springer-Verlag, Sweden, 2002.

Want, Roy et al, "The Active Badge Location System," Cambridge, England.
Ward, Andy et al, "The Bat Ultrasonic Location System." Cambridge, UK, 2004.

Bah], Paramvir & Padmanabhan, N. Venkata, "RADAR: An In-Building RF-
based User Location and Tracking System" in Microsoft Research.

Welch, Greg et a], "The HiBall Tracking System," in High-Performance Wide-
Area Optical Tracking, Chapel Hill, NC, 2001.

Randell, Cliff & Muller Henk, "Low Cost Indoor Positioning System,” Bristol,
UK 2003.

Youssef, Moustafa & Agrawala, Ashok, "The Horus WLAN Location
Determination System," College Park, MD.

Wallbaurn, Michael & Diepolder, Stefan, "Benchmarking Wireless LAN Location
Systems," in Proceedings of the 2005 Second IEEE International Workshop on
Mobile Commerce and Services, 2005

Hightower, Jeffrey & Borriello, Gaetano, "Location Systems for Ubiquitous
Computing," in IEEE Computer, August 2001.

www.ubisense.net/Product/whitepapers&downloads.htrnl
Curry, M et a1, "Indoor angle of arrival using wide-band frequency diversity with
experimental results and EM propagation modeling" in IEEE-APS Conference on

Antennas and Propagation for Wireless Communications, 6-8 Nov, 2000, pp 65-
68

75

[15]

[16]

[17]

[18]

[19]

[20]

[21]

[22]

[23]

[24]
[25]

[26]

[27]

[28]
[29]
[30]

[31]

Fischer, 8., & Kangas, A., "Time-arrival estimation for E-OTD location in
GERAN" in IEEE lntemational Symposium on Personal, Indoor and Mobile
Radion Communications, Sep/Oct 2001, pp F-121-F-125.

Baker, Bonnie, "Temperature Sensing Technologies," Microchip Technology Inc.,
1991.

Baker, Bonnie, "Thermistors in Single Supply Temperature Sensing Circuits,"
Microchip Technology Inc., 1999.

Evans, Rick, "Floating Point to ASCII Conversion," Microchip Technology Inc.,
2001.

Palmer, Mark, "Using the PORTB Interrupt on Change as an External Interrupt,"
Microchip Technology Inc., 1997.
Garbutt, Mike, "Asynchronous Communications with the PICmicro USART,"
Microchip Technology Inc., 2003.

Schreiber, Robert, "Implementing Ultrasonic Ranging," Microchip Technology
Inc., 1997.

Palmer, Mark, "Using Timer] in Asynchronous Clock Mode," Microchip
Technology Inc., 1997.

Dietz, Ken, "Interfacing SPI Serial EEPROMS to PIC16 Devices," Microchip
Technology Inc., 2005.

Microchip Technology Inc., "Software Interrupt Techniques," 1997.
Microchip Technology Inc., "EEPROM Memory Programming Specification".

Microchip Technology Inc., "Hardware Techniques for PICMicrO
Microcontrollers".

Predko Myke, Programming and Customizing PICmicro Microcontrollers,
McGraw-Hill Companies, Inc., USA, 2001.

Testa J ., Floating Point Math Functions, Microchip Technology Inc., 1997.
http://www.glenbrook.k12.il.us/gbssci/phys/Class/sound/ul 112c.html
Kobitone Audio Company, "Ultrasonic Transducers" datasheet, 05/ 1 1/06

Zetex, "ZXM61P03 datasheet", July, 1999

76

[32]

[33]

[34]

[35]
[36]

[37]

[33]

[39]
[40]
[41]
[42]
[43]

[44]

[45]

Vishay Semiconductors, "IR Receiver Modules for Remote Control Systems",
08/08/2007

Vishay Semiconductors, "Infrared Transceiver Module (SIR, 115.2 kbits/s) for
IrDA applications", 01/26/2004

Microchip Technology Inc., "PICmicrO Mid-Range MCU Family Reference
Manual", December, 1997

Microchip Technology Inc., "PIC16F87X Datasheet", 2001

www.st.com, July, 2005

Fairchild Semiconductor, "DM74L800 Quad 2-Input NAND Gate Datasheet",
March, 2000

MAXIM, "Dual 256-Tap, Low-Drift, digital Potentiometers in 10 uMAX",
04/2001

MAXIM, "SPI/Microwire-Compatible UART in QSOP-l6", 12/2001
MAXIM, "Dual-Output Charge Pump with Shutdown", 03/1996

MAXIM, "MAX232 Datashee ", March, 2004

National Semiconductor, "LMV822 Dual op amp datasheet", December, 2000
National Semiconductor, "LMC7215 datasheet", August, 1996

National Semiconductor, "LM34 Precision Fahrenheit Temperature Sensors
datashee ", November 2000

ECS Inc. lntemational, "1.8432 quartz crystal datasheet".

77

 
      

  

ill

 
 

lliiiiijiliji

3 5