Smart Dust Sensor Network Applications

Architecture and Design 1st Edition Mohammad

Ilyas pdf download

https://ebookname.com/product/smart-dust-sensor-network-
applications-architecture-and-design-1st-edition-mohammad-ilyas/

Get Instant Ebook Downloads – Browse at https://ebookname.com

Instant digital products (PDF, ePub, MOBI) available
Download now and explore formats that suit you...

Handbook of Sensor Networks Compact Wireless and Wired

Sensing Systems 1st Edition Mohammad Ilyas

https://ebookname.com/product/handbook-of-sensor-networks-
compact-wireless-and-wired-sensing-systems-1st-edition-mohammad-
ilyas/

The Handbook of Ad hoc Wireless Networks 1st Edition

Mohammad Ilyas

https://ebookname.com/product/the-handbook-of-ad-hoc-wireless-
networks-1st-edition-mohammad-ilyas/

Sensor Network Protocols 1st Edition Imad Mahgoub

https://ebookname.com/product/sensor-network-protocols-1st-
edition-imad-mahgoub/

The First World War Geoffrey Jukes

https://ebookname.com/product/the-first-world-war-geoffrey-jukes/
Emergence of a Bureaucracy The Florentine Patricians
1530 1790 R. Burr Litchfield

https://ebookname.com/product/emergence-of-a-bureaucracy-the-
florentine-patricians-1530-1790-r-burr-litchfield/

Arkansas A Narrative History 2nd Edition Jeannie M.

Whayne

https://ebookname.com/product/arkansas-a-narrative-history-2nd-
edition-jeannie-m-whayne/

Immaterial Architecture 1st Edition Jonathan Hill

https://ebookname.com/product/immaterial-architecture-1st-
edition-jonathan-hill/

Dictionary of Food Compounds with CD ROM Additives

Flavors and Ingredients Shmuel Yannai

https://ebookname.com/product/dictionary-of-food-compounds-with-
cd-rom-additives-flavors-and-ingredients-shmuel-yannai/

Getting Started in ZBrush an Introduction to Digital

Sculpting and Illustration 1st Edition Johnson

https://ebookname.com/product/getting-started-in-zbrush-an-
introduction-to-digital-sculpting-and-illustration-1st-edition-
johnson/
Closed form Solutions for Drug Transport through
Controlled Release Devices in Two and Three Dimensions
1st Edition Laurent Simon

https://ebookname.com/product/closed-form-solutions-for-drug-
transport-through-controlled-release-devices-in-two-and-three-
dimensions-1st-edition-laurent-simon/
 Smart Dust:
Sensor Network Applications,
Architecture, and Design

Imad Mahgoub
Mohammad Ilyas

© 2006 by Taylor & Francis Group, LLC

 7037_Discl.fm Page 1 Tuesday, November 22, 2005 11:03 AM

The material was previously published in Handbook of Sensor Networks: Compact Wireless and Wired Sensing Systems.
© CRC Press LLC 2005.

Published in 2006 by
CRC Press
Taylor & Francis Group
6000 Broken Sound Parkway NW, Suite 300
Boca Raton, FL 33487-2742

© 2006 by Taylor & Francis Group, LLC

CRC Press is an imprint of Taylor & Francis Group
No claim to original U.S. Government works
Printed in the United States of America on acid-free paper
10 9 8 7 6 5 4 3 2 1
International Standard Book Number-10: 0-8493-7037-X (Hardcover)
International Standard Book Number-13: 978-0-8493-7037-3 (Hardcover)
Library of Congress Card Number 2005022133
This book contains information obtained from authentic and highly regarded sources. Reprinted material is quoted with
permission, and sources are indicated. A wide variety of references are listed. Reasonable efforts have been made to publish
reliable data and information, but the author and the publisher cannot assume responsibility for the validity of all materials
or for the consequences of their use.

No part of this book may be reprinted, reproduced, transmitted, or utilized in any form by any electronic, mechanical, or
other means, now known or hereafter invented, including photocopying, microfilming, and recording, or in any information
storage or retrieval system, without written permission from the publishers.

For permission to photocopy or use material electronically from this work, please access www.copyright.com
(http://www.copyright.com/) or contact the Copyright Clearance Center, Inc. (CCC) 222 Rosewood Drive, Danvers, MA
01923, 978-750-8400. CCC is a not-for-profit organization that provides licenses and registration for a variety of users. For
organizations that have been granted a photocopy license by the CCC, a separate system of payment has been arranged.

Trademark Notice: Product or corporate names may be trademarks or registered trademarks, and are used only for
identification and explanation without intent to infringe.

Library of Congress Cataloging-in-Publication Data

Smart dust : sensor network applications, architecture, and design / editors Imad Mahgoub, and Mohammad
Ilyas.
p. cm.
Includes bibliographical references and index.
ISBN 0-8493-7037-X (9780849370373 : alk. paper)
1. Sensor networks. I. Ilyas, Mohammad, 1953- II. Mahgoub, Imad.

TK7872.D48S63 2006
681'.2--dc22 2005022133

Visit the Taylor & Francis Web site at

http://www.taylorandfrancis.com
Taylor & Francis Group and the CRC Press Web site at
is the Academic Division of Informa plc. http://www.crcpress.com

© 2006 by Taylor & Francis Group, LLC

 7037_C000.fm Page v Tuesday, November 22, 2005 10:13 AM

Preface

Advances in wireless communications and microelectronic mechanical system technologies have enabled
the development of networks of a large number of small inexpensive, low-power multifunctional sensors.
These networks nicknamed “Smart Dust” present a very interesting and challenging area and have
tremendous potential applications.
Wireless sensor networks consist of a large number of sensor nodes that may be randomly and densely
deployed. Sensor nodes are small electronic components capable of sensing many types of information
from the environment including temperature, light, humidity, radiation, the presence or nature of
biological organisms, geological features, seismic vibrations, specific types of computer data, and more.
Recent advancements have made it possible to make these components small, powerful, and energy
efficient, and they can now be manufactured cost-effectively in quantity for specialized telecommunica-
tion applications. The sensor nodes are very small in size and are capable of gathering, processing, and
communicating information to other nodes and to the outside world.
This handbook is expected to capture the current state of sensor networks, and specifically address
the architecture, applications, and design of such networks. This handbook has a total of 17 chapters
written by experts from around the world.
The targeted audience for this handbook includes professionals who are designers and planners for
emerging telecommunication networks, researchers (faculty members and graduate students), and those
who would like to learn about this field.
Although this handbook is not precisely a textbook, it can certainly be used as a textbook for graduate
courses and research-oriented courses that deal with wireless sensor networks. Any comments from the
readers will be highly appreciated.
Many people have contributed to this handbook in their unique ways. The first and the foremost group
that deserves immense gratitude is the group of highly-talented and skilled researchers who have con-
tributed to this handbook. All of them have been extremely cooperative and professional. It has also been
a pleasure to work with Nora Konopka, Helena Redshaw, and Allison Taub of Taylor & Francis, and we
are extremely gratified for their support and professionalism. Our families have extended their uncon-
ditional love and strong support throughout this project and they all deserve very special thanks.
Imad Mahgoub and Mohammad Ilyas
Boca Raton, Florida

© 2006 by Taylor & Francis Group, LLC

v
 7037_C000.fm Page vii Tuesday, November 22, 2005 10:13 AM

Editors

Imad Mahgoub, Ph.D., received his B.Sc. degree in electrical engineering from the University of Khartoum,
Khartoum, Sudan, in 1978. From 1978 to 1981, he worked for the Sudan Shipping Line Company, Port
Sudan, Sudan, as an electrical and electronics engineer. He received his M.S. in applied mathematics in
1983 and his M.S. in electrical and computer engineering in 1986, both from North Carolina State
University. In 1989, he received his Ph.D. in computer engineering from The Pennsylvania State University.
Since August 1989, Dr. Mahgoub has been with the College of Engineering at Florida Atlantic Uni-
versity, Boca Raton, Florida, where he is currently professor of computer science and engineering. He is
the director of the Computer Science and Engineering Department Mobile Computing Laboratory at
Florida Atlantic University.
Dr. Mahgoub has conducted successful research in various areas, including mobile computing; inter-
connection networks; performance evaluation of computer systems; and advanced computer architecture.
He has published more than 80 research articles and supervised three Ph.D. dissertations and 22 M.S.
theses to completion. He has served as a consultant to industry. Dr. Mahgoub served as a member of the
executive committee/program committee of the 1998, 1999, and 2000 IEEE International Performance,
Computing and Communications Conferences. He has served on the program committees of several
international conferences and symposia. He was the vice chair of the 2003, 2004, and 2005 International
Symposium on Performance Evaluation of Computer and Telecommunication Systems. Dr. Mahgoub is
a senior member of IEEE and a member of ACM.

Mohammad Ilyas, Ph.D., received his B.Sc. degree in electrical engineering from the University of
Engineering and Technology, Lahore, Pakistan, in 1976. From March 1977 to September 1978, he worked
for the Water and Power Development Authority in Pakistan. In 1978, he was awarded a scholarship for
his graduate studies and he completed his M.S. degree in electrical and electronic engineering in June
1980 at Shiraz University, Shiraz, Iran. In September 1980, he joined the doctoral program at Queen’s
University in Kingston, Ontario, Canada; he completed his Ph.D. degree in 1983. Dr. Ilyas’s doctoral
research was about switching and flow control techniques in computer communication networks. Since
September 1983, he has been with the College of Engineering at Florida Atlantic University, Boca Raton,
Florida, where he is currently associate dean for graduate studies and research. From 1994 to 2000, he
was chair of the department. During the 1993–1994 academic year, he was on sabbatical leave with the
Department of Computer Engineering, King Saud University, Riyadh, Saudi Arabia.
Dr. Ilyas has conducted successful research in various areas, including traffic management and con-
gestion control in broadband/high-speed communication networks; traffic characterization; wireless
communication networks; performance modeling; and simulation. He has published one book, three
handbooks, and more than 140 research articles. He has supervised 10 Ph.D. dissertations and more than
35 M.S. theses to completion. Dr. Ilyas has been a consultant to several national and international
organizations; a senior member of IEEE, he is an active participant in several IEEE technical committees
and activities.

© 2006 by Taylor & Francis Group, LLC

vii
 7037_C000.fm Page ix Tuesday, November 22, 2005 10:13 AM

Contributors

Özgür B. Akan Joel I. Goodman Sheng-Po Kuo

Georgia Institute of MIT Lincoln Laboratory National Chiao-Tung University
Technology Lexington, Massachusetts Hsin-Chu, Taiwan
Atlanta, Georgia
Martin Haenggi Antonio A.F. Loureiro
Cristian Borcea University of Notre Dame Federal University of Minas
Rutgers University Notre Dame, Indiana Gerais
Piscataway, New Jersey Belo Horizonte, Brazil
Hossam Hassanein
David R. Martinez
Athanassios Boulis Queen’s University
MIT Lincoln Laboratory
University of California at Kingston, Ontario, Canada
Lexington, Massachusetts
Los Angeles
Los Angeles, California Chi-Fu Huang Amitabh Mishra
National Chiao-Tung University Virginia Polytechnic Institute
Erdal Cayirci Hsin-Chu, Taiwan and State University
Istanbul Technical University Blacksburg, Virginia
Istanbul, Turkey Liviu Iftode
Rutgers University José Marcos Nogueira
Anantha Chandrakasan Piscataway, New Jersey Federal University of Minas
Engim, Inc. Gerais
Acton, Massachusetts Chaiporn Jaikaeo Belo Horizonte, Brazil
University of Delaware
Duminda Dewasurendra Newark, Delaware Miodrag Potkonjak
Virginia Polytechnic Institute University of California at
and State University Porlin Kang Los Angeles
Blacksburg, Virginia Rutgers University Los Angeles, California
Piscataway, New Jersey
Albert I. Reuther
Jessica Feng MIT Lincoln Laboratory
University of California at Zdravko Karakehayov
Lexington, Massachusetts
Los Angeles Technical University of Sofia
Los Angeles, California Sofia, Bulgaria Linnyer Beatrys Ruiz
Pontifical Catholic University
Vicente Farinaz Koushanfar of Paraná
González–Millán University of California at Curitiba, Brazil
University of Valencia Berkeley and Federal University of
Valencia, Spain Berkeley, California Minas Gerais
Belo Horizonte, Brazil

© 2006 by Taylor & Francis Group, LLC

ix
 7037_C000.fm Page x Tuesday, November 22, 2005 10:13 AM

Ayad Salhieh Sasha Slijepcevic Quanhong Wang

Wayne State University University of California at Queen’s University
Detroit, Michigan Los Angeles Kingston, Ontario, Canada
Los Angeles, California
Enrique Sanchis-Peris Brett Warneke
University of Valencia Chavalit Dust Networks
Valencia, Spain Srisathapornphat Berkeley, California
University of Delaware
Loren Schwiebert Newark, Delaware
Wayne State University
Jennifer L. Wong
University of California at Los
Detroit, Michigan Weilian Su Angeles
Georgia Institute of
Chien-Chung Shen Los Angeles, California
Technology
University of Delaware Atlanta, Georgia
Newark, Delaware Kenan Xu
Yu-Chee Tseng Queen’s University
Amit Sinha National Chiao-Tung Kingston, Ontario,
Engim, Inc. University Canada
Acton, Massachusetts Hsin-Chu, Taiwan

x
© 2006 by Taylor & Francis Group, LLC
 7037_C000.fm Page xi Tuesday, November 22, 2005 10:13 AM

Contents

1 Opportunities and Challenges in Wireless Sensor Networks Martin Haenggi

1.1 Introduction......................................................................................................................... 1-1
1.2 Opportunities....................................................................................................................... 1-2
1.3 Technical Challenges ........................................................................................................... 1-4
1.4 Concluding Remarks ......................................................................................................... 1-11

2 Next-Generation Technologies to Enable Sensor Networks Joel I. Goodman,

Albert I. Reuther, David R. Martinez
2.1 Introduction......................................................................................................................... 2-1
2.2 Goals for Real-Time Distributed Network Computing for Sensor Data Fusion ............. 2-5
2.3 The Convergence of Networking and Real-Time Computing .......................................... 2-6
2.4 Middleware......................................................................................................................... 2-11
2.5 Network Resource Management....................................................................................... 2-11
2.6 Experimental Results ......................................................................................................... 2-16

3 Sensor Network Management Linnyer Beatrys Ruiz, José Marcos Nogueira,

Antonio A. F. Loureiro
3.1 Introduction......................................................................................................................... 3-1
3.2 Management Challenges...................................................................................................... 3-2
3.3 Management Dimensions ................................................................................................... 3-3
3.4 MANNA as an Integrating Architecture........................................................................... 3-15
3.5 Putting It All Together....................................................................................................... 3-25
3.6 Conclusion ......................................................................................................................... 3-25

4 Models for Programmability in Sensor Networks Athanassios Boulis

4.1 Introduction......................................................................................................................... 4-1
4.2 Differences between Sensor Networks and Traditional Data Networks........................... 4-2
4.3 Aspects of Efficient Sensor Network Applications............................................................. 4-2
4.4 Need for Sensor Network Programmability....................................................................... 4-3
4.5 Major Models for System-Level Programmability............................................................. 4-4
4.6 Frameworks for System-Level Programmability................................................................ 4-6
4.7 Conclusions........................................................................................................................ 4-12

© 2006 by Taylor & Francis Group, LLC

xi
 7037_C000.fm Page xii Tuesday, November 22, 2005 10:13 AM

5 Miniaturizing Sensor Networks with MEMS Brett Warneke

5.1 Introduction......................................................................................................................... 5-1
5.2 MEMS Basics........................................................................................................................ 5-2
5.3 Sensors.................................................................................................................................. 5-4
5.4 Communication................................................................................................................... 5-5
5.5 Micropower Sources .......................................................................................................... 5-10
5.6 Packaging............................................................................................................................ 5-12
5.7 Systems ............................................................................................................................... 5-13
5.8 Conclusion ......................................................................................................................... 5-15

6 Sensor Network Architecture and Applications Chien-Chung Shen, Chaiporn Jaikaeo,

Chavalit Srisathapornphat
6.1 Introduction......................................................................................................................... 6-1
6.2 Sensor Network Applications.............................................................................................. 6-1
6.3 Functional Architecture for Sensor Networks.................................................................... 6-3
6.4 Sample Implementation Architectures............................................................................... 6-4
6.5 Summary ............................................................................................................................ 6-12

7 A Practical Perspective on Wireless Sensor Networks Quanhong Wang,

Hossam Hassanein, Kenan Xu
7.1 Introduction......................................................................................................................... 7-1
7.2 WSN Applications................................................................................................................ 7-2
7.3 Classification of WSNs ........................................................................................................ 7-6
7.4 Characteristics, Technical Challenges, and Design Directions.......................................... 7-7
7.5 Technical Approaches........................................................................................................ 7-11
7.6 Conclusions and Considerations for Future Research .................................................... 7-22

8 Sensor Network Architecture Jessica Feng, Farinaz Koushanfar, Miodrag Potkonjak

8.1 Overview............................................................................................................................... 8-1
8.2 Motivation and Objectives .................................................................................................. 8-1
8.3 SNs — Global View and Requirements.............................................................................. 8-3
8.4 Individual Components of SN Nodes................................................................................. 8-4
8.5 Sensor Network Node.......................................................................................................... 8-8
8.6 Wireless SNs as Embedded Systems.................................................................................. 8-13
8.7 Summary ............................................................................................................................ 8-16

9 Power-Efficient Topologies for Wireless Sensor Networks Ayad Salhieh,

Loren Schwiebert
9.1 Motivation............................................................................................................................ 9-1
9.2 Background .......................................................................................................................... 9-2
9.3 Issues for Topology Design ................................................................................................. 9-3
9.4 Assumptions......................................................................................................................... 9-8

xii
© 2006 by Taylor & Francis Group, LLC
 7037_C000.fm Page xiii Tuesday, November 22, 2005 10:13 AM

9.5 Analysis of Power Usage .................................................................................................... 9-10

9.6 Directional Source-Aware Routing Protocol (DSAP) ..................................................... 9-13
9.7 DSAP Analysis.................................................................................................................... 9-15
9.8 Summary ............................................................................................................................ 9-19

10 Overview of Communication Protocols for Sensor Networks Weilian Su,

Erdal Cayirci, Özgür B. Akan
10.1 Introduction...................................................................................................................... 10-1
10.2 Applications/Application Layer Protocols....................................................................... 10-2
10.3 Localization Protocols ...................................................................................................... 10-4
10.4 Time Synchronization Protocols ..................................................................................... 10-5
10.5 Transport Layer Protocols................................................................................................ 10-7
10.6 Network Layer Protocols.................................................................................................. 10-9
10.7 Data Link Layer Protocols .............................................................................................. 10-11
10.8 Conclusion ...................................................................................................................... 10-14

11 Positioning and Location Tracking in Wireless Sensor Networks Yu-Chee Tseng,

Chi-Fu Huang, Sheng-Po Kuo
11.1 Introduction...................................................................................................................... 11-1
11.2 Fundamentals.................................................................................................................... 11-2
11.3 Positioning and Location Tracking Algorithms.............................................................. 11-4
11.4 Experimental Location Systems ..................................................................................... 11-10
11.5 Conclusions..................................................................................................................... 11-12

12 Comparison of Data Processing Techniques in Sensor Networks

Vicente González-Millán, Enrique Sanchis-Peris
12.1 Sensor Networks: Organization and Processing ............................................................. 12-1
12.2 Architectures for Sensor Integration ............................................................................... 12-3
12.3 Example of Architecture Evaluation in High-Energy Physics...................................... 12-18

13 Cooperative Computing in Sensor Networks Liviu Iftode, Cristian Borcea,

Porlin Kang
13.1 Introduction...................................................................................................................... 13-1
13.2 The Cooperative Computing Model................................................................................ 13-3
13.3 Node Architecture............................................................................................................. 13-4
13.4 Smart Messages ................................................................................................................. 13-5
13.5 Programming Interface .................................................................................................... 13-7
13.6 Prototype Implementation and Evaluation..................................................................... 13-8
13.7 Applications .................................................................................................................... 13-12
13.8 Simulation Results .......................................................................................................... 13-14
13.9 Related Work................................................................................................................... 13-15
13.10 Conclusions..................................................................................................................... 13-18

© 2006 by Taylor & Francis Group, LLC

xiii
 7037_C000.fm Page xiv Tuesday, November 22, 2005 10:13 AM

14 Dynamic Power Management in Sensor Networks Amit Sinha,

Anantha Chandrakasan
14.1 Introduction...................................................................................................................... 14-1
14.2 Idle Power Management................................................................................................... 14-2
14.3 Active Power Management............................................................................................... 14-5
14.4 System Implementation.................................................................................................... 14-6
14.5 Results.............................................................................................................................. 14-12

15 Design Challenges in Energy-Efficient Medium Access Control for

Wireless Sensor Networks Duminda Dewasurendra, Amitabh Mishra
15.1 Introduction...................................................................................................................... 15-1
15.2 Unique Characteristics of Wireless Sensor Networks..................................................... 15-2
15.3 MAC Protocols for Wireless Ad Hoc Networks.............................................................. 15-4
15.4 Design Challenges for Wireless Sensor Networks......................................................... 15-10
15.5 Medium Access Protocols for Wireless Sensor Networks ............................................ 15-13
15.6 Open Issues ..................................................................................................................... 15-22
15.7 Conclusions..................................................................................................................... 15-24

16 Security and Privacy Protection in Wireless Sensor Networks Sasha Slijepcevic,

Jennifer L. Wong, Miodrag Potkonjak
16.1 Introduction...................................................................................................................... 16-1
16.2 Unique Security Challenges in Sensor Networks and Enabling Mechanisms............... 16-2
16.3 Security Architectures....................................................................................................... 16-4
16.4 Privacy Protection........................................................................................................... 16-11
16.5 Conclusion ...................................................................................................................... 16-15

17 Low-Power Design for Smart Dust Networks Zdravko Karakehayov

17.1 Introduction...................................................................................................................... 17-1
17.2 Location............................................................................................................................. 17-1
17.3 Sensing............................................................................................................................... 17-2
17.4 Computation..................................................................................................................... 17-2
17.5 Hardware–Software Interaction....................................................................................... 17-5
17.6 Communication................................................................................................................ 17-7
17.7 Orientation...................................................................................................................... 17-10
17.8 Conclusion ...................................................................................................................... 17-10

xiv
© 2006 by Taylor & Francis Group, LLC
 7037_C001.fm Page 1 Tuesday, November 1, 2005 12:46 PM

1
Opportunities and
Challenges in Wireless
Sensor Networks

1.1 Introduction ...................................................................... 1-1

1.2 Opportunities .................................................................... 1-2
Growing Research and Commercial Interest • Applications
1.3 Technical Challenges ......................................................... 1-4
Performance Metrics • Power Supply • Design of Energy-
Efficient Protocols • Capacity/Throughput • Routing • Channel
Access and Scheduling • Modeling • Connectivity • Quality of
Martin Haenggi Service • Security • Implementation • Other Issues
University of Notre Dame 1.4 Concluding Remarks....................................................... 1-11

1.1 Introduction
Due to advances in wireless communications and electronics over the last few years, the development of
networks of low-cost, low-power, multifunctional sensors has received increasing attention. These sensors
are small in size and able to sense, process data, and communicate with each other, typically over an RF
(radio frequency) channel. A sensor network is designed to detect events or phenomena, collect and
process data, and transmit sensed information to interested users. Basic features of sensor networks are:

• Self-organizing capabilities
• Short-range broadcast communication and multihop routing
• Dense deployment and cooperative effort of sensor nodes
• Frequently changing topology due to fading and node failures
• Limitations in energy, transmit power, memory, and computing power

These characteristics, particularly the last three, make sensor networks different from other wireless ad
hoc or mesh networks.
Clearly, the idea of mesh networking is not new; it has been suggested for some time for wireless
Internet access or voice communication. Similarly, small computers and sensors are not innovative
per se. However, combining small sensors, low-power computers, and radios makes for a new tech-
nological platform that has numerous important uses and applications, as will be discussed in the next
section.

1-1
© 2006 by Taylor & Francis Group, LLC
 7037_C001.fm Page 2 Tuesday, November 1, 2005 12:46 PM

1-2 Smart Dust

1.2 Opportunities
1.2.1 Growing Research and Commercial Interest
Research and commercial interest in the area of wireless sensor networks are currently growing expo-
nentially, which is manifested in many ways:
• The number of Web pages (Google: 26,000 hits for sensor networks; 8000 for wireless sensor
networks in August 2003)
• The increasing number of
• Dedicated annual workshops, such as IPSN (information processing in sensor networks);
SenSys; EWSN (European workshop on wireless sensor networks); SNPA (sensor network
protocols and applications); and WSNA (wireless sensor networks and applications)
• Conference sessions on sensor networks in the communications and mobile computing com-
munities (ISIT, ICC, Globecom, INFOCOM, VTC, MobiCom, MobiHoc)
• Research projects funded by NSF (apart from ongoing programs, a new specific effort now
focuses on sensors and sensor networks) and DARPA through its SensIT (sensor information
technology), NEST (networked embedded software technology), MSET (multisensor exploi-
tation), UGS (unattended ground sensors), NETEX (networking in extreme environments),
ISP (integrated sensing and processing), and communicator programs
Special issues and sections in renowned journals are common, e.g., in the IEEE Proceedings [1] and signal
processing, communications, and networking magazines. Commercial interest is reflected in investments
by established companies as well as start-ups that offer general and specific hardware and software
solutions.
Compared to the use of a few expensive (but highly accurate) sensors, the strategy of deploying a large
number of inexpensive sensors has significant advantages, at smaller or comparable total system cost:
much higher spatial resolution; higher robustness against failures through distributed operation; uniform
coverage; small obtrusiveness; ease of deployment; reduced energy consumption; and, consequently,
increased system lifetime. The main point is to position sensors close to the source of a potential problem
phenomenon, where the acquired data are likely to have the greatest benefit or impact.
Pure sensing in a fine-grained manner may revolutionize the way in which complex physical systems
are understood. The addition of actuators, however, opens a completely new dimension by permitting
management and manipulation of the environment at a scale that offers enormous opportunities for
almost every scientific discipline. Indeed, Business 2.0 (http://www.business2.com/) lists sensor robots
as one of “six technologies that will change the world,” and Technology Review at MIT and Globalfuture
identify WSNs as one of the “10 emerging technologies that will change the world” (http://www.global-
future.com/mit-trends2003.htm). The combination of sensor network technology with MEMS and nan-
otechnology will greatly reduce the size of the nodes and enhance the capabilities of the network.
The remainder of this chapter lists and briefly describes a number of applications for wireless sensor
networks, grouped into different categories. However, because the number of areas of application is
growing rapidly, every attempt at compiling an exhaustive list is bound to fail.

1.2.2 Applications
1.2.2.1 General Engineering
• Automotive telematics. Cars, which comprise a network of dozens of sensors and actuators, are
networked into a system of systems to improve the safety and efficiency of traffic.
• Fingertip accelerometer virtual keyboards. These devices may replace the conventional input
devices for PCs and musical instruments.
• Sensing and maintenance in industrial plants. Complex industrial robots are equipped with up
to 200 sensors that are usually connected by cables to a main computer. Because cables are

© 2006 by Taylor & Francis Group, LLC

 7037_C001.fm Page 3 Tuesday, November 1, 2005 12:46 PM

Opportunities and Challenges in Wireless Sensor Networks 1-3

expensive and subject to wear and tear caused by the robot’s movement, companies are replacing
them by wireless connections. By mounting small coils on the sensor nodes, the principle of
induction is exploited to solve the power supply problem.
• Aircraft drag reduction. Engineers can achieve this by combining flow sensors and blowing/sucking
actuators mounted on the wings of an airplane.
• Smart office spaces. Areas are equipped with light, temperature, and movement sensors, micro-
phones for voice activation, and pressure sensors in chairs. Air flow and temperature can be
regulated locally for one room rather than centrally.
• Tracking of goods in retail stores. Tagging facilitates the store and warehouse management.
• Tracking of containers and boxes. Shipping companies are assisted in keeping track of their goods,
at least until they move out of range of other goods.
• Social studies. Equipping human beings with sensor nodes permits interesting studies of human
interaction and social behavior.
• Commercial and residential security.
1.2.2.2 Agriculture and Environmental Monitoring
• Precision agriculture. Crop and livestock management and precise control of fertilizer concentra-
tions are possible.
• Planetary exploration. Exploration and surveillance in inhospitable environments such as remote
geographic regions or toxic locations can take place.
• Geophysical monitoring. Seismic activity can be detected at a much finer scale using a network of
sensors equipped with accelerometers.
• Monitoring of freshwater quality. The field of hydrochemistry has a compelling need for sensor
networks because of the complex spatiotemporal variability in hydrologic, chemical, and ecological
parameters and the difficulty of labor-intensive sampling, particularly in remote locations or under
adverse conditions. In addition, buoys along the coast could alert surfers, swimmers, and fishermen
to dangerous levels of bacteria.
• Zebranet. The Zebranet project at Princeton aims at tracking the movement of zebras in Africa.
• Habitat monitoring. Researchers at UC Berkeley and the College of the Atlantic in Bar Harbor
deployed sensors on Great Duck Island in Maine to measure humidity, pressure, temperature,
infrared radiation, total solar radiation, and photosynthetically active radiation (see http://
www.greatduckisland.net/).
• Disaster detection. Forest fire and floods can be detected early and causes can be localized precisely
by densely deployed sensor networks.
• Contaminant transport. The assessment of exposure levels requires high spatial and temporal
sampling rates, which can be provided by WSNs.
1.2.2.3 Civil Engineering
• Monitoring of structures. Sensors will be placed in bridges to detect and warn of structural
weakness and in water reservoirs to spot hazardous materials. The reaction of tall buildings to
wind and earthquakes can be studied and material fatigue can be monitored closely.
• Urban planning. Urban planners will track groundwater patterns and how much carbon dioxide
cities are expelling, enabling them to make better land-use decisions.
• Disaster recovery. Buildings razed by an earthquake may be infiltrated with sensor robots to locate
signs of life.
1.2.2.4 Military Applications
• Asset monitoring and management. Commanders can monitor the status and locations of troops,
weapons, and supplies to improve military command, control, communications, and computing
(C4).

© 2006 by Taylor & Francis Group, LLC

 7037_C001.fm Page 4 Tuesday, November 1, 2005 12:46 PM

1-4 Smart Dust

• Surveillance and battle-space monitoring. Vibration and magnetic sensors can report vehicle and
personnel movement, permitting close surveillance of opposing forces.
• Urban warfare. Sensors are deployed in buildings that have been cleared to prevent reoccupation;
movements of friend and foe are displayed in PDA-like devices carried by soldiers. Snipers can be
localized by the collaborative effort of multiple acoustic sensors.
• Protection. Sensitive objects such as atomic plants, bridges, retaining walls, oil and gas pipelines,
communication towers, ammunition depots, and military headquarters can be protected by intel-
ligent sensor fields able to discriminate between different classes of intruders. Biological and
chemical attacks can be detected early or even prevented by a sensor network acting as a warning
system.
• Self-healing minefields. The self-healing minefield system is designed to achieve an increased resistance
to dismounted and mounted breaching by adding a novel dimension to the minefield. Instead of a
static complex obstacle, the self-healing minefield is an intelligent, dynamic obstacle that senses
relative positions and responds to an enemy’s breaching attempt by physical reorganization.
1.2.2.5 Health Monitoring and Surgery
• Medical sensing. Physiological data such as body temperature, blood pressure, and pulse are sensed
and automatically transmitted to a computer or physician, where it can be used for health status
monitoring and medical exploration. Wireless sensing bandages may warn of infection. Tiny
sensors in the blood stream, possibly powered by a weak external electromagnetic field, can
continuously analyze the blood and prevent coagulation and thrombosis.
• Microsurgery. A swarm of MEMS-based robots may collaborate to perform microscopic and
minimally invasive surgery.
The opportunities for wireless sensor networks are ubiquitous. However, a number of formidable chal-
lenges must be solved before these exciting applications may become reality.

1.3 Technical Challenges

Populating the world with networks of sensors requires a fundamental understanding of techniques for
connecting and managing sensor nodes with a communication network in scalable and resource-efficient
ways. Clearly, sensor networks belong to the class of ad hoc networks, but they have specific characteristics
that are not present in general ad hoc networks.
Ad hoc and sensor networks share a number of challenges such as energy constraints and routing. On
the other hand, general ad hoc networks most likely induce traffic patterns different from sensor networks,
have other lifetime requirements, and are often considered to consist of mobile nodes [2–4]. In WSNs,
most nodes are static; however, the network of basic sensor nodes may be overlaid by more powerful
mobile sensors (robots) that, guided by the basic sensors, can move to interesting areas or even track
intruders in the case of military applications.
Network nodes are equipped with wireless transmitters and receivers using antennas that may be
omnidirectional (isotropic radiation), highly directional (point-to-point), possibly steerable, or some
combination thereof. At a given point in time, depending on the nodes’ positions and their transmitter
and receiver coverage patterns, transmission power levels, and cochannel interference levels, a wireless
connectivity exists in the form of a random, multihop graph between the nodes. This ad hoc topology
may change with time as the nodes move or adjust their transmission and reception parameters.
Because the most challenging issue in sensor networks is limited and unrechargeable energy provision,
many research efforts aim at improving the energy efficiency from different aspects. In sensor networks,
energy is consumed mainly for three purposes: data transmission, signal processing, and hardware
operation [5]. It is desirable to develop energy-efficient processing techniques that minimize power
requirements across all levels of the protocol stack and, at the same time, minimize message passing for
network control and coordination.

© 2006 by Taylor & Francis Group, LLC

 7037_C001.fm Page 5 Tuesday, November 1, 2005 12:46 PM

Opportunities and Challenges in Wireless Sensor Networks 1-5

1.3.1 Performance Metrics

To discuss the issues in more detail, it is necessary to examine a list of metrics that determine the
performance of a sensor network:
• Energy efficiency/system lifetime. The sensors are battery operated, rendering energy a very scarce
resource that must be wisely managed in order to extend the lifetime of the network [6].
• Latency. Many sensor applications require delay-guaranteed service. Protocols must ensure that
sensed data will be delivered to the user within a certain delay. Prominent examples in this class
of networks are certainly the sensor-actuator networks.
• Accuracy. Obtaining accurate information is the primary objective; accuracy can be improved
through joint detection and estimation. Rate distortion theory is a possible tool to assess accuracy.
• Fault tolerance. Robustness to sensor and link failures must be achieved through redundancy and
collaborative processing and communication.
• Scalability. Because a sensor network may contain thousands of nodes, scalability is a critical
factor that guarantees that the network performance does not significantly degrade as the network
size (or node density) increases.
• Transport capacity/throughput. Because most sensor data must be delivered to a single base station
or fusion center, a critical area in the sensor network exists (the gray area in Figure 1.1.), whose
sensor nodes must relay the data generated by virtually all nodes in the network. Thus, the traffic
load at those critical nodes is heavy, even when the average traffic rate is low. Apparently, this area
has a paramount influence on system lifetime, packet end-to-end delay, and scalability.
Because of the interdependence of energy consumption, delay, and throughput, all these issues and
metrics are tightly coupled. Thus, the design of a WSN necessarily consists of the resolution of numerous
trade-offs, which also reflects in the network protocol stack, in which a cross-layer approach is needed
instead of the traditional layer-by-layer protocol design.

1.3.2 Power Supply

The most difficult constraints in the design of WSNs are those regarding the minimum energy consumption
necessary to drive the circuits and possible microelectromechanical devices (MEMS) [5, 7, 8]. The energy
problem is aggravated if actuators are present that may be substantially hungrier for power than the sensors.
When miniaturizing the node, the energy density of the power supply is the primary issue. Current
technology yields batteries with approximately 1 J/mm3 of energy, while capacitors can achieve as much as
1 mJ/mm3. If a node is designed to have a relatively short life span, for example, a few months, a battery is
a logical solution. However, for nodes that can generate sensor readings for long periods of time, a charging

BS critical nodes

FIGURE 1.1 Sensor network with base station (or fusion center). The gray-shaded area indicates the critical area
whose nodes must relay all the packets.

© 2006 by Taylor & Francis Group, LLC

 7037_C001.fm Page 6 Tuesday, November 1, 2005 12:46 PM

1-6 Smart Dust

method for the supply is preferable. Currently, research groups are investigating the use of solar cells to
charge capacitors with photocurrents from the ambient light sources. Solar flux can yield power densities
of approximately 1 mW/mm2. The energy efficiency of a solar cell ranges from 10 to 30% in current
technologies, giving 300 μW in full sunlight in the best-case scenario for a 1-mm2 solar cell operating at
1 V. Series-stacked solar cells will need to be utilized in order to provide appropriate voltages.
Sensor acquisition can be achieved at 1 nJ per sample, and modern processors can perform compu-
tations as low as 1 nJ per instruction. For wireless communications, the primary candidate technologies
are based on RF and optical transmission techniques, each of which has its advantages and disadvantages.
RF presents a problem because the nodes may offer very limited space for antennas, thereby demanding
very short-wavelength (i.e., high-frequency) transmission, which suffers from high attenuation. Thus,
communication in that regime is not currently compatible with low-power operation. Current RF
transmission techniques (e.g., Bluetooth [9]) consume about 100 nJ per bit for a distance of 10 to 100 m,
making communication very expensive compared to acquisition and processing.
An alternative is to employ free-space optical transmission. If a line-of-sight path is available, a well-
designed free-space optical link requires significantly lower energy than its RF counterpart, currently
about 1 nJ per bit. The reason for this power advantage is that optical transceivers require only simple
baseband analog and digital circuitry and no modulators, active filters, and demodulators. Furthermore,
the extremely short wavelength of visible light makes it possible for a millimeter-scale device to emit a
narrow beam, corresponding to an antenna gain of roughly five to six orders of magnitude compared to
an isotropic radiator. However, a major disadvantage is that the beam needs to be pointed very precisely
at the receiver, which may be prohibitively difficult to achieve.
In WSNs, where sensor sampling, processing, data transmission, and, possibly, actuation are involved,
the trade-off between these tasks plays an important role in power usage. Balancing these parameters
will be the focus of the design process of WSNs.

1.3.3 Design of Energy-Efficient Protocols

It is well acknowledged that clustering is an efficient way to save energy for static sensor networks [10–13].
Clustering has three significant differences from conventional clustering schemes. First, data compression
in the form of distributed source coding is applied within a cluster to reduce the number of packets to
be transmitted [14, 15]. Second, the data-centric property makes an identity (e.g., an address) for a
sensor node obsolete. In fact, the user is often interested in phenomena occurring in a specified area
[16], rather than in an individual sensor node. Third, randomized rotation of cluster heads helps ensure
a balanced energy consumption [11].
Another strategy to increase energy efficiency is to use broadcast and multicast trees [6, 17, 18], which
take advantage of the broadcast property of omnidirectional antennas. The disadvantage is that the high
computational complexity may offset the achievable benefit. For sensor networks, this one-to-many
communication scheme is less important; however, because all data must be delivered to a single desti-
nation, the traffic scheme (for application traffic) is the opposite, i.e., many to one. In this case, clearly
the wireless multicast advantage offers less benefit, unless path diversity or cooperative diversity schemes
are implemented [19, 20].
The exploitation of sleep modes [21, 22] is imperative to prevent sensor nodes from wasting energy
in receiving packets unintended for them. Combined with efficient medium access protocols, the “sleep-
ing” approach could reach optimal energy efficiency without degradation in throughput (but at some
penalty in delay).

1.3.4 Capacity/Throughput
Two parameters describe the network’s capability to carry traffic: transport capacity and throughput.
The former is a distance-weighted sum capacity that permits evaluation of network performance.
Throughput is a traditional measure of how much traffic can be delivered by the network [23–30]. In a

© 2006 by Taylor & Francis Group, LLC

 7037_C001.fm Page 7 Tuesday, November 1, 2005 12:46 PM

Opportunities and Challenges in Wireless Sensor Networks 1-7

packet network, the (network-layer) throughput may be defined as the expected number of successful
packet transmissions of a given node per timeslot.
The capacity of wireless networks in general is an active area of research in the information theory
community. The results obtained mostly take the form of scaling laws or “order-of ” results; the prefactors
are difficult to determine analytically. Important results include the scaling law for point-to-point coding,
which shows that the throughput decreases with 1 / N for a network with N nodes [23]. Newer results
[28] permit network coding, which yields a slightly more optimistic scaling behavior, although at high
complexity. Grossglauser and Tse [26] have shown that mobility may keep the per-node capacity constant
as the network grows, but that benefit comes at the cost of unbounded delay.
The throughput is related to (error-free) transmission rate of each transmitter, which, in turn, is upper
bounded by the channel capacity. From the pure information theoretic point of view, the capacity is
computed based on the ergodic channel assumption, i.e., the code words are long compared to the
coherence time of the channel. This Shannon-type capacity is also called throughput capacity [31].
However, in practical networks, particularly with delay-constrained applications, this capacity cannot
provide a helpful indication of the channel’s ability to transmit with a small probability of error.
Moreover, in the multiple-access system, the corresponding power allocation strategies for maximum
achievable capacity always favor the “good” channels, thus leading to unfairness among the nodes.
Therefore, for delay-constrained applications, the channel is usually assumed to be nonergodic and the
capacity is a random variable, instead of a constant in the classical definition by Shannon. For a delay-
bound D, the channel is often assumed to be block fading with block length D, and a composite channel
model is appropriate when specifying the capacity. Correspondingly, given the noise power, the channel
state (a random variable in the case of fading channels), and power allocation, new definitions for delay-
constrained systems have been proposed [32–35].

1.3.5 Routing
In ad hoc networks, routing protocols are expected to implement three main functions: determining and
detecting network topology changes (e.g., breakdown of nodes and link failures); maintaining network
connectivity; and calculating and finding proper routes. In sensor networks, up-to-date, less effort has
been given to routing protocols, even though it is clear that ad hoc routing protocols (such as destination-
sequenced distance vector (DSDV), temporally-ordered routing algorithm (TORA), dynamic source rout-
ing (DSR), and ad hoc on-demand distance vector (AODV) [4, 36–39]) are not suited well for sensor
networks since the main type of traffic in WSNs is “many to one” because all nodes typically report to
a single base station or fusion center. Nonetheless, some merits of these protocols relate to the features
of sensor networks, like multihop communication and QoS routing [39]. Routing may be associated with
data compression [15] to enhance the scalability of the network.

1.3.6 Channel Access and Scheduling

In WSNs, scheduling must be studied at two levels: the system level and the node level. At the node level,
a scheduler determines which flow among all multiplexing flows will be eligible to transmit next (the
same concept as in traditional wired scheduling); at the system level, a scheme determines which nodes
will be transmitting. System-level scheduling is essentially a medium access (MAC) problem, with the
goal of minimum collisions and maximum spatial reuse — a topic receiving great attention from the
research community because it is tightly coupled with energy efficiency and throughput.
Most of the current wireless scheduling algorithms aim at improved fairness, delay, robustness (with
respect to network topology changes) and energy efficiency [62, 64, 65, 66]. Some also propose a distrib-
uted implementation, in contrast to the centralized implementation in wired or cellular networks, which
originated from general fair queuing. Also, wireless (or sensor) counterparts of other wired scheduling
classes, like priority scheduling [67, 68] and earliest deadline first (EDF) [69], confirm that prioritization
is necessary to achieve delay balancing and energy balancing.

© 2006 by Taylor & Francis Group, LLC

 7037_C001.fm Page 8 Tuesday, November 1, 2005 12:46 PM

1-8 Smart Dust

The main problem in WSNs is that all the sensor data must be forwarded to a base station via multihop
routing. Consequently, the traffic pattern is highly nonuniform, putting a high burden on the sensor
nodes close to the base station (the critical nodes in Figure 1.1). The scheduling algorithm and routing
protocols must aim at energy and delay balancing, ensuring that packets originating close and far away
from the base station experience a comparable delay, and that the critical nodes do not die prematurely
due to the heavy relay traffic [40].
At this point, due to the complexity of scheduling algorithms and the wireless environment, most
performance measures are given through simulation rather than analytically. Moreover, medium access
and scheduling are usually considered separately. When discussing scheduling, the system is assumed to
have a single user; whereas in the MAC layer, all flows multiplexing at the node are treated in the same
way, i.e., a default FIFO buffer is assumed to schedule flows. It is necessary to consider them jointly to
optimize performance figures such as delay, throughput, and packet loss probability.
Because of the bursty nature of the network traffic, random access methods are commonly employed
in WSNs, with or without carrier sense mechanisms. For illustrative purposes, consider the simplest
sensible MAC scheme possible: all nodes are transmitting packets independently in every timeslot with
the same transmit probability p at equal transmitting power levels; the next-hop receiver of every packet
is one of its neighbors. The packets are of equal length and fit into one timeslot. This MAC scheme was
considered in Silvester and Kleinrock [41], Hu [42], and Haenggi [43]. The resulting (per-node) through-
put turns out to be a polynomial in p of order N, where N is the number of nodes in the network.
A typical throughput polynomial is shown in Figure 1.2. At p = 0, the derivative is 1, indicating that,
for small p, the throughput equals p. This is intuitive because there are few collisions for small p and the
throughput g(p) is approximately linear. The region in which the packet loss probability is less than 10%
can be denoted as the collisionless region. It ranges from 0 to about pmax/8. The next region, up to pmax,
is the practical region in which energy consumption (transmission attempts) is traded off against through-
put; it is therefore called the trade-off region. The difference p – g(p) is the interference loss. For small
networks, all N nodes interfere with each other because spatial reuse is not possible: If more than one
node is transmitting, a collision occurs and all packets are lost. Thus, the (per-node) throughput is
p(1 – p)N–1, and the optimum transmit probability is 1/N. The maximum throughput is (1 – 1/N)N–1/N.
With increasing N, the throughput approaches 1/(eN), as pointed out in Silvester and Kleinrock [41]
and LaMaire et al. [44]. Therefore the difference pmax – 1/N is the spatial reuse gain (see Figure 1.2).
This simple example illustrates the concepts of collisions, energy-throughput trade-offs, and spatial reuse,
which are present in every MAC scheme.

1.3.7 Modeling
The bases for analysis and simulations and analytical approaches are accurate and tractable models.
Comprehensive network models should include the number of nodes and their relative distribution; their
degree and type of mobility; the characteristics of the wireless link; the volume of traffic injected by the
sources and the lifespan of their interaction; and detailed energy consumption models.
1.3.7.1 Wireless Link
An attenuation proportional to dα, where d is the distance between two nodes and α is the so-called path
loss exponent, is widely accepted as a model for path loss. Alpha ranges from 2 to 4 or even 5 [45],
depending on the channel characteristics (environment, antenna position, frequency). This path loss
model, together with the fact that packets are successfully transmitted if the signal-to-noise-and-inter-
ference ratio (SNIR) is bigger than some threshold [8], results in a deterministic model often used for
analysis of multihop packet networks [23, 26, 41, 42, 46–48]. Thus, the radius for a successful transmission
has a deterministic value, irrespective of the condition of the wireless channel. If only interferers within
a certain distance of the receiver are considered, this “physical model” [23] turns into a “disk model.”
The stochastic nature of the fading channel and thus the fact that the SINR is a random variable are
mostly neglected. However, the volatility of the channel cannot be ignored in wireless networks [5, 8];

© 2006 by Taylor & Francis Group, LLC

 7037_C001.fm Page 9 Tuesday, November 1, 2005 12:46 PM

Opportunities and Challenges in Wireless Sensor Networks 1-9

interference loss pmax –gmax

spatial reuse gain pmax –1/N

p
maximum throughput gmax
∍
collisionless region p [0,pmax /8]
Throughput g
trade-off region p [ pmax /8,pmax ]
∍

g(
p)
1 pmax 1
0 0.5
N
Transmit probability p

FIGURE 1.2 Generic throughput polynomial for a simple random MAC scheme.

Sousa and Silvester have also pointed out the inaccuracy of disk models [49] and it is easily demonstrated
experimentally [50, 51]. In addition, this “prevalent all-or-nothing model” [52] leads to the assumption
that a transmission over a multihop path fails completely or is 100% successful, ignoring the fact that
end-to-end packet loss probabilities increase with the number of hops. Although fading has been con-
sidered in the context of packet networks [53, 54], its impact on the throughput of multihop networks
and protocols at the MAC and higher layers is largely an open problem.
A more accurate channel model will have an impact on most of the metrics listed in Section 1.3.1. In
the case of Rayleigh fading, first results show that the energy benefits of routing over many short hops
may vanish completely, in particular if latency is taken into account [20, 55, 56]. The Rayleigh fading
model not only is more accurate than the disk model, but also has the additional advantage of permitting
separation of noise effects and interference effects due to the exponential distribution of the received
power. As a consequence, the performance analysis can conveniently be split into the analysis of a zero-
interference (noise-analysis) and a zero-noise (interference-analysis) network.
1.3.7.2 Energy Consumption
To model energy consumption, four basic different states of a node can be identified: transmission,
reception, listening, and sleeping. They consist of the following tasks:
• Acquisition: sensing, A/D conversion, preprocessing, and perhaps storing
• Transmission: processing for address determination, packetization, encoding, framing, and maybe
queuing; supply for the baseband and RF circuitry (The nonlinearity of the power amplifier must
be taken into account because the power consumption is most likely not proportional to the
transmit power [56].)
• Reception: Low-noise amplifier, downconverter oscillator, filtering, detection, decoding, error
detection, and address check; reception even if a node is not the intended receiver
• Listening: Similar to reception except that the signal processing chain stops at the detection
• Sleeping: Power supply to stay alive
Reception and transmission comprise all the processing required for physical communication and net-
working protocols. For the physical layer, the energy consumption depends mostly on the circuitry, the
error correction schemes, and the implementation of the receiver [57]. At the higher layers, the choice

© 2006 by Taylor & Francis Group, LLC

 7037_C001.fm Page 10 Tuesday, November 1, 2005 12:46 PM

1-10 Smart Dust

of protocols (e.g., routing, ARQ schemes, size of packet headers, number of beacons and other infra-
structure packets) determines the energy efficiency.
1.3.7.3 Node Distribution and Mobility
Regular grids (square, triangle, hexagon) and uniformly random distributions are widely used analytically
tractable models. The latter can be problematic because nodes can be arbitrarily close, leading to unre-
alistic received power levels if the path attenuation is assumed to be proportional to dα. Regular grids
overlaid with Gaussian variations in the positions may be more accurate. Generic mobility models for
WSNs are difficult to define because they are highly application specific, so this issue must be studied
on a case-by-case basis.
1.3.7.4 Traffic
Often, simulation work is based on constant bitrate traffic for convenience, but this is most probably not
the typical traffic class. Models for bursty many-to-one traffic are needed, but they certainly depend
strongly on the application.

1.3.8 Connectivity
Network connectivity is an important issue because it is crucial for most applications that the network
is not partitioned into disjoint parts. If the nodes’ positions are modeled as a Poisson point process in
two dimensions (which, for all practical purposes, corresponds to a uniformly random distribution), the
problem of connectivity has been studied using the tool of continuum percolation theory [58, 59]. For
large networks, the phenomenon of a sharp phase transition can be observed: the probability that the
network percolates jumps abruptly from almost 0 to almost 1 as soon as the density of the network is
bigger than some critical value. Most such results are based on the geometric disk abstraction. It is
conjectured, though, that other connectivity functions lead to better connectivity, i.e., the disk is appar-
ently the hardest shape to connect [60]. A practical consequence of this conjecture is that fading results
in improved connectivity. Recent work [61] also discusses the impact of interference. The simplifying
assumptions necessary to achieve these results leave many open problems.

1.3.9 Quality of Service

Quality of service refers to the capability of a network to deliver data reliably and timely. A high quantity
of service, i.e., throughput or transport capacity, is generally not sufficient to satisfy an application’s delay
requirements. Consequently, the speed of propagation of information may be as crucial as the throughput.
Accordingly, in addition to network capacity, an important issue in many WSNs is that of quality-of-
service (QoS) guarantees. Previous QoS-related work in wireless networks mostly focused on delay (see,
for example, Lu et al. [62], Ju and Li [63], and Liu et al. [64]). QoS, in a broader sense, consists of the
triple (R, Pe, D), where R denotes throughput; Pe denotes reliability as measured by, for example, bit
error probability or packet loss probability; and D denotes delay. For a given R, the reliability of a
connection as a function of the delay will follow the general curve shown in Figure 1.3.

reliability

100%
2 3
1

delay

FIGURE 1.3 Reliability as a function of the delay. The circles indicate the QoS requirements of different possible
traffic classes.

© 2006 by Taylor & Francis Group, LLC

 7037_C001.fm Page 11 Tuesday, November 1, 2005 12:46 PM

Opportunities and Challenges in Wireless Sensor Networks 1-11

Note that capacity is only one point on the reliability-delay curve and therefore not always a relevant
performance measure. For example, in certain sensing and control applications, the value of information
quickly degrades as the latency increases. Because QoS is affected by design choices at the physical,
medium-access, and network layers, an integrated approach to managing QoS is necessary.

1.3.10 Security
Depending on the application, security can be critical. The network should enable intrusion detection
and tolerance as well as robust operation in the case of failure because, often, the sensor nodes are not
protected against physical mishandling or attacks. Eavesdropping, jamming, and listen-and-retransmit
attacks can hamper or prevent the operation; therefore, access control, message integrity, and confiden-
tiality must be guaranteed.

1.3.11 Implementation
Companies such as Crossbow, Ember, Sensoria, and Millenial are building small sensor nodes with
wireless capabilities. However, a per-node cost of $100 to $200 (not including sophisticated sensors) is
prohibitive for large networks. Nodes must become an order of magnitude cheaper in order to render
applications with a large number of nodes affordable. With the current pace of progress in VLSI and
MEMS technology, this is bound to happen in the next few years. The fusion of MEMS and electronics
onto a single chip, however, still poses difficulties. Miniaturization will make steady progress, except for
two crucial components: the antenna and the battery, where it will be very challenging to find innovative
solutions. Furthermore, the impact of the hardware on optimum protocol design is largely an open topic.
The characteristics of the power amplifier, for example, greatly influence the energy efficiency of routing
algorithms [56].

1.3.12 Other Issues

• Distributed signal processing. Most tasks require the combined effort of multiple network nodes,
which requires protocols that provide coordination, efficient local exchange of information, and,
possibly, hierarchical operation.
• Synchronization and localization. The notion of time is critical. Coordinated sensing and actuating
in the physical world require a sense of global time that must be paired with relative or absolute
knowledge of nodes’ locations.
• Wireless reprogramming. A deployed WSN may need to be reprogrammed or updated. So far,
no networking protocols are available to carry out such a task reliably in a multihop network.
The main difficulty is the acknowledgment of packets in such a joint multihop/multicast
communication.

1.4 Concluding Remarks

Wireless sensor networks have numerous exciting applications in virtually all fields of science and
engineering, including health care, industry, military, security, environmental science, geology, agricul-
ture, and social studies. In particular, the combination with macroscopic or MEMS-based actuators is
intriguing because it permits manipulation of the environment in an unprecedented manner. Researchers
and operators currently face a number of critical issues that need be resolved before these applications
become reality. Wireless networking and distributed data processing of embedded sensing/actuating
nodes under tight energy constraints demand new approaches to protocol design and hardware/software
integration.

© 2006 by Taylor & Francis Group, LLC

 7037_C001.fm Page 12 Tuesday, November 1, 2005 12:46 PM

1-12 Smart Dust

References
1. Sensor networks and applications, IEEE Proc., 8, Aug. 2003.
2. Internet Engineering Task Force, Mobile ad-hoc networks (MANET). See http://www.ietf.org/
html.charters/manet-charter.html.
3. Z.J. Haas et al., Eds., Wireless ad hoc networks, IEEE J. Selected Areas Commun., 17, Aug. 1999.
Special ed.
4. C.E. Perkins, Ed., Ad Hoc Networking. Addison Wesley, Reading, MA, 2000.
5. A.J. Goldsmith and S.B. Wicker, Design challenges for energy-constrained ad hoc wireless networks,
IEEE Wireless Commun., 9, 8–27, Aug. 2002.
6. A. Ephremides, Energy concerns in wireless networks, IEEE Mag. Wireless Commun., 9, 48–59,
Aug. 2002.
7. V. Rodoplu and T.H. Meng, Minimum energy mobile wireless networks, IEEE J. Selected Areas
Commun., 17(8), 1333–1344, 1999.
8. A. Ephremides, Energy concerns in wireless networks, IEEE Wireless Commun., 9, 48–59, Aug. 2002.
9. Bluetooth wireless technology. Official Bluetooth site: http://www.bluetooth.com.
10. S. Tilak, N.B. Abu–Ghazaleh, and W. Heinzelman, A taxonomy of wireless micro-sensor network
models, ACM Mobile Computing Commun. Rev., 6(2), 28–36, 2002.
11. W.B. Heinzelman, A.P. Chandrakasan, and H. Balakrishnan, An application-specific protocol archi-
tecture for wireless microsensor networks, IEEE Trans. Wireless Commun., 1, 660–670, Oct. 2002.
12. J. Kulik, W. Heinzelman, and H. Balakrishnan, Negotiation-based protocols for disseminating
information in wireless sensor networks, Wireless Networks, 8, 169–185, March–May 2002.
13. A.B. McDonald and T.F. Znati, A mobility-based framework for adaptive clustering in wireless ad-
hoc networks, IEEE J. Selected Areas Commun., 17, 1466–1487, Aug. 1999.
14. S.S. Pradhan, J. Kusuma, and K. Ramchandran, Distributed compression in a dense microsensor
network, IEEE Signal Process. Mag., 19, 51–60, Mar. 2002.
15. A. Scaglione and S. Servetto, On the interdependence of routing and data compression in multi-
hop sensor networks, in Proc. ACM Int. Conf. Mobile Comp. Networks (MobiCom’02), Atlanta,
GA, 140–147, Sept. 2002.
16. C. Intanagowiwat, R. Govindan, and D. Estrin, Directed diffusion: a scalable and robust commun-
ication paradigm for sensor networks, in ACM Int. Conf. Mobile Computing Networking (Mobi-
Com’00), Boston, MA, 56–67, Aug. 2000.
17. J.E. Wieselthier, G.D. Nguyen, and A. Ephremides, On the construction of energy-efficient broadcast
and multicast trees in wireless networks, in IEEE INFOCOM, Tel Aviv, Israel, 585–594, Mar. 2000.
18. J.E. Wieselthier, G.D. Nguyen, and A. Ephremides, An insensitivity property of energy-limited
wireless networks for session-based multicasting, in IEEE ISIT, Washington, D.C., June 2001.
19. J. Laneman, D. Tse, and G. Wornell, Cooperative diversity in wireless networks: efficient protocols
and outage behavior, IEEE Trans. Inf. Theory. Accepted for publication. Available at: http://
www.nd.edu/jnl/pubs/it2002.pdf.
20. M. Haenggi, A formalism for the analysis and design of time and path diversity schemes in wireless
sensor networks, in 2nd Int. Workshop Inf. Process. Sensor Networks (IPSN’03), Palo Alto, CA,
417–431, Apr. 2003. Available at http://www.nd.edu/mhaenggi/ipsn03.pdf.
21. C.S. Raghavendra and S. Singh, PAMAS — power aware multi-access protocol with signaling for
ad hoc networks, 1999. ACM Computer Commun. Rev. Available at: http://citeseer.nj.nec.com/
460902.html.
22. C.-K. Toh, Maximum battery life routing to support ubiquitous mobile computing in wireless ad
hoc networks, IEEE Commun. Mag., 39, 138–147, June 2001.
23. P. Gupta and P.R. Kumar, The capacity of wireless networks, IEEE Trans. Inf. Theory, 46, 388–404,
Mar. 2000.
24. P. Gupta and P.R. Kumar, Towards an information theory of large networks: an achievable rate
region, in IEEE Int. Symp. Inf. Theory, Washington, D.C., 159, 2001.

© 2006 by Taylor & Francis Group, LLC

 7037_C001.fm Page 13 Tuesday, November 1, 2005 12:46 PM

Opportunities and Challenges in Wireless Sensor Networks 1-13

25. L.-L. Xie and P.R. Kumar, A network information theory for wireless communication: scaling laws
and optimal operation, Apr. 2002. submitted to IEEE Trans. Inf. Theory. Available at: http://
black1.csl.uiuc.edu/prkumar/publications.html.
26. M. Grossglauser and D. Tse, Mobility increases the capacity of ad-hoc wireless networks, in IEEE
INFOCOM, Anchorage, AL, 2001.
27. D. Tse and S. Hanly, Effective bandwidths in wireless networks with multiuser receivers, in IEEE
INFOCOM, 35–42, 1998.
28. M. Gastpar and M. Vetterli, On the capacity of wireless networks: the relay case, in IEEE INFO-
COM, New York, 2002.
29. G. Mergen and L. Tong, On the capacity of regular wireless networks with transceiver multipacket
communication, in IEEE Int. Symp. Inf. Theory, Lausanne, Switzerland, 350, 2002.
30. S. Toumpis and A. Goldsmith, Capacity regions for wireless ad hoc networks, IEEE Trans. Wireless
Commun., 2, 736–748, July 2003.
31. D.N.C. Tse and S.V. Hanly, Multiaccess fading channels — part I: polymatroid structure, optimal
resource allocation and throughput capacities, IEEE Trans. Inf. Theory, 44(7), 2796–2815, 1998.
32. S.V. Hanly and D.N.C. Tse, Multiaccess fading channels — part II: delay-limited capacities, IEEE
Trans. Inf. Theory, 44(7), 2816–2831, 1998.
33. R. Negi and J.M. Cioffi, Delay-constrained capacity with causal feedback, IEEE Trans. Inf. Theory,
48, 2478–2494, Sept. 2002.
34. R.A. Berry and R.G. Gallager, Communication over fading channels with delay constraints, IEEE
Trans. Inf. Theory, 48, 1135–1149, May 2002.
35. D. Tuninetti, On multiple-access block-fading channels, Mar. 2002. Ph.D. thesis, Institut EURE-
COM. Available at: http://www.eurecom.fr/tuninett/publication.html.
36. J. Broch, D. Maltz, D. Johnson, Y. Hu, and J. Jetcheva, A performance comparison of multi-hop
wireless ad hoc network routing protocols, in ACM Int. Conf. Mobile Computing Networking
(MobiCom), Dallas, TX, 85–97, Oct. 1998.
37. P. Johansson, T. Larsson, and N. Hedman, Scenario-based performance analysis of routing proto-
cols for mobile ad-hoc networks, in ACM MobiCom, Seattle, WA, Aug. 1999.
38. S.R. Das, C.E. Perkins, and E.M. Royer, Performance comparison of two on-demand routing
protocols for ad hoc networks, in IEEE INFOCOM, Mar. 2000.
39. C.R. Lin and J.-S. Liu, QoS Routing in ad hoc wireless networks, IEEE J. Selected Areas Commun.,
17, 1426–1438, Aug. 1999.
40. M. Haenggi, Energy-balancing strategies for wireless sensor networks, in IEEE Int. Symp. Circuits
Syst. (ISCAS’03), Bangkok, Thailand, May 2003. Available at http://www.nd.edu/mhaenggi/
iscas03.pdf.
41. J.A. Silvester and L. Kleinrock, On the capacity of multihop slotted ALOHA networks with regular
structure, IEEE Trans. Commun., COM-31, 974–982, Aug. 1983.
42. L. Hu, Topology control for multihop packet networks, IEEE Trans. Commun., 41(10), 1474–1481,
1993.
43. M. Haenggi, Probabilistic analysis of a simple MAC scheme for ad hoc wireless networks, in IEEE
CAS Workshop on Wireless Communications and Networking, Pasadena, CA, Sept. 2002.
44. R.O. LaMaire, A. Krishna, and H. Ahmadi, Analysis of a wireless MAC protocol with client–server
traffic and capture, IEEE J. Selected Areas Commun., 12(8), 1299–1313, 1994.
45. T.S. Rappaport, Wireless Communications — Principles and Practice, 2nd ed. Prentice Hall,
Englewood Cliffs, NJ.
46. H. Takagi and L. Kleinrock, Optimal transmission ranges for randomly distributed packet radio
terminals, IEEE Trans. Commun., COM-32, 246–257, Mar. 1984.
47. J.L. Wang and J.A. Silvester, Maximum number of independent paths and radio connectivity, IEEE
Trans. Commun., 41, 1482–1493, Oct. 1993.
48. C. Schurgers, V. Tsiatsis, S. Ganeriwal, and M. Srivastava, Optimizing sensor networks in the
energy–latency–density design space, IEEE Trans. Mobile Computing, 1(1), 70–80, 2002.
© 2006 by Taylor & Francis Group, LLC
 7037_C001.fm Page 14 Tuesday, November 1, 2005 12:46 PM

1-14 Smart Dust

49. E.S. Sousa and J.A. Silvester, Optimum transmission ranges in a direct-sequence spread-spectrum
multihop packet radio network, IEEE J. Selected Areas Commun., 8, 762–771, June 1990.
50. D.A. Maltz, J. Broch, and D.B. Johnson, Lessons from a full-scale multihop wireless ad hoc network
testbed, IEEE Personal Commun., 8, 8–15, Feb. 2001.
51. D. Ganesan, B. Krishnamachari, A. Woo, D. Culler, D. Estrin, and S. Wicker, An empirical study
of epidemic algorithms in large scale multihop wireless networks, 2002. Intel Research Report IRB-
TR-02-003. Available at www.intel-research.net/Publications/Berkeley/05022002170319.pdf.
52. T.J. Shepard, A channel access scheme for large dense packet radio networks, in ACM SIGCOMM,
Stanford, CA, Aug. 1996. Available at: http://www.acm.org/sigcomm/sigcomm96/papers/shep-
ard.ps.
53. M. Zorzi and S. Pupolin, Optimum transmission ranges in multihop packet radio networks in the
presence of fading, IEEE Trans. Commun., 43, 2201–2205, July 1995.
54. Y.Y. Kim and S. Li, Modeling multipath fading channel dynamics for packet data performance
analysis, Wireless Networks, 6, 481–492, 2000.
55. M. Haenggi, On routing in random rayleigh fading networks, IEEE Trans. Wireless Commun.,
2003. Submitted for publication. Available at http://www.nd.edu/mhaenggi/routing.pdf.
56. M. Haenggi, The impact of power amplifier characteristics on routing in random wireless networks,
in IEEE Global Commun. Conf. (GLOBECOM’03), San Francisco, CA, Dec. 2003. Available at
http://www.nd.edu/mhaenggi/globecom03.pdf.
57. H. Meyr, M. Moenecleay, and S.A. Fechtel, Digital Communication Receivers: Synchronization,
Channel Estimation, and Signal Processing. Wiley Interscience, 1998.
58. R. Meester and R. Roy, Continuum Percolation. Cambridge University Press, New York, 1996.
59. B. Bollobás, Random Graphs, 2nd ed. Cambridge University Press, New York, 2001.
60. L. Booth, J. Bruck, M. Cook, and M. Franceschetti, Ad hoc wireless networks with noisy links, in
IEEE Int. Symp. Inf. Theory, Yokohama, Japan, 2003.
61. O. Dousse, F. Baccelli, and P. Thiran, Impact of interferences on connectivity in ad-hoc networks,
in IEEE INFOCOM, San Francisco, CA, 2003.
62. S. Lu, V. Bharghavan, and R. Srikant, Fair scheduling in wireless packet networks, IEEE/ACM
Trans. Networking, 7, 473–489, Aug. 1999.
63. J.-H. Ju and V.O.K. Li, TDMA scheduling design of multihop packet radio networks based on Latin
squares, IEEE J. Selected Areas Commun., 1345–1352, Aug. 1999.
64. H. Luo, S. Lu, and V. Bharghavan, A new model for packet scheduling in multihop wireless
networks, in ACM Int. Conf. Mobile Computing Networking (MobiCom’00), Boston, MA, 76–86,
2000.
65. H. Luo, P. Medvedev, J. Cheng, and S. Lu, A self-coordinating approach to distributed fair queueing
in ad hoc wireless networks, IEEE INFOCOM, Anchorage, Apr. 2001.
66. A.E. Gamal, C. Nair, B. Prabhakar, E. Uysal-Biyikoglu, and S. Zahedi, Energy-efficient scheduling
of packet transmissions over wireless networks, IEEE INFOCOM, New York, 2002, pp. 1773–1782.
67. S. Bhatnagar, B. Deb, and B. Nath, Service differentiation in sensor networks, Fourth International
Symposium on Wireless Personal Multimedia Communications, Sept. 2001.
68. V. Kanodia, C. Li, A. Sabharwal, B. Sadeghi, and E. Knightly, Distributed multi-hop scheduling
and medium access with delay and throughput constraints, ACM MobiCom, Rome, July 2001.
69. A. Striegel and G. Manimaran, Best-effort scheduling of (m, k)-firm real-time streams in multihop
networks, Workshop of Parallel and Distributed Real-Time Systems (WPDRTS) at IPDPS 2000,
Apr. 2000.

© 2006 by Taylor & Francis Group, LLC

 7037_C002.fm Page 1 Tuesday, November 1, 2005 12:20 PM

2
Next-Generation
Technologies to Enable
Sensor Networks*

2.1 Introduction ...................................................................... 2-1

Geolocation and Identification of Mobile Targets • Long-Term
Architecture
2.2 Goals for Real-Time Distributed Network Computing
for Sensor Data Fusion ..................................................... 2-5
2.3 The Convergence of Networking and Real-Time
Computing......................................................................... 2-6
Guaranteeing Network Resources • Guaranteeing Storage
Buffer Resources • Guaranteeing Computational Resources
Joel I. Goodman 2.4 Middleware ...................................................................... 2-11
MIT Lincoln Laboratory Control and Command of System • Parallel Processing
Albert I. Reuther 2.5 Network Resource Management .................................... 2-11
Graph Generator • Metrics Object • Graph Search • NRM
MIT Lincoln Laboratory
Agents • Sensor Interface • Mapping Database • Topology
David R. Martinez Database • NRM Federation • NRM Fault Tolerance
MIT Lincoln Laboratory 2.6 Experimental Results....................................................... 2-16

2.1 Introduction
Several important technical advances make extracting more information from intelligence, surveillance,
and reconnaissance (ISR) sensors very affordable and practical. As shown in Figure 2.1, for the radar
application the most significant advancement is expected to come from employing collaborative and
network centric sensor netting. One important application of this capability is to achieve ultrawideband
multifrequency and multiaspect imaging by fusing the data from multiple sensors. In some cases, it is
highly desirable to exploit multimodalities, in addition to multifrequency and multiaspect imaging.
Key enablers to fuse data from disparate sensors are the advent of high-speed fiber and wireless
networks and the leveraging of distributed computing. ISR sensors need to perform enough on-board
computation to match the available bandwidth; however, after some initial preprocessing, the data will
be distributed across the network to be fused with other sensor data so as to maximize the information
content. For example, on an experimental basis, MIT Lincoln Laboratory has demonstrated a virtual
radar with ultrawideband frequency [1]. Two radars, located at the Lincoln Space Surveillance Complex

*This work is sponsored by the United States Air Force under Air Force contract F19628-00-C-002. Opinions,
interpretations, conclusions, and recommendations are those of the authors and are not necessarily endorsed by the
U.S. government.

2-1
© 2006 by Taylor & Francis Group, LLC
 Another Random Document on
Scribd Without Any Related Topics
by another meeting she will have left the college—left it for the lack
of just such recognition as membership in the society will give her."
Cornelia Burt was a born orator. Never was she so happy as when
she felt an audience, however small, given over to her, eyes and
ears, for the moment. She stood straight as a reed, and looked
easily over their faces, holding by very force of personality their
attention. She spoke without the slightest hesitation, yet perfectly
simply and after no set form. Insensibly the girls around her felt
conviction in her very presence: they agreed with her against their
will, while she was speaking.
"Before I go any farther, I want to tell you that Miss Hastings is no
friend of mine," said Neal. "I hardly know her. Only lately I have
learned the circumstances that led me to take this step. I feel that I
must do this thing. I feel that we are letting go from the college a
girl whose failure in life, if she fails, will be in our hands. We can
elect these others later: Winifred Hastings leaves the college next
week. And, speaking as editor of the college paper, I must say that
she carries with her some of the best literary material in the college.
You ask me why we have never seen it—I tell you, because she is a
girl who needs encouragement, and she has never had it. She can
do her best only when it is called for. Some of you may think you
know her—may think that she is proud and solitary and
disagreeable: she is not. This is the real girl!"
And, stepping farther into the circle, Cornelia, by an effort of
memory she has never equalled since, told them, with the simplest
eloquence, the pathetic story of Winifred Hastings' life, as she had
written it. She did not comment—she only related. Her keen literary
appreciation had caught the most effective parts, and she had the
dramatic sense to which every successful speaker owes so much.
Under her touch the haughty, solitary figure of a scarcely known girl
melted away before them, and they saw a baffled, eager, hungry
soul that had fought desperately, and was going silently away—
beaten.
Cornelia Burt had made speeches before, and she made them
afterward, to larger and more excited college audiences, but she
never held so many hearts in her hand as she did that night. She
was not a particularly unselfish girl, but no one who heard her then
ever called her egotistic afterward. Her whole nature was thrown
with all its force into this fight—for it was a fight.
Perhaps there is nowhere an audience less sentimental and more
critical than a group of clever college girls. They see clearly for the
most part, and, like all clever youth, somewhat cruelly. They object
to being ruled by any but their chosen, and however they admired
her, Cornelia was not their chosen leader. It was not because her
speech was able, but because it was so evident that she believed
herself only the means of preventing a calamity that she was striving
with all her soul to avert, that she impressed them so deeply.
For she did impress them. When she ended, it was very quiet in the
room. "I have broken a confidence in telling this," she said. "The girl
herself would rather die than have you know it, I'm sure, and now—I
feel afraid. It has been a bold stroke; if I have lost, I shall never
forgive myself. But oh! I cannot have her go!"
She sat down quickly and stared into her lap. The spell of her voice
was gone, the girls looked at each other, and a tall, keen-eyed girl
with glasses got up. "I wish to say," she said, "that while Miss Burt's
story is terribly convincing, still this may be a little exaggerated, and,
at any rate, think of the precedent! If this should be done very often
—"
"But it won't be!" cried some one with a somewhat husky voice, and
Patsy rudely interrupted the speaker. Dear Patsy! She crushed her
handkerchief in her hand and said good-by to Kate: she would have
liked to put her pin in Kate's shirt-waist, and now—now Phi Kappa
would get her! When Patsy spoke, it was with the voice of eleven,
for she carried at least ten of the leading set in the Alpha with her.
"I think we are all very glad to realize that there won't be many such
cases—most people have compensations—we ought to be willing to
break the constitution again for such a thing, anyhow—and, Miss
President, I move that Miss Hastings be voted upon by acclamation!"
"I second the motion," said the vice-president, quickly.
"It is moved and seconded that Miss Hastings be voted upon by
acclamation," said the president. "All in favor—"
"Miss Hastings has yet to be proposed," said some one, after the
vote.
The president looked at Cornelia.
"I propose Winifred Hastings, '9-, as a member of the Alpha Society,"
said Cornelia, with flaming cheeks and downcast eyes. She dared
not look at them. Were they going to punish her? She heard the
motion announced, she heard the name put up.
"All in favor please signify by rising," said the president, and only
when the Alpha rose in a body did Cornelia lift her eyes.
They were all looking at her, and she stepped a little back.
"I cannot thank you," she said, so low that they leaned forward to
hear. "It was no affair of mine, as I said. But—I think you—we—shall
never regret this election." And then they applauded so loudly that
the freshmen on the campus could not forbear peeping under the
blinds to see what they were doing. They saw only the president,
however, as she stepped back to the table and said with an air of
relief—for, after all, emotion is very wearing—"We will now proceed
to the literary programme of the evening!"
"But Neal, dear," said Patsy, as they settled themselves to listen, "do
you think she'll stay? (Oh, Neal! I'm so proud of you!)"
"Shut up, Patsy!" said Neal, rudely. Then, as she thought of what
Miss Henderson had told her of Winifred Hastings: "You are the only
girl whose friendship"—she blushed. Then, assuming a bored
expression, she looked at the girl who was reading. "I fear there's no
doubt she will!" said Cornelia Burt.

THE THIRD STORY

 MISS BIDDLE OF BRYN MAWR

III
MISS BIDDLE OF BRYN MAWR
"I wouldn't have minded so much," explained Katherine, dolefully,
and not without the suspicion of a sob, "if it wasn't that I'd asked
Miss Hartwell and Miss Ackley! I shall die of embarrassment—I shall!
Oh! why couldn't Henrietta Biddle have waited a week before she
went to Europe?"
Her room-mate, Miss Grace Farwell, sank despairingly on the pile of
red floor-cushions under the window. "Oh, Kitten! you didn't ask
them? Not really?" she gasped, staring incredulously at the tangled
head that peered over the screen behind which Katherine was
splashily conducting her toilet operations.
"But I did! I think they're simply grand, especially Miss Hartwell, and
I'll never have any chance of meeting her, I suppose, and I thought
this was a beautiful one. So I met her yesterday on the campus and
I walked up to her—I was horribly scared, but I don't think I showed
it—and, said I, 'Oh, Miss Hartwell, you don't know me, of course, but
I'm Miss Sewall, '9-, and I know Henrietta Biddle of Bryn Mawr, and
she's coming to see me for two or three days, and I'm going to
make a little tea for her—very informal—and I've heard her speak of
you and Miss Ackley as about the only girls she knew here, and I'd
love to have you meet her again!'"
Miss Farwell laughed hysterically. "And did she accept?" she inquired.
Katherine wiped her face for the third time excitedly. "Oh, yes! She
was as sweet as peaches and cream! 'I shall be charmed to meet
Miss Biddle again, and in your room, Miss Sewall,' she said, 'and
shall I bring Miss Ackley?' Oh, Grace, she's lovely! She is the most—"
"Yes, I've no doubt," interrupted Miss Farwell, cynically; "all the
handsome seniors are. But what are you going to say to her to-
day?"
Katherine buried her yellow head in the towel. "I don't know! Oh,
Grace! I don't know," she mourned. "And they say the freshmen are
getting so uppish, anyway, and if we carry it off well, and just make
a joke of it, they'll think we're awfully f-f-fresh!" Here words failed
her, and she leaned heavily on the screen, which, as it was old and
probably resented having been sold third-hand at a second-hand
price, collapsed weakly, dragging with it the Bodenhausen Madonna,
a silver rack of photographs, and a Gibson Girl drawn in very black
ink on a very white ground.
"And if we are apologetic and meek," continued Miss Farwell, easily,
apparently undisturbed by the confusion consequent to the downfall
of a piece of furniture known to be somewhat erratic, "they'll laugh
at us or be bored. We shall be known as the freshmen who invited
seniors and Faculty and town-people to meet—nobody at all! A
pretty reputation!"
"But, Grace, we couldn't help it! Such things will happen!" Katherine
was pinning the Gibson Girl to the wall, in bold defiance of the
matron's known views on that subject.
"Yes, of course. But they mustn't happen to freshmen!" her room-
mate returned sententiously. "How many Faculty did you ask?"
"I asked Miss Parker, because she fitted Henrietta for college, at
Archer Hall, and I asked Miss Williams, because she knows
Henrietta's mother—Oh! Miss Williams will freeze me to death when
she comes here and sees just us!—and I asked Miss Dodge, because
she knows a lot of Bryn Mawr people. Then Mrs. Patton on Elm
Street was a school friend of Mrs. Biddle's, and—oh! Grace, I can't
manage them alone! Let's tell them not to come!"
"And what shall we do with the sandwiches? And the little cakes?
And the lemons that I sliced? And the tea-cups and spoons I
borrowed? And that pint of extra thick cream?" Miss Farwell checked
off these interesting items on her fingers, and kicked the floor-
cushions to point the question.
"Oh! I don't know! Isn't there any chance—"
"No, goosey, there isn't. See here!" Grace pulled down a letter with a
special delivery stamp from the desk above her head, and read with
emphasis:
DEAR Kitten,—Just a line to say that Aunt Mary has sent
for me at three days' notice to go to Paris with her for a
year. It's now or never, you know, and I've left the college,
and will come back to graduate with '9-. So sorry I can't
see you before I go. Had looked forward to a very
interesting time, renewing my own freshman days, and all
that. Please send my blue cloth suit right on to
Philadelphia C. O. D. when it comes to you. I hope you
hadn't gotten anything up for me.
With much love,
Henrietta Biddle.
Bryn Mawr, March 5.
"I don't think there's much chance, my dear."
"No," said Katherine, sadly, and with a final pat administered to the
screen, which still wobbled unsteadily. "No, I suppose there isn't.
And it's eleven o'clock. They'll be here at four! Oh! and I asked that
pretty junior, Miss Pratt, you know. Henrietta knew her sister. She
was in '8-."
"Ah," returned Miss Farwell, with a suspicious sweetness, "why didn't
you ask a few more, Katherine, dear? What with the list we made
out together and these last extra ones—"
"But I thought there wasn't any use having the largest double room
in the house, if we couldn't have a decent-sized party in it! And think
of all those darling, thin little sandwiches!—Oh well, we might just as
well be sensible and carry the thing through, Gracie! But I am just as
afraid as I can be: I tell you that. And Miss Williams will freeze me
stiff." The yellow hair was snugly braided and wound around by now,
and a neat though worried maiden sat on the couch and punched
the Harvard pillow reflectively.
"Never mind her, Kitten, but just go ahead. You know Caroline Wilde
said it was all right to ask her if she was Miss Biddle's mother's
friend, and there wasn't time to take her all around, and you know
how nice Miss Parker was about it. We can't help it, as you say, and
we'll go and get the flowers as we meant to. Have you anything this
hour?"
With her room-mate to back her, to quote the young lady herself,
Miss Sewall felt equal to almost any social function. Terrifying as her
position appeared—and strangely enough, the seniors appalled her
far more than the Faculty—there was yet a certain excitement in the
situation. What should she say to them? Would they be kind about
it, or would they all turn around and go home? Would they think—
"Oh, nonsense!" interrupted Grace the practical, as these doubts
were thrust upon her. "If they're ladies, as I suppose they are, of
course they'll stay and make it just as pleasant for us as they can.
They'll see how it is. Think what we'd do, ourselves, you know!"
They went down the single long street, with the shops on either
side, a red-capped, golf-caped pair of friends, like nine hundred
other girls, yet different from them all. And they chattered of Livy
and little cakes and Trigonometry and pleated shirt-waists and
basket-ball and Fortnightly Themes like all the others, but in their
little way they were very social heroines, setting their teeth to carry
by storm a position that many an older woman would have found
doubtful.
They stopped at a little bakery, well down the street, to order some
rolls for the girl across the hall from them, who had planned to
breakfast in luxury and alone on chocolate and grape-fruit the next
morning. "Miss Carter, 24 Washburn," said Grace, carelessly, when
Katherine whispered, "Look at her! Isn't that funny? Why, Grace, just
see her!"
"See who—whom, I mean? (only I hate to say 'whom.') Who is it,
Kitten?"
Katherine was staring at the clerk, a tall, handsome girl, with masses
of heavy black hair and an erect figure. As she went down to the
back of the shop again, Katherine's eyes followed her closely.
"It's that girl that used to be in the Candy Kitchen—don't you
remember? I told you then that she looked so much like my friend
Miss Biddle. And then the Candy Kitchen failed and I suppose she
came here. And she's just Henrietta's height, too. You know
Henrietta stands very straight and frowns a little, and so did this girl
when you gave Alice's number and she said, 'Thirty-four or twenty-
four?' Isn't it funny that we should see her now?—Oh, dear! If only
she were Henrietta!"
Grace stared at the case of domestic bread and breathed quickly.
"Does she really look like her, Kitten?" she said.
"Oh yes, indeed. It's quite striking. Henrietta's quite a type, you
know—nothing unusual, only very dark and tall and all that. Of
course there are differences, though."
"What differences?" said Grace, still looking intently at the domestic
bread.
"Oh, Henrietta's eyes are brown, and this girl's are black. And
Henrietta hasn't any dimple, and her hands are prettier. And
Henrietta's waist isn't so small, and she hasn't nearly so much hair, I
should say. But then, I haven't seen her for a year, and probably
there's a greater difference than I think."
"How long is it since those seniors and the Faculty saw Henrietta?"
said Grace, staring now at a row of layer chocolate-cakes.
Her room-mate started. "Why—why, Grace, what do you mean? It's
two years, Henrietta wrote, I think. And Miss Parker and Miss
Williams haven't seen her for much longer than that. But—but—you
don't mean anything, Grace?"
Grace faced her suddenly. "Yes," she said, "I do. You may think that
because I just go right along with this thing, I don't care at all. But I
do. I'm awfully scared. I hate to think of that Miss Ackley lifting her
eyebrows—the way she will! And Miss Hartwell said once when
somebody asked if she knew Judge Farwell's daughter, 'Oh, dear me
—I suppose so! And everybody else in her class—theoretically! But
practically I rarely observe them!' Ugh! She'll observe me to-day, I
hope!"
"Yes, dear, I suppose she will. And me too. But—"
"Oh, yes! But if nobody knows how Miss Biddle looks, and she was
going to stay at the hotel, anyway, and it would only be for two
hours, and everything would be so simple—"
Katherine's cheeks grew very red and her breath came fast. "But
would we dare? Would she be willing? Would it be—"
"Oh, my dear, it's only a courtesy! And everybody will think it's all
right, and the thing will go beautifully, and Miss Biddle, if she has
any sense of humor—"
"Yes, indeed! Henrietta would only be amused—oh, so amused! And
it would be such a heavenly relief after all the worry. We could send
her off on the next train—Henrietta, you know—and dress makes
such a difference in a girl!"
"And I think she would if we asked her just as a favor—it wouldn't
be a question of money! Oh, Katherine! I could cry for joy if she
would!"
"She'd like to, if she has any fun in her—it would be a game with
some point to it! And will you ask her, or shall I?"
They were half in joke and half in earnest: it was a real crisis to
them. They were only freshmen, and they had invited the seniors
and the Faculty. And two of the most prominent seniors! Whom they
hadn't known at all! They had a sense of humor, but they were
proud, too, and they had a woman's horror of an unsuccessful social
function. They felt that they were doomed to endless joking at the
hands of the whole college, and this apprehension, though probably
exaggerated, nerved them to their coup d'état.
Grace walked down the shop. "I will ask her," she said.
Katherine stood with her back turned and tried not to hear. Suppose
the girl should be insulted? Suppose she should be afraid? Now that
there was a faint hope of success, she realized how frightened and
discouraged she had been. For it would be a success, she saw that.
Nobody would have had Miss Biddle to talk with for more than a few
minutes anyhow, they had asked such a crowd. And yet she would
have been the centre of the whole affair.
"Katherine," said a voice behind her, "let me introduce Miss Brooks,
who has consented to help us!"
Katherine held out her hands to the girl. "Oh, thank you! thank you!"
she said.
The girl laughed. "I think it's queer," she said, "but if you are in such
a fix, I'd just as lief help you as not. Only I shall give you away—I
shan't know what to say."
Grace glanced at Katherine. Then she proved her right to all the
praise she afterward accepted from her grateful room-mate. "That
will be very easy," she said sweetly. "Miss Biddle, whom you will—
will represent, speaks very rarely: she's not at all talkative!"
Katherine gasped. "Oh, no!" she said eagerly, "she's very
statuesque, you know, and keeps very still and straight, and just
looks in your eyes and makes you think she's talking. She says
'Really?' and 'Fancy, now!' and 'I expect you're very jolly here,' and
then she smiles. You could do that."
"Yes, I could do that," said the girl.
"Can you come to the hotel right after dinner?" said Grace,
competently, "and we'll cram you for an hour or so on Miss Biddle's
affairs."
The girl laughed. "Why, yes," she said, "I guess I can get off."
So they left her smiling at them from the domestic bread, and at two
o'clock they carried Miss Henrietta Biddle's dress-suit case to the
hotel and took Miss Brooks to her room. And they sat her on a sofa
and told her what they knew of her alma mater and her relatives
and her character generally. And she amazed them by a very
comprehensive grasp of the whole affair and an aptitude for mimicry
that would have gotten her a star part in the senior dramatics. With
a few corrections she spoke very good English, and "as she'd only
have to answer questions, anyhow, she needn't talk long at a time,"
they told each other.
She put up her heavy hair in a twisted crown on her head, and they
put the blue cloth gown on her, and covered the place in the front,
where it didn't fit, with a beautiful fichu that Henrietta had
apparently been led of Providence to tuck in the dress-suit case. And
she rode up in a carriage with them, very much excited, but with a
beautiful color and glowing eyes, and a smile that brought out the
dimple that Henrietta never had.
They showed her the room and the sandwiches and the tea, and
they got into their clothes, not speaking, except when a great box
with three bunches of English violets was left at their door with
Grace's card. Then Katherine said, "You dear thing!" And Miss Brooks
smiled as they pinned hers on and said softly, "Fancy, now!"
And then they weren't afraid for her any more.
When the pretty Miss Pratt came, a little after four, with Miss
Williams, she smiled with pleasure at the room, all flowers and tea
and well-dressed girls, with a tall, handsome brunette in a blue
gown with a beautiful lace bib smiling gently on a crowd of
worshippers, and saying little soft sentences that meant anything
that was polite and self-possessed.
Close by her was her friend Miss Sewall, of the freshman class, who
sweetly answered half the questions about Bryn Mawr that Miss
Biddle couldn't find time to answer, and steered people away who
insisted on talking with her too long. Miss Farwell, also of the
freshman class, assisted her room-mate in receiving, and passed
many kinds of pleasant food, laughing a great deal at what
everybody said and chatting amicably and unabashed with the two
seniors of honor, who openly raved over Miss Biddle of Bryn Mawr.
As soon as Katherine had said, "May I present Miss Hartwell—Miss
Ackley?" they took their stand by the stately stranger and talked to
her as much as was consistent with propriety.
"Isn't she perfectly charming!" they said to Miss Parker, and "Yes,
indeed," replied that lady, "I should have known Netta anywhere.
She is just what I had thought she would be!"
And Miss Williams, far from freezing the pretty hostess, patted her
shoulder kindly. "Henrietta is quite worth coming to see," she said
with her best and most exquisite manner. "I have heard of the Bryn
Mawr style, and now I am convinced. I wish all our girls had such
dignity—such a feeling for the right word!"
And they had the grace to blush. They knew who had taught
Henrietta Biddle Brooks that right word!
At six o'clock Miss Biddle had to take the Philadelphia express. She
had only stopped over for the tea. And so the girls of the house
could not admire her over the supper-table. But they probably
appreciated her more. For after all, as they decided in talking her
over later, it wasn't so much what she said, as the way she looked
when she said it!
But only a dress-suit case marked H. L. B. took the Philadelphia
express that night, and a tall, red-cheeked girl in a mussy checked
suit left the hotel with a bunch of violets in her hand and a
reminiscent smile on her lips.
"We simply can't thank you; we haven't any words. You've helped us
give the nicest party two freshmen ever gave, if it is any pleasure to
you to know that," said Katherine. "And now you're only not to
speak of it."
"Oh, no! I shan't speak of it," said the girl. "You needn't be afraid.
Nobody that I'd tell would believe me, very much, anyhow. I'm glad
I could help you, and I had a lovely time—lovely!"
She smiled at them: the slow, sweet smile of Henrietta Biddle, late
of Bryn Mawr. "You College ladies are certainly queer—but you're
smart!" said Miss Brooks of the bakery.

THE FOURTH STORY

 BISCUITS EX MACHINA

IV
BISCUITS EX MACHINA
B. S. Kitts—this was the signature she had affixed in a neat clerical
backhand to all her written papers since the beginning of freshman
year; and she had of course been called Biscuits as soon as she had
found her own particular little set of girls and settled down to that
peculiar form of intimacy which living in barracks, however
advantageously organized, necessitates. She had a sallow irregular
face, fine brown eyes surrounded with tiny wrinkles, a taste for
Thackeray, and a keen sense of humor. It was the last which was
subsequently responsible for this story about her.
She was quite unnoticed for two or three years, which is a very good
thing for a girl. During that time she quietly took soundings and laid
in material, presumably, for those satiric characterizations which
were the terror of her undergraduate enemies and the concealed
discomfort of those in high places. During her junior year she began
to be considered terribly clever, and though she was never what is
known as a Prominent Senior, she had her little triumphs here and
there, and in the matter of written papers she was a source of great
comfort to those whom custom compels to demand such tributes.
She was the kind of girl who, though well known in her own class, is
quite unobserved of the lower classes, and this, if it deprived her of
the admirations and attentions bestowed on the prominent, saved
her the many worries and wearinesses incident to trying to please
everybody at once—the business of the over-popular. She had a
great deal of time, which may seem absurd, but which is really quite
possible if one keeps positively off committees, is neither musical nor
athletic, and shuns courses involving laboratory work. It is of great
assistance also in this connection to elect English Literature
copiously, when one has read most of the works in question and can
send home for the reference books, thus saving an immense amount
of fruitless loitering about crowded libraries.
Biscuits employed the time thus gained in a fashion apparently
purposeless. She loafed about and observed, with Vanity Fair under
one arm and an apple in the other hand. She was never the subject
or the object of a violent friendship; she was one of five or six clever
girls who hung together consistently after sophomore year, bickering
amicably and indulging in mutual contumely when together,
defending one another promptly when apart. The house president
spoke of them bitterly as blasé and critical; the lady-in-charge
remarked suspiciously the unusual chance which invariably seated
them together at the end of the table at the regular drawing for
seats; the collector for missions found them sceptical and inclined to
ribaldry if pushed too far; but the Phi Kappa banked heavily on their
united efforts, and more than usually idiotic class meetings meekly
bowed to what they themselves scornfully referred to afterward as
"their ordinary horse-sense."
One of the members of this little group was Martha Augusta
Williams. Sometimes she retired from it and devoted herself to
solitude, barely replying to questions and obscurely intimating that
to ennui such as hers the prattle of the immature and inexperienced
could hardly be supposed even by themselves to be endurable;
sometimes she returned to it with the air of one willing to impart to
such a body the mellow cynicism of a tolerant if fatigued femme du
monde. In the intervals of her retirement she wrote furiously at
long-due themes, which took the form of Richard Harding Davis
stories—she did them very well—or modern and morbid verses of a
nature to disturb the more conservative of those who heard them. At
any expression of disturbance Martha would elaborately suppress a
three-volume smile and murmur something about "meat for babes;"
a performance which delighted her friends—especially Biscuits—
beyond measure. Her shelves bristled with yellow French novels, and
on her bureau a great ivory skull with a Japanese paper snake
carelessly twined through it impressed stray freshmen tremendously.
She cut classes elaborately and let her work drop ostentatiously in
the middle of the term, appearing at mid-years with ringed eyes and
an air of toleration strained to the breaking point. She slept till nine
and wandered lazily to coffee and toast at Boyden's an hour later, at
least three times a week, with an air that would have done credit to
one of Ouida's noblemen.
And yet, in spite of all this, Martha was not happy. The disapproval
of the lady-in-charge, the suspicions of the freshmen, the periodical
discussions with members of the Faculty, who "regretted to be
obliged to mark," etc., "when they realized perfectly that she was
capable," etc.,—all these alleviated her trouble a little, but the facts
remained that her own particular set would never treat her seriously,
and that her name was Martha Augusta Williams. Fancy feeling such
feelings, and thinking such thoughts, and bearing the name of
Martha Augusta Williams! It is, to say the least, dispiriting. And
nobody had ever called her anything else. Harriet Williams was
called, indifferently, Billie and Willie and Sillie. Martha Underhill took
her choice of Mattie, Nancy, and Sister. A girl whose name was Anna
Augusta. Something had been hailed as Gustavus Adolphus from her
freshman year on; but below her most daring flights of fiction must
ever appear those three ordinary, not to say stodgy, names. That
alone would have soured a temper not too inclined to regard life
with favor.
Martha might have lived down the name, but she was assured that
never while Bertha Kitts remained alive would she be able to appear
really wickedly interesting. For Biscuits would tell the Story. Tell it
with variations and lights and shades and explanations adapted to
the audience. And it never seemed to pall. Yet it was simple—
horribly simple.
Martha had invited a select body of sophomores to go with her to
the palm-reader's. There were two clever ones, who vastly admired
her Richard Harding Davis tales, two curious ones, who openly
begged for her opinions and thrilled at her epigrams on Love and
Life and Experience, and, in an evil hour, the Sutton twins, whom
she admitted into the occasion partly to impress them, and partly so
that if anything really fascinating should come to light, Kate Sutton
could impart it to her room-mate, Patsy Pattison.
When they were assembled in the palm-reader's parlor, Martha
gravely motioned the others to go before her, and they took their
innocent turns before the little velvet cushion. The Twins were
admirably struck off in a few phrases, to the delight of their friends,
and the palm-reader's reputation firmly established. In the case of
one of the curious girls, peculiar and private events were hinted at
that greatly impressed her, for "how could she have known that,
girls?" The clever girls were comforted with fame and large
"scribbler's crosses," also wealthy marriages and social careers, but
they looked enviously at Martha, nevertheless, and she smiled
maternally on them, as was right. There remained only the other
quiet little girl, and she modestly suggested waiting till another day,
"so there'll be lots of time for yours, Miss Williams;" but Martha
smiled kindly and waved her to the seat, suggesting that hers might
not be a long session, with an amused glance at the empty, little
pink palm.
The palm-reader turned and twisted and patted and asked her age,
and finally announced that it was a remarkable hand. The dying
interest revived, and even Martha's eyebrows went up with
amazement as the seer spoke darkly of immense influence; tact to
the nth degree; unusual amount of experience, or at the least,
"intuitional discoveries;" two great artistic means of expression;
previous affairs of the heart, and an inborn capacity for ruling the
destinies of others—marked resemblance to the hands of Cleopatra
and Sara Bernhardt. It was hands like that that moved the world,
she said. The sophomores regarded their friend with interest and
awe, noted that she blushed deeply at portions of the revelation,
recollected her Sunday afternoon improvisations at the piano and
her request for a more advanced course in harmony, and attached a
hitherto unfelt importance to her heavy mails.
Martha may have regretted her politeness, but she smothered her
surprise, sank, with an abstracted air, upon the chair before the
cushion, and with a face from which all emotion had been withdrawn
and eyes which defied any wildest revelation to disturb their settled
ennui, awaited the event. The palm-reader glanced at the back of
the slim hand, noted the face, touched the finger tips.
"How old are you, please?" she asked. Martha wearily announced
that she was twenty-one. She was conscious of its being a terribly
ordinary age. The palm-reader nodded. "Ah!" she said easily. "Well,
come to me again in a year or two. I can't really tell much now."
Martha gasped at her. "You can't tell much!"
The palm-reader took her hand again. "There's nothing much to
tell!" she explained. "The hand isn't really developed yet—it's the
opposite from the last young lady's, you might say."
She became conscious of a cold silence through the room, and
added a few details. "There's a good general ability; no particular
line of talent, I should say; orderly, regular habits; a very kind heart;
I can't see any events in particular; you've led a very quiet life, I
should say; fond of reading; I shouldn't say you'd met many people
or travelled much"—she scrutinized the hand more closely—"you'll
probably develop a strong religious feeling—"
She stopped and smiled deprecatingly. "It is really impossible to say
very much," she said, "just now. It's what we call an immature
hand!"
For months after that Martha woke in the night and tried to forget
the nightmare of a terrible figure that led her to an amphitheatre of
grinning enemies, and leered at her: It's what we call an Immature
Hand! She could have suppressed the others, but the Sutton twins
were beyond earthly and human suppression. It seemed to her that
she never met them or passed them in a corridor without hearing
their jovial assurance: "Oh, Martha Williams is all right! Why, the
idea! She's as kind a girl as ever lived—she's nothing like that story.
Gracious, no! She's never been to Paris—she lives in Portland. Why,
her father's a Sunday School Superintendent! Oh, bother! She's as
good as Alberta May, every bit! She has a strong religious—" and
somebody passed on, assured—heavens, perhaps admiring her
character! At such times Martha would read furiously in her French
novels or regard the skull pensively or sit up all night, which
annoyed her room-mate and the lady-in-charge. Her room-mate was
an absolutely unimportant person, and does not come into the story
at all.
It is now time to revert to the Twins. When they appeared in the
house, two solemn-eyed, pigtailed imps from Buffalo, they were
packed away together in a double room on the third floor, and
except for their amazing resemblance, were absolutely unnoted. The
matron uneasily fancied a certain undue disturbance on the third
floor, the evening of their arrival, but on going to that level she
found all as still as the grave, and immediately went back
downstairs. It is only due to her, however, to say that she never
again made such an error. From that time on any abnormal quiet in
the house was to her as the trumpet to the war-horse; and she
mounted unerringly to the all-too-certain scene of action. Their plans
for the first year were rather crude, though astonishingly effective at
the time. It was they who invented the paper bag of water dropped
from the fourth floor to burst far below, and waken the house with
the most ghastly hollow explosion; it was they who let a pair of
scissors down two flights to tap against the pane of an unfortunate
enemy in the senior class, and send her into convulsions of nervous
and, as they said, guilty fear. It was they who stuck new caramels to
their door-knob, and oblivious to the matron's admonitions of the
hour, waited till in exasperation she seized the knob, when they met
her disgust with soap and apologies; it was they who left the gas
brightly burning and the door temptingly ajar at 10.15, so that the
long-suffering woman pounced upon them with just recrimination,
only to find her stored-up wrath directed against two night-gowned
figures bowed over their little white beds, as it were two Infant
Samuels. It is doubtful if a devotional exercise ever before or since
has roused such mingled feelings in the bosom of the chance
spectator.
It was they who beyond a shadow of doubt won the basket-ball
game for the freshmen—an unprecedented victory—by their
marvellous intuition of each other's intentions and their manner of
being everywhere at once and playing into each other's hands with
an uncanny certainty. This gave them position and weight among
their mates, which they duly appreciated. They were the recognized
jesters of the class, and their merry, homely faces were sure of
answering grins wherever they appeared.
When they returned sophomore year more alike than ever, with
happy plans for the best double room on the second floor, they were
met by quite another kind of grin: its owner, Mrs. Harrow, would
have perhaps described it as firm and pleasant—the Twins referred
to it bitterly as hypocritical and disgusting.
"No, Martha, no. It's no use to coax me—I can't have it. I cannot go
through another such year. If you wish to remain in the house, you
must separate. You can have No. 10 with Alberta Bunting, and Kate
can go in with Margaret—she says she is perfectly willing, rather
than give up the room, and Helen is not coming back till next year.
Now, I don't want to have to argue about it; I think you are better
apart."
No one ever accused Mrs. Harrow of tact. Her placid firmness was
almost the most exasperating thing about her. Her decisions, if
apparently somewhat feather-beddish, ranked, nevertheless, with
those of the Medes and Persians, and the Twins walked haughtily
away—beaten but defiant.
Of course it never occurred to them to leave the house, and Kate,
after a time, grew quite contented, for Miss Pattison was eminently
pleasant and tactful, kept the room in beautiful order, and spent a
great deal of time in the Dewey with her sister, an instructor in the
college, and her great friend Cornelia Burt, who was off the campus.
This left the room to the Twins, who were almost as much together
as of yore. But Martha was in quite another case. In her the insult of
a dictated separation rankled continually, and her hitherto mild
contempt for Mrs. Harrow deepened into a positively appalling
enmity. Circumstances unfortunately assisted her feeling, for beyond
a doubt Alberta May Bunting was not adapted to her new room-
mate.
She was a wholesome, kindly creature, with high principles and no
particular waist-line. She drank a great deal of milk, and was a
source of great relief to her teachers, her recitations being practically
perfect. From her sophomore year she had been wildly, if solidly,
addicted to zoölogy, and to her, after hours spent in the successful
chase of the doomed insect, the grasshopper was literally a burden,
for she slew him by the basketful. She rendered the surrounding
territory frogless in her zeal for laboratory practice, and in her senior
year it was rumored that stray cats fled at her approach: "She'll cut
me up in my sleep," said Martha, gloomily, "and soak me in
formaline in the bath-tub—the idiot!"
For, although the "h'Arrow-that-flyeth-by-day-and-the-terror-that-
walketh-by-night," as Martha Williams, in a burst of inspiration, had
named her, could not, of course, have known it, Sutton M., as she
was most commonly called, loathed and despised bugs, reptiles, and
crawling and dismembered things generally, more than aught else
beside. She regarded an interest in such things as an indication of
mild insanity, and as a characteristic of Alberta May's such a
predilection assumed the proportions of a malignant insult.
"It's bad enough to have her drink milk like a cow, and eat graham
crackers like a—like a steam-engine," she confided to her
sympathetic sister, "and smell like a whole biological laboratory, and
glower at me, and bobble her head like a China image whenever I
open my mouth, and call me Mottha, which I despise, and say, 'Why,
the idea! Why, Mottha, the idea! What do you mean, Mottha?'
without putting little bottles of Things all around, and my having to
upset them. My gym suit made me sick to put on for a week
because I upset some nasty little claws all pickled in something per
cent. alcohol on the sleeve, and I kept thinking the legs were
walking on me—ugh! they were leggy claws!"
The h'Arrow-that-flyeth-by-day had fondly hoped that Alberta would
"do Martha Sutton a world of good," because of her exemplary,
regular habits and her calm, sensible nature, but this consummation,
though devoutly to be wished, was fated never to be witnessed.
Everyone heard the wails and gibes of Sutton M., but to few or none
were the woes of Alberta May made known. But that she must have
had them, her attitude at the time of the crisis conclusively proved.
The Twins, in the course of their mysterious loitering, overheard a
somewhat sentimental discussion between Evelyn Lyon and an
extremely stiff and correct young man from Amherst, as to whether
chivalry and openly expressed devotion to the fair were not
disappearing from the earth. "Men like shirt-waists and golf-shoes,"
Evelyn had been heard to murmur, with a glance at her fluffy chiffon
and bronze slippers, and the senior had protested that they did not,
and that emotion, if controlled, was as deep as in the balcony-
serenade days. "In fact," said he, finally, "Estabrook and I will
serenade you Wednesday night."
"You would never dare," said Evelyn, with a glance at his eye-glasses
and collar, which for height and circumference might have been a
cuff. "You'd be afraid the girls would laugh." The senior looked
nettled. "Expect us at ten on Wednesday next," said he. "It won't
necessarily be the Glee and Banjo Club, you understand, but it will
be a real, old-fashioned serenade." Then, as Evelyn smiled
maliciously, he added, "Only you must appear at the casement, and
throw flowers, you know—that's what they did." Evelyn frowned, but
agreed. "At the end of the song, I will," she said, with visions of the
night-watchman hasting to the scene.
The Twins were unaccountably strolling about as the senior left the
house, and wondered with great distinctness and repetition why on
earth Evelyn should say she'd be in 14 at the front when of course
she'd be in the East corner on the first floor. "She has some game
up," shrieked Martha, and Kate called back, "Of course she has—
some one will be awfully left, that's all!"
The senior listened, grinned, muttered that women told everything
they knew, and went his way. On next Wednesday night, the entire
house being congregated in the hall near No. 14, where Evelyn, not
to be found wanting in case they should get through a verse, was
sorting carnations, a husky burst of song enlivened the East corner,
a mandolin and a guitar having raced through a confused prelude
under the spur of a youth hopping with nervousness and sputtering
as he punched the mandolin-player: "Hang it all, Pete, get along, get
along! He'll be here in a minute—whoop it up, can't you?"
A muffled baritone began, standing so close to the window with a
light in it that its owner could have touched the sill with his
shoulder:

Last night the nightingale waked me,

Last night when all was—

The shade went up, the window followed, and the eyes of the
musicians beheld, below an audience of house-maids, the only
people at present on that side of the house, an enormous woman,
with gray hair in curling-kids, and a blanket-wrapper which added to
her size, grasping a lamp in her hand and regarding them with a
mingling of amazement, irritation, and authority that caused their
blood to curdle and their voices to cease. Pattering feet, a lantern
turned on them, and a voice: "'Ere, 'ere, what you doing? H'all h'off
the campus after ten—get along, now!" completed their confusion,
and they left, with an attempt at dignity and a slowness which they
had occasion to curse; for as they passed the front of the house,
from out of the air above their heads, apparently, two sweet and
boyish voices, a first and second soprano, lifted up to the fresh
October sky an ancient and beautiful hymn:

Sometimes a light surprises

The Christian while he sings,
It is—

A window banged forcibly, and the minstrels stood upon no order

but fled to their carriage and rattled out of town.
Evelyn Lyon, with set teeth and artistically loosened hair, rushed
down the hall behind Martha Sutton, who made the room she was
aiming for, slammed the door, realized that the key was lost, and
dragged the first piece of furniture that came to hand against it. This
was Alberta May's desk, and upon it were the collected results of her
vacation work at Wood's Holl. Six jars upset under the impact of
Evelyn's weight, a dozen mounted cross-sections jingled in the dark,
a pint bottle of ink soaked a thick and beautifully illustrated note-
book; and as the Terror-that-walketh-by-night headed Evelyn to her
door and mounted a flight to quell the rising tumult, Sutton M., with
a hysterical sob, for she was tingling with a delicious excitement,
huddled the desk back into the corner, hoped none of the bugs were
around the floor, and dropped into bed, wondering how ever Alberta
May could sleep through such a night.
And now—though perhaps you may have imagined that there was
never going to be any story—now we are coming to it, and though it
is short, all the characters appear. Alberta May, with an ugly brick-
red flush, told Sutton M. that she need never speak to her again, for
no answer would be forthcoming, and that she must have her things
out of the room before night. Martha was really horribly frightened,
and begged to be allowed to copy the note-book and hire some one
to make the slides and re-pickle the scattered Things; but Alberta
May merely shook her head, replied that she accepted apologies but
could not speak again, and kept her word, for she never noticed
Martha from then till the 22d of June.
The h'Arrow-that-flyeth-by-day gave Martha an address that reduced
her to a pulp, and having sent the Twins off to cry in each other's
arms till dinner-time and got the doctor for Evelyn, who had
sprained her ankle in the rush, she sat down to a cup of tea and
council.
To her entered Biscuits, and they talked of odds and ends till Mrs.
Harrow had grown a little calm. The girls in the house accused
Biscuits of a hypocritical and unnatural interest in the h'Arrow:
Biscuits denied this, alleging that she was merely ordinarily
courteous and saw no occasion for treating her like a dog, which
somewhat strong language was addressed with intention to a few of
her friends who certainly did not display any undue consideration in
their manner to the lady in question. She was wont to add calmly
that she saw no sense in having those in authority hate you when a
little politeness would so easily prevent it. And many times had she
successfully interceded for the offender and gained seats for guests
and obtained the parlor for dancing purposes on nights not
mentioned in the bond. On these accounts she made an unusually
fine house president in her senior year, and though as a sophomore
she had been but suspiciously regarded by that officer, she made as
firm a bond as is perhaps possible between powers so hostile as
those with which she struggled.
To-day she listened sympathetically as Mrs. Harrow held forth,
concluding with,—
"Now, Bertha, something must be done. I hate dreadfully to make a
change, so early in the year, too, but Alberta is decided, and says
that she will leave the house to-morrow unless Martha leaves to-
night. And Alberta is perfectly justified: nobody could be expected to
put up with it. I don't know whom to put her with: she certainly
can't be trusted with her friends, and I can't feel that I have any
right to put her anywhere else. I hate to have to admit that I can't
manage them—Miss Roberts insists that they're fine girls and will
outgrow it all, and I have great respect for her opinion, and yet—
think of that disgraceful performance last night! It would have done
credit to a boarding-school! I was so disgusted—"
"Yes, indeed, and I've talked to them, Mrs. Harrow, and told them
just how the house feels about it, but don't you think that it was
rather boarding-schoolish in Evelyn? She started it all, you know."
"Oh, well, of course. Evelyn shouldn't have—but then she is a good,
quiet girl, and—Oh, not that I would excuse her!"
"Certainly not," said Biscuits, briskly. This was good management on
her part, for Evelyn had one friend in the house to the Twins' ten,
though a favorite with Mrs. Harrow.
"Now, Mrs. Harrow, I've got an idea, and truly, I think it would
work," she added persuasively. When she had unfolded the idea, the
lady-in-charge could hardly believe her ears.
"Why, Bertha Kitts, you must be crazy! Nothing could induce me to
think of it for a moment—nothing! It would be the worst possible
influence!"
Biscuits argued gently. Her three years of consistent good sense and
politeness stood in her favor, and though Mrs. Harrow had no sense
of humor whatever, she was enabled to perceive a certain poetic
justice in the plan set before her.
"You know, Mrs. Harrow," she concluded, "that at bottom they're
both nice girls! They're awfully irritating at times, and of course you
feel that they've both occasioned a great deal of trouble; but they're
both honorable, and I'm sure it will be all right: truly, I'd be willing to
take the responsibility—if I can get them to consent to it!"
"Very well," said Mrs. Harrow, unwillingly, "you know them both
better than I do, Bertha, of course, and it certainly couldn't be any
worse than it is! But at the first outbreak I shall take the matter into
my own hands, and act very severely, if necessary!"
Biscuits went directly upstairs and sought out Martha Williams, who
lounged on the couch with Loti in her hand and a bag of chocolate
peppermints in her lap. Her room-mate, observing that Biscuits
glanced at the clock as she entered, murmured something about
getting a History note-book and obligingly disappeared.
"That's a good harmless creature," observed Biscuits, approvingly.
"Yes, she's in very good training," the creature's room-mate
returned. "Have a peppermint?"
"Pity she can't room with Alberta May," said Biscuits, lightly; "she'd
give her no trouble!"
"Lord, no!" Martha agreed; "she wouldn't trouble a fly!"
Biscuits wandered about the room and absent-mindedly picked up a
sheaf of papers.
"Themes back?" she inquired. Martha nodded.
"'Me see 'em?" Martha shrugged her shoulders in a manner to be
envied of the Continent.
Biscuits opened at a poem that caught her eye, and read it. Martha's
eyes were apparently fixed on Madame Chrysanthème, but they
wandered occasionally to Biscuits' face as she read. The poem was
called,—
 THE LIFTING VEIL

Do you love me now?

Ah, your mouth is cold!
Yet you taught me how—
Are we growing old?

Did you love me then?

Ah, your eyes are wet!
If the memory's sweet,
Why will you forget?

Could you love me still?

Hush! you shall not say!
Love is not of will—
Shall I go away?

Dare you love me now?

Let me burn my ships!
I, myself, am not so sure—
Am I worth your lips?

"Um—ah—yes," said Biscuits, "sounds something like Browning,

doesn't it?"
Martha looked only politely interested.
"Do you think so?" she said impersonally.
"Yes. I like that line about the ships," added Biscuits, tentatively; "it
—er—seems to—er—imply so much!"
Martha looked enigmatically at the skull. "Does it?" she asked.
Biscuits caught a glimpse of a long, hastily written story, and
gasped.
"Why, Martha, did you really hand that in?" she demanded.
"Certainly I did," said Martha; "why not?"
"Because it's really shocking, you know," Biscuits replied. "What did
she say?"
Martha hesitated, but a twinkle slipped into her eye and she smiled
as she replied. "Look and see," she said.
Biscuits turned to the last page, passing many an underlined word or
phrase by the way, and read in crimson ink at the bottom: Mallock
has done this better: you are getting very careless in your use of
relatives. At which Biscuits smiled wisely and reassured herself of an
announcement she had made in the middle of her junior year to the
effect that even among the Faculty one ran across occasional
evidences of real intelligence.
"Martha," she said abruptly, "I meant what I said about Mary and
Alberta—they'd make a very good pair."
"And Miss Sutton and I—" returned Martha, sardonically.
"Precisely," said Biscuits, "Miss Sutton and you. Oh, I know nobody
has the slightest right to ask it of you and we all supposed you
wouldn't, but at the same time I thought I'd just lay it before you. I
firmly believe, Martha, that you are the only person in this house
capable of managing Martha Sutton!"
"I?" And Madame Chrysanthème dropped to the floor.
"Yes, you. Now, Martha, just look at it: you know that the girl is a
perfect child—you know that she means well enough, and in her way
she has a keen sense of humor. Now you are much more mature
than the average girl up here and you take—er—broader views of
things than most of them. You wouldn't be so shocked at the things
Suttie does; you could, very gradually, you know, convey to her that
her ideas of humor were just a little crude, you know, and that
Welcome to our website – the ideal destination for book lovers and
knowledge seekers. With a mission to inspire endlessly, we offer a
vast collection of books, ranging from classic literary works to
specialized publications, self-development books, and children's
literature. Each book is a new journey of discovery, expanding
knowledge and enriching the soul of the reade

Our website is not just a platform for buying books, but a bridge
connecting readers to the timeless values of culture and wisdom. With
an elegant, user-friendly interface and an intelligent search system,
we are committed to providing a quick and convenient shopping
experience. Additionally, our special promotions and home delivery
services ensure that you save time and fully enjoy the joy of reading.

Let us accompany you on the journey of exploring knowledge and

personal growth!

ebookname.com