(FREE PDF Sample) Wireless Sensor Networks 1st Edition Liam I. Farrugia Ebooks

Full download ebook at ebookgate.
com
Wireless Sensor Networks 1st Edition Liam I.

Farrugia
https://ebookgate.com/product/wireless-sensor-
networks-1st-edition-liam-i-farrugia/
Download more ebook from https://ebookgate.com

More products digital (pdf, epub, mobi) instant
download maybe you interests ...
Protocols and architectures for wireless sensor

networks 1st Edition Holger Karl
https://ebookgate.com/product/protocols-and-architectures-for-
wireless-sensor-networks-1st-edition-holger-karl/
Cooperating Embedded Systems and Wireless Sensor

Networks 1st Edition Michel Banatre
https://ebookgate.com/product/cooperating-embedded-systems-and-
wireless-sensor-networks-1st-edition-michel-banatre/
Security for Wireless Sensor Networks using Identity

Based Cryptography Patil
https://ebookgate.com/product/security-for-wireless-sensor-
networks-using-identity-based-cryptography-patil/
Mobile Wireless and Sensor Networks Technology

Applications and Future Directions 1st Edition Rajeev
Shorey
https://ebookgate.com/product/mobile-wireless-and-sensor-
networks-technology-applications-and-future-directions-1st-
edition-rajeev-shorey/
RFID and Sensor Networks Architectures Protocols
Security and Integrations Wireless Networks and Mobile
Communications 1st Edition Yan Zhang
https://ebookgate.com/product/rfid-and-sensor-networks-
architectures-protocols-security-and-integrations-wireless-
networks-and-mobile-communications-1st-edition-yan-zhang/
Synchronization in Wireless Sensor Networks Parameter

Estimation Peformance Benchmarks and Protocols 1st
Edition Erchin Serpedin
https://ebookgate.com/product/synchronization-in-wireless-sensor-
networks-parameter-estimation-peformance-benchmarks-and-
protocols-1st-edition-erchin-serpedin/
Co Operative and Energy Efficient Body Area and

Wireless Sensor Networks for Healthcare Applications
1st Edition Akram Alomainy
https://ebookgate.com/product/co-operative-and-energy-efficient-
body-area-and-wireless-sensor-networks-for-healthcare-
applications-1st-edition-akram-alomainy/
Deploying wireless networks Wilton A.
https://ebookgate.com/product/deploying-wireless-networks-
wilton-a/
Technological breakthroughs in modern wireless sensor

applications 1st Edition Hamid Sharif
https://ebookgate.com/product/technological-breakthroughs-in-
modern-wireless-sensor-applications-1st-edition-hamid-sharif/
Copyright © 2010. Nova Science Publishers, Incorporated. All rights reserved.
Wireless Sensor Networks, Nova Science Publishers, Incorporated, 2010. ProQuest Ebook Central,
COMPUTER SCIENCE, TECHNOLOGY AND APPLICATIONS
WIRELESS SENSOR NETWORKS

No part of this digital document may be reproduced, stored in a retrieval system or transmitted in any form or
by any means. The publisher has taken reasonable care in the preparation of this digital document, but makes no
expressed or implied warranty of any kind and assumes no responsibility for any errors or omissions. No
liability is assumed for incidental or consequential damages in connection with or arising out of information
contained herein. This digital document is sold with the clear understanding that the publisher is not engaged in
rendering legal, medical or any other professional services.
COMPUTER SCIENCE, TECHNOLOGY
AND APPLICATIONS
Additional books in this series can be found on Nova’s website

under the Series tab.
Additional E-books in this series can be found on Nova’s website

under the E-book tab.
COMPUTER SCIENCE, TECHNOLOGY AND APPLICATIONS
WIRELESS SENSOR NETWORKS
LIAM I. FARRUGIA
EDITOR
Nova Science Publishers, Inc.

New York
Copyright © 2011 by Nova Science Publishers, Inc.
All rights reserved. No part of this book may be reproduced, stored in a retrieval system or transmitted in
any form or by any means: electronic, electrostatic, magnetic, tape, mechanical photocopying, recording or
otherwise without the written permission of the Publisher.
For permission to use material from this book please contact us:
Telephone 631-231-7269; Fax 631-231-8175
Web Site: http://www.novapublishers.com
NOTICE TO THE READER

The Publisher has taken reasonable care in the preparation of this book, but makes no expressed or implied
warranty of any kind and assumes no responsibility for any errors or omissions. No liability is assumed for
incidental or consequential damages in connection with or arising out of information contained in this book.
The Publisher shall not be liable for any special, consequential, or exemplary damages resulting, in whole or
in part, from the readers’ use of, or reliance upon, this material. Any parts of this book based on government
reports are so indicated and copyright is claimed for those parts to the extent applicable to compilations of
such works.
Independent verification should be sought for any data, advice or recommendations contained in this book. In
addition, no responsibility is assumed by the publisher for any injury and/or damage to persons or property
arising from any methods, products, instructions, ideas or otherwise contained in this publication.
This publication is designed to provide accurate and authoritative information with regard to the subject
matter covered herein. It is sold with the clear understanding that the Publisher is not engaged in rendering
legal or any other professional services. If legal or any other expert assistance is required, the services of a
competent person should be sought. FROM A DECLARATION OF PARTICIPANTS JOINTLY ADOPTED
BY A COMMITTEE OF THE AMERICAN BAR ASSOCIATION AND A COMMITTEE OF
PUBLISHERS.
Additional color graphics may be available in the e-book version of this book.
LIBRARY OF CONGRESS CATALOGING-IN-PUBLICATION DATA
Wireless sensor networks / editors, Liam I. Farrugia.

p. cm.
Includes index.
ISBN: (eBook)
1. Wireless sensor networks. I. Farrugia, Liam I.
TK7872.D48W573 2010
681'.2--dc22
2010014106
Published by Nova Science Publishers, Inc. † New York
CONTENTS
Preface vii
Chapter 1 Improvement in Lifetime of Wireless Sensor Networks Using
Perfect Difference Set-Networks 1
G. Sudhir Akojwar and M. Rajendra Patrikar
Chapter 2 Key Management Protocol for Heterogeneous Sensor Networks 27
Ma Chun-Guang, Wang Jiu-Ru, Geng Gui-Ning, Shang Zhi-Guo
and Zhang Bing-Zheng
Chapter 3 Data Aggregation in WSNs: State of the Art and a Multi-Criteria
Proposal 43
Ahmad Sardouk, Leïla Merghem-Boulahia, Rana Rahim-Amoud
and Dominique Gaïti
Chapter 4 Detection of Malicious Code Injection Attacks in Sensor Networks
Using Sequential Analysis 89
Jun-Won Ho
Chapter 5 The Simulation on Wireless Sensor Networks 95
Bai Li, Longxiang Gao and Binbin Zhou
Chapter 6 Continuous Data Collection In Wireless Sensor Networks 113
Dan Wang, Qian Zhang and Jiangchuan Liu
Chapter 7 On Load-Balanced Data Gathering for Lifetime Maximization in
Wireless Sensor Networks 133
Punyasha Chatterjee and Nabanita Das
Commentary Future Direction of Network Operators: Wireless Machine-to-
Machine Communications Network Infrastructure 157
Hiroshi Saito
Index 163
PREFACE
A wireless sensor network (WSN) consists of spatially distributed autonomous sensors to

cooperatively monitor physical or environmental conditions, such as temperature, sound,
vibration, pressure, motion or pollutants. The development of wireless sensor networks was
motivated by military applications such as battlefield surveillance. They are now used in
many industrial and civilian application areas, including industrial process monitoring and
control, machine health monitoring, environment and habitat monitoring, healthcare
applications, home automation, and traffic control. This new book gathers and presents
topical research data from around the globe in the field of wireless sensor networks.
Chapter 1 - Wireless Sensor Networks (WSN) are becoming more popular data gathering
networks for monitoring a wide range of parameters. WSN are being used for monitoring
various parameters such as environmental, industrial, in-house domestic parameters, medical
and health parameters etc. WSN populates a huge number of sensor nodes. A typical
architecture of sensor node is shown in Figure 1. Sensor nodes are tiny devices, which consist
of three functional units: Sensing Unit, Computational Unit and Communication Unit.
Sensing unit is composed of an array of sensors, which may include temperature, light, sound,
humidity, seismic, and position sensors. Computational unit is usually a microcontroller

which processes the sensor data. Communication unit is a radio device used for
communicating the data to the neighboring sensor node. WSN populates a huge number of
sensor nodes. These nodes form a self-organized network. These nodes coordinate and
communicate to perform a common task of data gathering. For this purpose, different routing
algorithms are used.
Chapter 2 - In this chapter, the authors will talk about Heterogeneous Sensor Networks
key Management Protocol.
At the first of this part, the authors will talk about what is Heterogeneous Sensor
Networks (HSN), and how to divide key management protocol by different method.
The second part of this chapter, based on different network threats and assumptions, the
authors will introduce one key pre-distributed protocol, two key management protocols.
About the key pre-distributed protocol, the key ring of nodes is based on divisible core pair-
wise balanced design. The authors design two key pre-distributed protocol, Divisible Core
Pair-wise Balanced Design Scheme shorted by DCPBD, and variant DCPBD shorted by
VDCPBD. One of the key management protocols, the authors propose a key management
protocol for heterogeneous sensor networks to improve the random key pre-distribution
scheme using deployment knowledge of nodes and the prior area deployment information.
viii Liam I. Farrugia
Another key management protocol, the authors propose a Location-aware and secret share
based dynamic key management protocol to effectively replace the compromised central node
and enhance the security level of the network.
At the last of this chapter, the authors will summarize this chapter, and predict the future
development of key management.
Chapter 3 – The basic role of Wireless Sensor Network (WSN) is to collect data from the
environment by many sensor nodes. The WSN could be constituted by hundreds or thousands
of sensor nodes; hence, it could generate a huge amount of data. The communication of all
the sensed data causes high power consumption limiting, as a result, the lifetime of the sensor
nodes and consequently the WSN lifetime. Whereas, processing data locally within the sensor
node consumes much less than its communication. Thus, it is possible to reduce the power
consumption of the sensor nodes and, hence, to extend the network lifetime, by reducing the
number and size of the communicated data. This reduction of communicated data is possible,
if the sensor nodes' data are aggregated instead of being sent directly to the sink (the base
station in WSNs).
This chapter studies several methods of data aggregation in WSNs. It classifies them into
three categories. The first category represents structure-free mechanisms which abstract the
organization of the nodes within the network. It allows the end-user of the network to query
each sensor node, if needed. In the second category, the authors find the structure-based
mechanisms that solve the structure-free problems in terms of scalability. It organizes the
sensor nodes into a well structured form, such as a tree, or a set of groups. It defines
aggregator nodes to aggregate other nodes data. This aggregator node may be one head-node
per group or one parent for several children in a tree. Then, the end-user of the network has
the aggregated data through the aggregator nodes. The third one is a Multi-criteria Data
aggregation (McDa) mechanism. It combines characteristics from the structure-free and the
structure-based categories. It does not suppose any network structure. However,
spontaneously during the same data aggregation session, it is possible to have cooperation
between single nodes and groups of nodes. It eliminates the overhead and signaling needed in
structure-based category to create and maintain an organization of the sensor nodes.

Finally, a discussion of the three methods of data aggregation is made in order to
underline, according to their characteristics, their advantages and limitations. In addition, the
chapter studies some methods to secure the aggregated data by treating confidentiality and
integrity issues.
Chapter 4 – In wireless sensor networks, an attacker can physically capture and
compromise sensor nodes because they are deployed in an unattended manner. However, it
will take substantial time and effort for the attacker to widely spread node compromise by
only physically capturing sensor nodes. To facilitate wide-spread node compromise, an
attacker may launch \textit{malicious code injection attacks} in which he forces a few
compromised nodes to inject malicious codes into a set of benign nodes, leading to the
compromise of those benign nodes. To detect malicious code injection attacks, the authors
propose a malicious code detection scheme by using the Sequential Probability Ratio Test in
wireless sensor networks. The authors analytically show that the proposed scheme achieves
effective and efficient malicious code detection capability.
Chapter 5 – Wireless sensor networks (WSNs) have attracted significant attention in the
past few years and have been widely implemented in battlefield communication,
environmental sensing, and traffic monitoring. They consist of large numbers of cooperating
Preface ix
small-scale motes, each only capable of limited processing performance, short-range radio
transmission, and low power sensing capacities. As such, there has been an increasing need to
define and develop simulation frameworks for carrying out highfidelity WSN simulation,
which should be easily chosen and employed by researchers as well. This book chapter
surveys the recent simulation work on WSN and presents the available up-to-date simulators.
Each simulation tool is described based on some certain criteria and compared under the
appropriate evaluation standards in terms of different task requirements. This also provides
the readers with a better understanding of the latest development of WSN simulation and
helps them to incorporate their own WSN research work into the suitable simulation
development environment in the future.
Chapter 6 - Wireless sensor networks have been widely used for surveillance in harsh
environ-ments. In many such applications, the environmental data are continuously sensed,
and data collection by a server is only performed occasionally. Hence, the sensor nodes have
to temporarily store the data, and provide easy and on-hand access for most updated data
when the server approaches. Given the expensive server-to-sensor communications, the large
amount of sensors, and the limited storage space at each tiny sensor, continuous data
collection becomes a challenging problem.
In this paper, the authors present partial network coding (PNC) as a generic tool for the
above applications. PNC generalizes the existing network coding (NC) paradigm, an elegant
solution for ubiquitous data distribution and collection. Yet, PNC enables ef-fiand collectio
replacement for continuous data, which is a major deficiency of the conventional NC. The
authors prove that the performance of PNC is quite close to NC, ex-cept for a sub-linear
overhead on storage and communications. The authors then address a set of practical concerns
toward PNC-based continuous data collection and replacement in sensor networks. Its
feasibility and superiority are further demonstrated through simulation results.
Chapter 7 - A wireless sensor network (WSN) is an autonomous self-organizing system
of sensor nodes, distributed over an area to sense data from its surroundings and gather it to a
sink node for further processing. Sensor nodes are typically battery-operated and therefore
severely energy constrained. Besides limited computation and data sensing, each sensor node
dissipates most of its energy in transmitting and receiving data packets en route the sink node.
The shortest path routing is good for overall energy efficiency since energy needed to
transmit a packet is correlated to the path length. However this may load some nodes heavily,
causing earlier deaths and creating holes in the network, which in the worst case may leave
the network disconnected. Hence, load balanced data routing, considering load in terms of
power consumption at individual nodes, emerges as an important issue in WSN’s to enhance
the network lifetime. Unfortunately, the problem to compute the most balanced routes for
data gathering is NP-hard even in a very simple network. Extensive research has been done so
far to solve the problem under various scenarios of traffic patterns, data aggregation
techniques and node distribution policies. This chapter aims to present a comprehensive study
of several load-balancing schemes for wireless sensor networks, proposed in recent research
literatures, with special emphasis on the distributed ones that may lead the researchers to
explore newer avenues for load balanced data gathering in WSN’s to enhance network
lifetime.
Commentary - Due to the saturation of the telecommunication markets in technologically
advanced countries, network operators need to create a new market. One promising market is
for mobile wireless machine-to-machine (M2M) communications. Current networks do not
x Liam I. Farrugia
work well with many M2M applications; therefore, a new network is needed. To differentiate
this new network from others, it should have the following features: 1) low power
consumption, 2) be low-end-terminal friendly, 3) low cost, and 4) ubiquitous coverage.
Current networks do not satisfy these features. Standardization bodies and forums may also
be interested in the mobile wireless M2M communications market, but their main target at
present seems to be the current network infrastructure with short range radio equipment
purchased by customers. To implement the new network as a network infrastructure
satisfying the above-mentioned features, R&D efforts are essential for developing a large cell
with wireless terminals of low transmission power and a simplified security protocol
implementable in a low-end wireless terminal. A Wide Area Ubiquitous Network has been
proposed as one solution, and application experiments on this network have been conducted
in Tokyo, Japan.
In: Wireless Sensor Networks ISBN: 978-1-61728-125-9
Editor: Liam I. Farrugia © 2011 Nova Science Publishers, Inc.
Chapter 1
IMPROVEMENT IN LIFETIME OF WIRELESS SENSOR

NETWORKS USING PERFECT DIFFERENCE SET-
NETWORKS
G. Sudhir Akojwar*1 and M. Rajendra Patrikar**2

1
Department of Electronics Engineering, Rajiv Gandhi College of Engineering
Research and Technology, Chandrapur - 442402, India
2
Department of Electronics and Computer Science Engineering,
Vesvevsaraya National Institute of Technology, Nagpur- 441010, India
1. WIRELESS SENSOR NETWORKS

1.1. Review and Introduction
Wireless Sensor Networks (WSN) are becoming more popular data gathering networks
for monitoring a wide range of parameters. WSN are being used for monitoring various
parameters such as environmental, industrial, in-house domestic parameters, medical and
health parameters etc. WSN populates a huge number of sensor nodes. A typical architecture
of sensor node is shown in Figure 1. Sensor nodes are tiny devices, which consist of three
functional units: Sensing Unit, Computational Unit and Communication Unit. Sensing unit is
composed of an array of sensors, which may include temperature, light, sound, humidity,
seismic, and position sensors. Computational unit is usually a microcontroller which
processes the sensor data. Communication unit is a radio device used for communicating the
data to the neighboring sensor node. WSN populates a huge number of sensor nodes. These
nodes form a self-organized network. These nodes coordinate and communicate to perform a
common task of data gathering. For this purpose, different routing algorithms are used.
Wireless sensor networks have become more popular because advances in sensor
technology, low-power electronics, low-power radio frequency design, and VLSI & MEMS
* Corresponding author: [email protected], [email protected].

** [email protected].
2 G. Sudhir Akojwar and M. Rajendra Patrikar
technology, have enabled the development of small, relatively inexpensive and low-power
sensors, called microsensors. These wireless microsensor networks represent a new paradigm
for extracting data from the environment and enabling the reliable monitoring of a variety of
environments for applications that include surveillance, machine failure diagnosis,
chemical/biological detection, habitat monitoring, environmental monitoring etc. WSN are
effective ubiquitous computing systems used for medical and health care applications.
The sensor network is composed of a large number of sensor nodes which are densely
deployed either inside the phenomenon or in its proximity. The sensor nodes may be
randomly deployed in inaccessible terrain or disaster relief operations. Hence, sensor network
protocols and algorithms must possess self-organizing capabilities. Wireless sensor nodes are
battery powered and thus face energy constraint. Every node has limited power which decides
the life span of the node and, in turn, the entire network. Many times, WSN include redundant
nodes to provide fault tolerant capability. A few sensor nodes may die because of power
exhaustion or any other failure. Even if few nodes die, the sensor network continues to
function because of its fault-tolerant capability. One of the constraints on sensor nodes is the
low power consumption. Hence, sensor network protocols focus on power conservation.
Since the sensor nodes are often inaccessible, the lifetime of a sensor node must be assured.
Sensor node lifetime depends on the lifetime of power resources; hence, power scarcity must
be effectively managed. Figure 1 shows architecture of a typical sensor node. It basically
consists of: sensing unit, computation or data processing unit and communication or radio
unit. Daniele et al. [1] has discussed many hardware specifications/features of sensor
node/mote for radio, microcontroller and the platform used.
Mem
Sensors
Battery
DC - DC
RADIO
ADC
MC
Algorithms and
Real time
Figure 1. Architecture of a typical Sensor Node.
Typical sensor node block diagram is shown in Figure 1. It basically consists of: Sensing
unit (sensors), computation or data processing unit (MCU: micro controller unit) and
communication unit (RADIO). Sensor data is converted to digital streams using ADC.
Microcomputer units (MCU) process data streams by executing algorithms, such as
classification algorithm. Then the processed data is communicated over the network by radio
unit. The entire sensor node works under a typical real time operating system such as TINY
OS. The MCU also includes a protocol stack for efficiently routing the data over the sensor
network.
Improvement in Lifetime of Wireless Sensor Networks Using Perfect … 3
Power consumption can be divided into three domains: sensing, communication and data
processing. Sensing power varies with the nature of applications. Sporadic sensing might
consume less power than constant event monitoring. Of the three domains, a sensor node
expends maximum energy in data communication by radio unit. This involves both data
transmission and reception. Sensor node architecture, wireless sensor network protocols, its
applications, and wireless-networking issues are very well described in Akyildiz et al. [2].
Many categories of applications of WSN- such as industrial applications, domestic
applications, military, public utilities, agriculture etc. are discussed by Carlos et al. [3]. Lewis
[4] has described many topologies for WSN and has discussed network structures and
hierarchical networks. Chong et al. [5] has elaborated on the history of research on sensor
network research. He has discussed the prospects of sensor network research in 21st century.
The main functional blocks in sensor nodes that consume power are: sensing unit,
computation unit and communication unit. An important challenge in the design of these
networks is that two key resources - communication bandwidth and energy - are significantly
more limited than in a tethered network environment. These constraints require innovative
design techniques to use the available bandwidth and energy efficiently. The communication
consumes the major part of the energy budget, thus attempts must be made to implement
techniques to save energy on communications. Every aspect of WSN is designed with energy
constraints. There are three basic units in sensor nodes which consume power:
communication, computation and sensing unit. Hence, all the attempts are being made to
reduce the energy consumption at each unit. Many energy aware protocols are developed for
reduced energy consumption and thereby improve the life span of the WSN. Wireless sensor
network bears very specific characteristics that are described in following section.
1.2. Characteristics and Requirements of Wireless Sensor Networks

1. Low bit rates: Sensor networks have low bit rates (~1kbps). The low bit rate is due
to the fact that the sensor parameter varies at very low rates. This is a very important
feature of WSN which needs to be fully explored at each design step. This feature
makes it possible to keep the WSN in deep sleep mode.
2. Low duty cycle : The sensor nodes are kept in sleep mode for most of the time. The
duty cycle for sleep and wake up mode is typically around 1%. Thus, the node is kept
on only when it is sensing data with its sensor or transmitting/receiving data from
other nodes. The strategy is to keep the sensor node in sleep mode for most of the
time to reduce energy consumption.
3. Size constraints: In order to deploy the sensor nodes unobtrusively for environment
monitoring, the sensor nodes need to have miniature size. The current focus of the
research community is to achieve dust-sized sensor nodes. Advances in VLSI and
MEMS are leading to smaller sensor nodes. Further advances in nanotechnology and
organic sensors are likely to help in reduction in the size of the nodes. Currently, the
power supply (battery or energy scavenging technique) is leading to larger sensor
nodes. It is possible to employ renewable energy resources within the nodes.

Research in this area would definitely help in achieving smaller sensor nodes.
4. Low cost: Sensor nodes should be cheap. Since this network will have hundreds or
thousands of sensor nodes, these devices should be low cost. Even the research
community is expecting and striving for sensor nodes with negligible cost as
compared to network infrastructure cost.
5. Energy constraints: Energy constraint is the most important factor influencing the
design of the sensor nodes and the choice of algorithms for the sensor networks. Due
to energy constraints, sensor nodes need to have ultra-low power circuitry. The
algorithms for sensor networks need to be energy-efficient.
6. Data-centric: WSN are data-centric in nature, meaning that the emphasis is on data
to be routed. This can be contrasted with traditional wireless networks which are
essentially node-centric. WSN are basically data gathering networks.
7. Multi hop: The signal from a wireless sensor node may not reach the base station
straight. The solution is to communicate through multi-hop. Another advantage is
that radio signal power is proportional to r4, where r is the distance of
communication. Thus, depending on radio parameters, it can be more energy
economic to transmit many short-distance messages than one-long distance message.
8. Localization: The data from a sensor node is meaningless if it is not associated with
the position of the sensor node. Localization is important for wireless sensor
networks and the data from a node is sent along with the co-ordinates of the sensor
node.
9. Scalability: Wireless sensor networks consist of thousands of sensor nodes, while

traditional wireless networks contain a maximum of hundreds of nodes. As a result,
care needs to be taken in the design of algorithms for WSN. The sensor network
should scale from ten to thousands or millions of sensor nodes. This needs automatic-
configuration, maintenance, and upgrading of individual devices. Another important
attribute is scalability and adaptivity to changes in network size, node density and
topology. Some nodes may die over time; some new nodes may join later; some
nodes may move to different locations. Hence, WSN must be scalable and adaptive.
10. Distributed Network and Distributed Processing: WSN consists of large numbers
of nodes that are distributed over a region of interest. Each sensor node should be
able to process local data, using filtering, data fusion and aggregation algorithms to
collect data from the environment and aggregate this data, transforming it to
information.
11. Dynamic Networks: Sensor nodes are mobile so that during deployment they can be
placed as per needs of the application. Due to node mobility, environmental
obstructions, restricted resources etc., the sensor networks exhibit a highly dynamic
network topology.
12. Redundancy in data: There is inherent redundancy in the readings from the sensors.
This redundancy can be further characterized as spatial redundancy and temporal
redundancy. Spatial redundancy arises from the fact that sensor nodes, which are
physically placed close to each other, will give similar value of the measured
quantity. In addition, the value of measured physical quantity will not change
abruptly in short span of time, and, therefore, the data measured over a short time
span is likely to remain same. This gives rise to temporal redundancy. Hence, WSN
must have efficient data processing and management algorithms.
13. Application-specific: WSN are deployed in a wide-range of scenarios. Hence, they

would be application-specific in nature. For instance, a WSN deployed in a forest for
wildlife or habitat monitoring is likely to have different requirements than a WSN
deployed in indoor spaces for temperature monitoring.
From the literature review, we identified that the communication unit consumes major
part of the energy available in the sensor node, as discussed by Raghunathan et al. [6]. This
research work discusses classification technique using ART1 (Adaptive Resonance Theory)
and Fuzzy ART (FA) neural networks model. ART1/FA has a specific feature that allows it to
classify real time data, hence being called real time classifier. Real time classifier classifies
the sensor readings and then the classified sensor data is communicated further. This saves a
sufficient amount of energy.
The next sections describe the issue of fixed deployment of sensor nodes, introduce the
research approach, and briefly describe UbiSens: Project at VLSI Lab VNIT.
1.3. Fixed (Non Mobile) Deployment of Sensor Nodes
Deployment of sensor nodes includes many possibilities. Traditionally, sensor nodes are
randomly deployed over a geographical area. This includes routing protocols which use
random deployment principle. Random deployment needs a location finding system in the
node, which could find the neighboring nodes and decide whether to connect and transmit
data. In such protocols, a large part of the energy is consumed in executing long programs by
the computation unit. We pondered over the idea of reducing energy consumption on
executing routing protocols. Today, WSN are also becoming popular in applications such as
smart houses, intelligent offices, mines, nuclear installations, large buildings, government
installations, etc. Hence, for such applications, fixed geometrical deployment could be
employed. Sensor nodes while deployment adheres to some geometrical form. We chose to
implement the concept of Perfect difference set networks for wireless sensor nodes. In PDS
networks, any node can communicate with other node within a hop count of one or two. This
makes it possible to use simplest possible routing technique. This routing technique can be
effectively implemented in VLSI while designing the node. This routing technique reduces
the computation time of the processor and hence in turn saves power consumed by the
computation unit.
1.4. Research Approach
We propose to use ANN (artificial neural network) based ART1 neural network model
for classification of sensor data. By classification, huge sensor data traffic can be reduced and
hence it lowers the energy consumption on the communication unit and also improves the
bandwidth of the network. This classifier is supposed to be a part of the (ubiquitous sensing)
UbiSens: Low power sensor node. When this node is deployed in WSN, the life span of the
network is improved by about 40%. Many classification and clustering algorithms are
available and are being developed for data management in sensor networks.
Most of the energy-aware routing algorithms are designed for the random deployment of
sensor networks, which are suitable for large geographical deployment. Fixed geometrical
deployment strategy can be more efficient and economical for deploying sensor network in
smart houses, offices, hospitals, mines, nuclear installations, and many other domestic
applications.
We propose to use Perfect Difference Set (PDS) network for the fixed geometrical
deployment of sensor networks. In this research work, we propose to reduce the energy
consumption by computation unit by using PDS-Networks for deploying WSN. By using
classification techniques, a large amount of sensor data traffic can be reduced and, thereby,
energy spent on communication unit is saved. The lifetime of WSN with popular cooperative
routing is compared with PDS-WSN. The ART1 neural network model is used for the
classification of sensor data. The lifetime of WSN with and without classification in both the
cases – with PDS-WSN and with cooperative routing is also presented.
In this research work we attempt to reduce the power consumption on communication
unit by using classification techniques and by deploying the sensor nodes as per PDS –
network power consumption by computation unit can be reduced. PDS-WSN takes care of
geometrical deployment of the nodes.
Many research efforts are being carried out for designing low power sensor nodes. One
such sensor node architecture – UBISENS – is proposed and undertaken at VLSI Lab, VNIT,
Nagpur. The proposed classifier in this work is to be used in this sensor node. With
classification technique, it is proposed that a Low power sensor node with an improved
lifetime can be designed.
The low power sensor node was achieved using the classification technique. Then we
thought of deploying the sensor node with effective and least power consumption. Why we
should not have a network strategy where we have minimum hops and simpler routing? We
ponder over the concept and we used the concept of perfect difference set networks to be used
for WSN.
PDS-WSN strategy is explored and experimented for its fixed (non mobile) geometrical
deployment. PDS-WSN facilitates minimum hop counts (one or two) to reach cluster head.
This feature provides low power consumption by the network and simplest routing technique,
which leads to improvement in the lifetime of WSN. Thus, the low power sensor network can
be realized. Hence, it can be called a low power network. A further advantage of PDS-WSN
is that the routing technique is simple and can be realized as VLSI implementation. We had
VLSI framework in mind while proceeding on this research work. ART1/ Fuzzy ART
classifier algorithms are proposed which provides features for its easy VLSI implementation.
This ART1 classifier can be effectively used in noisy industrial environments by using DWT
input layer filter. Many possible VLSI implementations of DWT are available. Thus, the
research output provides a framework for VLSI implementation of Low power Sensor node
with Low power Sensor network.
We have implemented cooperative routing techniques for data dissemination in the
network, which give the advantage of equitable distribution of energy over the network and
thereby improving the average lifespan of the network. We have deployed two separate
radios; a low bit rate (1Kbps) MAC Radio with 100% duty cycle and a high bit rate (10Kbps)
data transmission Radio with low duty cycle (1%). The MAC radio is always ON for
maintaining the network while the data transmission radio is under sleep or wake up mode. It
wakes up only during data transmission. This strategy also contributes to the saving of energy
consumption by the communication unit. This technique is suggested in Berkeley Motes and
discussed by Miller et al. [7].
The lifetime comparison for cooperative routing and PDS-WSN with and without
classification techniques is presented in this chapter. The WSN is modeled [8] in Ptolemy-II:
Visualsense [9,10]. The ART1 neural network model is implemented in MATLAB. ART1
classifier implemented in MATLAB is interfaced with WSN model implemented in Ptolemy-
II.
1.5. UbiSens: Project at VLSI Lab VNIT Nagpur
The UbiSens (Ubiquitous Sensing) project [11] at VNIT is targeted towards exploring
novel architecture for ultra-low power sensor nodes. Typically, a sensor node consists of a
microcontroller, a power supply, sensor, ADC, memory and RF transceiver to achieve

wireless connectivity between different nodes. The design of sensor node circuitry is driven
by the need for ultra-low power operation. We propose the strategies to use in the UbiSens
project to achieve low power operation. Our approach is to partition the system in such a
manner that each sub-system part contributes towards the overall reduction of power
consumption. In addition, sensor networks can be used in numerous applications and hence,
they are application-specific. Thus, partitioning the system into several subsystems can make
it possible to tailor certain subsystems to suit the application for which the sensor node is
intended. A model in which, instead of having a single microcontroller, there are a number of
subsystems that are meant for a specific purpose, is proposed.
These subsystems are referred to as “processing elements”. This division of the
computation between parallel processing elements is beneficial since each processing element
may be tailored for a particular task and specific techniques could be used for reducing power
consumption. Further ambient RF energy scavenging to ensure long lifetime of the sensor
nodes is used. The proposed architecture for sensor nodes is shown in Figure 2. The various
processing elements in the proposed architecture are:
1. Central controller: This performs generic computation and is responsible for

controlling the operation of the entire sensor node. This controller has low speed of
operation (~100-200 KHz) and is power efficient. It directs the operation of the other
processing elements.
2. Data compression engine: This processing element is responsible for compressing the
measured data by the sensor. This compression of data leads to significant reduction
in the transmit energy spent and hence, leads to low power operation.
3. Networking module: This module controls the networking operation of each sensor
node. The DLL and Network layer operations are performed by this processing
element.
4. Power controller: This controller efficiently regulates the power being supplied to the
sensor node circuitry. Effective techniques such as dynamic voltage scaling may be
employed to reduce the power consumption of the sensor node.
In addition to the above mentioned strategies, various software techniques and efficient
scheduling can help to reduce the power consumption. The classifier developed under this
research work is supposed to be embedded with the sensor node as shown in Figure 3.
The classifier functional block may be VLSI implemented hardware. If the node
processor and memory permits, the classifier may be a software module. However, to gain
speed, reduce memory requirement and save power, we could have VLSI implementation of
classifier. The sensor data is classified at classifier and then the data is communicated through
transceiver to the neighbouring sensor nodes. Under this architecture, the data compression
engine becomes optional.
Control and Processing Unit

Data
Power Controller Central Controller Compression

Engine
Networking Module
Sensors ADC MEMORY TRANSCEIVER
Power Storage
Power Source
Figure 2. UbiSens: Wireless Sensor Node architecture.
Control and ANN Based Classifier

Processing Unit Optional
Central Controller Data
Power Controller
Compression
Engine
Networking Module
Sensors ADC MEMORY TRANSCEIVER
Power Source Power Storage
Figure 3.Wireless Sensor Node Architecture with ANN Based classifier.
2. PERFECT DIFFERENCE SET NETWORKS

FOR WIRELESS SENSOR NETWORKS
The role of Projective Geometry in WSN is discussed in this chapter. The mathematical
notion of Perfect Difference set is derived from Projective Geometry which simplifies node
deployment and routing technique. The concept of Perfect Difference set-Network for
Wireless Sensor Networks is introduced in this chapter. A cluster of 12 sensor nodes is
formed and sensor network with 4 such clusters is implemented. A similar sensor network
with two tires of PDS-WSN with 6 clusters is also implemented. Lifetime computation is
carried out for both cases with and without classification.
2.1. Projective Geometry in WSN
The parallel computer architecture needs efficient interconnection techniques. Such

parallel architectures involve sparse matrix and their efficient computations. Amrutur et al.
[12] had discussed the projective geometry architecture for scientific computation. In this
paper, Amrutur et al. had presented efficient computation of sparse matrix and
interconnection schemes. This paper discusses perfect patterns and perfect sequences for
interconnection networks. The concept of interconnection network can be used for sensor
networks.
A wireless ad-hoc network can be roughly considered as one consisting of a collection of
mobile nodes, which adopts a non-center, self-organized structure. Pingyi et al. [13] had
described geometrical characteristics of wireless Ad-Hoc networks. In contrast to other
networks, all the nodes in a wireless ad-hoc network can transmit, receive and forward
messages. Thus, it does not require supports of backbone networks. These characteristics
make it superior to those schemes requiring infrastructure supports and are especially useful
in battlefield, disaster relief, and scientific exploration. Unfortunately, since the locations of
nodes are random and time varying, it is more difficult to analyze the performance of ad-hoc
networks. In general, wireless ad-hoc networks can be modeled in the framework of a random
graph, where network nodes and links can be thought of as vertices and edges of a random
graph.
A Distributed Geometric Routing Algorithm for Ad Hoc Wireless Networks is presented
by Rashid et al [14]. This paper presents a fully distributed algorithm to compute a planar
subgraph of the underlying wireless connectivity graph. It considers the idealized unit disk
graph model in which nodes are assumed to be connected if and only if nodes are within their
transmission range. The main contribution of this paper is a fully distributed algorithm to
extract the connected, planar graph for routing in the wireless networks. The communication
cost of the proposed algorithm is O(d log d) bits, where d is the degree of a node. This paper
presents a geometric routing algorithm. The algorithm is fully distributed and nodes know
only the position of other nodes and can communicate with neighboring nodes in their
transmission range.
This review confirms that projective geometry concepts can be explored which facilitates
node deployment strategies and can provide possibly simple routing techniques. We choose to
explore and use Perfect Difference set – Networks for WSN.
2.2. Fixed Geometrical Deployment of Sensor Nodes
In Wireless Sensor Networks, nodes are randomly deployed. Most of the protocols in
WSN are designed for its random deployment. Projective Geometry can be used for the fixed-
geometrical deployment of wireless sensor nodes. This strategy can be used for deploying
nodes in smart houses, hotels, offices, underground mines, etc. For environmental monitoring
on large geographical areas, random deployment is used. In random deployment, many
redundant sensors are used which increases the cost. In fixed geometrical deployment, exact
number of nodes with their specific location can be deployed. This reduces the cost and
optimization of the network can be achieved. We propose the use of Perfect Difference Set in
the paradigm of WSN. Wireless Sensor Nodes are battery powered and their life span
depends on how energy consumption can be effectively managed for lower consumption at all
functional blocks. Amongst the three units of Wireless Sensor Node: sensing, computation
and communication, communication unit consumes the largest energy budget, hence all
research efforts are being made to reduce the energy consumption on communication. Many
energy aware routing protocols are designed on this line.
We propose the strategy to use PDS- networks in the paradigm of WSN. In PDS–
Network, nodes are interconnected to allow a minimum number of hops during
communication. Any node can send data to a cluster head node within a hop count of one or
two. Because of simplicity of connection amongst nodes, the routing algorithm becomes very
simple. In such cases, the relative position of nodes is always known. Hence the routing
algorithm will become very short, occupy less memory space and will be executed faster by
the node’s processor. This reduces the energy consumed by the computation unit.
Wireless sensor nodes also consist of a location finding system to know the location of
neighboring node. If the WSN are deployed using PDS-Networks, then all the nodes in the
network know their relative as well as absolute position. This removes the need of Location
finding system in the WSN which otherwise could consume additional energy. The fixed
deployment strategy could use PDS- Network for deployment of sensor nodes.
2.3. Perfect Difference Set
Projective geometry gives many mathematical approaches for arranging nodes in a

graphical manner, which is also the need in WSN. Perfect Difference Sets provide the
mathematical tools for achieving the optimum number of nodes. Perfect Difference Sets were
first discussed by Singer in 1938. The formulation was in terms of points and lines in a finite
projective plane. Perfect Difference Networks were brought forward by Behrooz et al.
[15,16]. More mathematical theorems and formulations in terms of projective geometry and
perfect difference set are discussed by Rey Casse [17].
The theory of finite projective planes is highly developed, but these mathematical notions
are not required for applications. We first present a theorem that forms the basis of the
definition of perfect difference sets and, then, the definition itself.
Theorem 1. - A sufficient condition that there exist +1 integers s0,s1 ,……., s, having
the property that their 2+  difference si – sj, ij, 0i, j, are congruent, modulo 2+ +1, to
the integers 1,2,….., 2+  in some order is that  be a power of a prime.
Definition 1. - Perfect Difference Set (PDS) – A set {s0,s1 ,……., s } of δ+1 integers
having the property that their 2+  differences si – sj, 0ij, are congruent, modulo 2+ +1,
to the integers 1,2,….., 2+  in some order is a perfect difference set of order δ. Perfect
difference sets are sometimes called simple difference sets, given that they correspond to the
special λ=1 case of different sets for which each of the possible difference is formed in
exactly λ ways.
PDS need not contain an integer outside the interval [0, 2+ ] because any integer outside
the interval can be replaced by another integer in the interval without affecting the defining
property of the PDS.
Theorem 2. Given a PDS {s0, s1, . . . . ,sδ} of order δ, the set {as0 + b, as1 + b, . . . . , asδ +
b}, where a is relatively prime to δ2 + δ+1, also forms a perfect difference set
By definition, any perfect difference set contains a pair of integers su and sv such that sv -
su ≡ 1 mod δ2 + δ+1. By Theorem 2 and the observation that preceded it, subtracting su from
all integers in such a PDS yields another PDS that contains 0 and 1.
Definition 2. Normal PDS: A PDS {s0, s1, . . . . ,sδ} is reduced if it contains the integers 0
and 1. A reduced PDS is in normal form if it satisfies si < si+1 ≤ δ2 + δ, 0 ≤ i < δ.
Definition 3. Equivalent PDSs: two different PDSs are equivalent if they have the same
normal form {s0, s1, . . . . ,sδ}.
Henceforth, we deal only with PDSs in normal form, some examples of which appear in
Table 2.1. Here the values of δ shown are powers of prime numbers and n = δ2 + δ+1. The
following section depicts the properties of PDS.
Property 1. Existence: Theorem 1 guarantees that a PDS exists for any number n that is
of the form δ2 + δ+1, where δ = ph and p is a prime number. It is suspected, though not yet
proven for arbitrarily large values of n that PDSs do not exist for other values of n. However,
practically speaking, this is not alarming, given that primes and their powers are quite
abundant, both in the range of practical interest for interconnection network size and
asymptotically. For example, there are 197 primes and powers of primes under 1,000.
Property 2. Multiplicity: For some values of δ, there exist more than one PDS. For
example, we have the following PDSs of order δ = 3:
0, 1, 3, 9 and 0, 1, 4, 6.
It is easily verified that all numbers in the interval [1, 12] can be formed as the mod 13
difference of numbers in each of the two sets above:
Links PDS(0,1,3,9) PDS(0,1,4,6)

1 ≡ 1–0 ≡ 1-0
2 ≡ 3–1 ≡ 6–4
3 ≡ 3–0 ≡ 4–1
4 ≡ 0–9 ≡ 4–0
5 ≡ 1–9 ≡ 6–1
6 ≡ 9–3 ≡ 6–0
7 ≡ 3–9 ≡ 0–6
8 ≡ 9–1 ≡ 1–6
9 ≡ 9–0 ≡ 0–4
10 ≡ 0–3 ≡ 1–4
11 ≡ 1–3 ≡ 4–6
12 ≡ 0–1 ≡ 0–1
Property 3. Generation: A PDS of order δ = ph, where p is a prime number, represents a

set of n points and n lines in the 3D Euclidian space such that each point is on δ + 1 lines and
each line contains δ + 1points. This geometric interpretation leads to a PDS of order δ = ph
being generated from an irreducible degree-3 polynomial in GF (ph). Here, we take it for
granted that a PDS of order δ = ph can be easily generated when required.
Property 4. Relationship with perfect partitions: PDSs are closely related to perfect
partitions, which have an even longer history. Take any PDS in normal form and find the
mod-n differences si+ 1 - si between consecutive numbers in it, including the difference s0 -sδ.
For example:
PDS (δ = 3, n = 13): 0, 1, 3, 9
mod-n differences si+ 1 - si : 1, 2, 6, 4.
Viewing this last sequence of integers as a circular one and adding subsequences of
length 1, 2, and 3 beginning with each term, yields each of the sums in the interval [1, 12]
exactly once as given below -
1 = 1
2 = 2
3 = 1+2
4 = 4
5 = 4+1
6 = 6
7 = 4+1+2
8 = 2+6
9 = 1+2+6
10 = 6+4
11 = 6+4+1
12 = 2+6+4
Such a mod-n sequence, which is also known as a perfect partition, ideal code, or ideal
ring proportions, can be used in synthesizing PDN type structures. However, PDSs provide a
more straightforward and efficient tool in this regard. Note that a PDS is transformed to a
corresponding perfect partition via modular subtraction of consecutive terms, while the
reverse transformation involves computing modular prefix sums.
A PDS allows us to express a large set of integers via a set of much smaller size, in a
simple and highly regular fashion. When used in the design of interconnection networks, this
property translates to a reduced number of links and switching elements or to more efficient
use of bandwidth.
2.4. Perfect Difference Networks
Consider the normal form PDS {0,1, s2 ,……., s } of order δ. We can construct a direct
interconnection network with n=2+ +1 nodes based on this PDS as follows
Definition 4. – Perfect difference network (PDN) based on the PDS {0,1, s2 ,……., s }-
There are n=2+ +1 nodes, numbered 0 to n-1. Node i is connected via directed links to nodes
i±1 and i± sj (mod n), for 2≤ j≤δ. The preceding connectivity leads to a chordal ring of in and
out-degree d = 2δ and diameter D=2. Because, for each link from node i to j, the reverse link
from j to node I also exists, the network can be drawn as an undirected graph.
An example PDN for n = 7, based on the PDS {0, 1, 3}, is depicted in Figure-4. Every
normal-form PDS contains 1 as a member. Therefore, PDNs based on normal-form PDSs are
special types of chordal rings. In the terminology of chordal rings, the links connecting
consecutive nodes i and i+1 are ring links, while those that connect nonconsecutive nodes i
and i + sj, 2 ≤ j ≤ δ, are skip links or chords. The link connecting nodes i and i + sj is a
forward skip link of node i and a backward skip link of node i + sj. Similarly, the ring link
connecting nodes i and i + 1 is a forward ring link for i and backward ring link for i + 1.
As seen in Figure 4, any two nodes in a PDN are either connected by a link directly or via
a path of length 2 through an intermediate node. This property is elaborated upon in Figure 5,
where a shortest path from node 0 to each of the other nodes is highlighted and labeled with
the associated difference si – sj. Given the node symmetry of the network, shortest paths
between other pairs of nodes are obtained by simply adding the index of the source node to all
path labels as seen in Figure 5.
0
6
1
5 2
4 3
Figure 4. Chordal Ring Structure of the PDN with 7 nodes on the Perfect Difference Set {0, 1, 3}.
Figure 5. PDN with n=7, =2, depicting shortest paths from node 0 and labeled with corresponding
differences.
Table 1 shows the formation of various PDS with  as power of prime numbers and n=2+
+1. Thus, there can be a minimum of 7 nodes with δ=2 or with δ=16 a network of 273 nodes
can be possible. PDS–Networks are better suited with small number of nodes in a wireless
environment.
Table 2 summarizes various parameter for PDS-Network. Network diameter D, defined
as the longest of the internodes distances, is an important figure of merit for networks. The
diameter D indicates the worst-case number of hops in sending a message from one node to
another. Network diameter plays a key role in communication latency, albeit in an indirect
way. This is best understood by considering the case of short and long messages separately.
For short worms, the travel time of the head, which is proportional to the hop distance,
dominates the overall message latency. For long messages, a significant number of links,
perhaps the entire source-to-destination path, is occupied by the worm carrying the message.
In networks with large diameters, the worms tend to be longer and, thus, occupy a greater
portion of the aggregate network bandwidth. This either increases the possibility of deadlock
or else forces us to use less aggressive routing algorithms.
Table 1. Formation of Various PDS
 n PDS of order  in normal form

2 7 0,1,3
3 13 0,1,3,9
4 21 0,1,4,14,16
5 31 0,1,3,8,12,18
7 57 0,1,3,13,32,36,54,63
8 73 0,1,3,7,15,31,36,54,63
9 91 0,1,3,9,27,49,56,61,77,81
11 133 0,1,3,12,20,34,38,81,88,94,104,109
13 183 0,1,3,16,23,28,42,76,82,86,119,137,154,175
16 273 0,1,3,7,15,31,63,90,116,127,136,181,194,204,233,238,255
Table 2. Summary of parameters for PDS-Network
Number of Nodes(n) δ2 + δ+1

Degree 2δ
Number of Links δ( δ2 + δ+1)

Network Diameter (D) 2
Average internode distance (∆) 2δ/( δ+1)
Bisection width (B) ≥ ( δ+1)( δ2 + δ+1)/4
Average internodes distance ∆ is defined as the average of the lengths of the distances
between all n(n-1) pairs of nodes, or perhaps between all n2 pairs of nodes when the distance
of each node to itself is also included in the averaging. The average internodes distance ∆ is
representative of average or expected communication latencies, whereas D represents the
worst case. For such networks, the average internodes distance ∆ generally grows in
proportion to D, even though the relationship between the two parameters is not strictly
linear. Put in another way, if the diameter of the network is quadrupled, the average
internodes distance at least doubles.
The bisection width B of a network is the minimum number of links whose removal cuts
the network in two parts, with n=2c nodes on one side of the partition and dn=2e nodes on the
other. Bisection bandwidth is defined in terms of link bandwidths, rather than multiplicity, for
networks in which the links have varying communication capacities. A large bisection (band)
width is an indicator of large aggregate network capacity for routing random traffic patterns
between arbitrary network nodes. For example, the hypercube with a bisection width of n=2
can achieve a higher communication performance for random traffic than a 2D mesh with
bisection width n1/2. The mesh is, in turn better than constant-bisection networks such as rings
and trees.
These parameters for the interconnection networks are elaborated in [16] and these
parameters need to be analyzed for WSN.
2.5. Collective Communication in PDS Networks
In PDS networks, there are different possibilities of collective communications – one to

all, all to all, complete exchange amongst all pairs of nodes and all to one. One to all
broadcasting is possible from any node as shown in Figure 6. The maximum number of
messages that can be broadcast will be δ+ δ( δ-1)+ δ= δ+ δ2 messages. Nodes 5 and 2 are
connected through 2 hops while all other nodes are connected through 1 hop. All to all
broadcasting is possible as shown in Figure 7. Node 0 can broadcast with all the nodes.
Complete exchange of messages amongst all pairs of nodes is possible as depicted below:
1. x → x+si and x → x-si (2 δ messages)

2. x → x+si- sj (2 δ (δ-1) messages)
6
1
5 2
4 3
Figure 6. One-to-All broadcasting.
6
1
5 2
4 3
Figure 7. All-to-All broadcasting.
0
6
1
5 2
4 3
Figure 8. All to one PDS-WSN with 7 nodes on the Perfect Difference Set {0, 1, 3}.
0123456 0123456
 
3456012 2345601
n -message 2n Message
Figure 9. Shift Permutation routing of messages
Cluster Head or
6 Gateway Node
5 1
0
2
4
Figure 10. Cluster of PDS-WSN with 7 nodes on the Perfect Difference Set {0, 1, 3} with circular
deployment.
In all to one broadcasting, all nodes can broadcast to the selected central node as shown
in Figure 8. This particular topology is of use in WSN. Routing in PDS-Network is very
simple and efficient, one called shift permutation routing. In shift permutation routing,
messages can be routed amongst nodes through one hop or two hop connectivity i.e. n or 2n
messages as depicted in Figure 9.
2.6. Perfect Difference Set - Wireless Sensor Network

(PDS-WSN)
For a better understanding of the PDS-WSN, we described a cluster of 7 nodes of WSN,

as shown in Figure 5. Nodes 1, 3, 4 & 6 have single hop connectivity with node 0, while
nodes 5 & 2 have two hop connectivity with node 0. Seven node PDS networks can be
arranged in a circular fashion as shown in Figure 10. Central node 0 is designated as cluster
head. All nodes send the collected data to the cluster head from time to time as scheduled.
We implemented WSN with 48 nodes in 4 clusters each of 12 nodes and one cluster head.
This gives WSN of 48(12x4) nodes, we call it WSN (48).
Figure 11. Cluster of 12 nodes with one cluster head implemented in Ptolemy-II- Visual Sense.
Figure 12. WSN (48) with 4 clusters, each with 12 nodes as shown in Figure -11.
2.7. Routing in PDS-WSN
Perfect Difference Set can be used for WSN [18] for fixed geometrical deployment,
where any node can send data to cluster head with one or two hop counts. WSN with 48
nodes scattered in four clusters is implemented in visual sense of Ptolemy-II. There are four
clusters each of 12 nodes and one cluster head as shown in Figure 12. From the cluster head,
data is collected at base station. In another implementation PDS – WSN with δ=3, n=13,
PDS= {0, 1, 3, 9} is implemented for each cluster; the node interconnections are shown in
Figure 2.10. During routing, each node knows the path towards cluster head. Shift
Permutation Routing is used, wherein either n or 2n messages are communicated to the
cluster head. In a shift permutation, each node x sends a message to node x + c, where c is a
constant. When c = ±si for some i, shift permutation can be performed in one step with no
conflict. In all other cases two steps are needed, with all intermediate nodes x + si and thus the
links used are distinct. Actually, a more general form of shift permutation in which the shift
amount is +c for some nodes and -c for some others can be performed in the same number of
steps, given that the forward moving and backward-moving messages do not interfere with
each other.
2.8. Clustering Wireless Sensor Network with PDS-WSN
PDS-WSN facilitates clustering in wireless sensor nodes. One such experiment is

performed here. We have a basic cluster of 12 nodes with one cluster head i.e. with 13 nodes
PDS network as shown in Figure 12 and Figure 13. SIX such clusters are formed; cluster head
of each of these clusters are connected with second tire of PDS network of 7 nodes as shown
in Figure -14. Thus, there are 72 active nodes in this network called WSN (72). There can be
as many such clusters as needed by the application. All bottom layers of these clusters are
then connected with second tire of PDS- network. Thus, it is possible to deploy huge
population of senor nodes under the PDS-WSN.
The architecture of ART1[19] classifiers also supports PDS-WSN. The two tire classifier
architecture can be effectively implemented with PDS-WSN. At the ground layer, clusters
there can be ART1[20] classifier in each nodes so that classified data can be send to its cluster
head. At the second tire of PDS-network, we can implement second tire of classifier. Here,
ART1[21] classifier is implemented in each cluster head so that sensor data classified at each
ground layer node is further classified at second layer. This can greatly reduce the data traffic
communicated to base station. This peculiar architecture can be used for a huge population of
sensor nodes specifically for monitoring environmental parameters over large geographical
area. The second tire of classifier at cluster head is optional.
The experimental results of counting lifetime of WSN with different formulations of
PDS-WSN are presented in the next section.
0
12 1
2
11
3
10
4
9
5
8
7 6
Figure 13. PDS Network with =3, n=13, PDS= {0, 1, 3, 9}.
Figure 14. WSN (72) with 6 clusters, each with 12 nodes as shown in Figure 2.8.
2.9. Experimental Results
The experimental results are presented in the forms of charts. Chart 2.1 presents the
network lifetime of the proposed algorithm in comparison with the Cooperative Routing
algorithm[19,20]. The difference in all the sets of simulation is the seed used for
randomisation. The seeds for randomisation determine the network topology. The update
cycles in the Gateway node that measures the time right from setup phase until the first node
dies, calculates the lifetime. It is evident from the above table that there is an appreciable
improvement in the network lifetime. The improvement varies according to the network
topology. In Chart 2.1, the improvement in lifetime is consistently around 5%. The maximum
network lifetime improvement is found to be 6.21%.
Chart 2.1 Improvement in Lifetime of PDS-WSN(48)
200
195 Lifetime with
190
Life Time
Cooperative
185 routing
180
175 Lifetime with
170 PDS-WSN(48)
165
160
155
150
145
Iterations of 1 2
Randomisation
3 4 5 6
Chart 2.2 describes the Lifetime of Network for PDS-WSN with and without
Classification. Here with classification the network lifetime is improved by around 41.08%
with maximum improvement of 42.52% and minimum 39.56% with different seeds of
randomisation of the nodes.
Chart 2.3. Describes the lifetime of PDS-WSN with 72 nodes. When clusters are
increased to 7, the lifetime of the overall network falls a little bit as compared to that of PDS-
WSN with 48 nodes. Still, it is showing the increase as compared to cooperative routing with
50 nodes.
Chart 2.4 describes the lifetime improvement in PDS-WSN (72) with classification
techniques, which is about 40.82% on average.
CHART 2.2 - Improvement in L ifetime of PDS-WSN(48) with

Classification
300
250 PDS-WSN(48)
Without
200
Lifetime
Classification
150
PDS-WSN(48)
100 With
Classification
50
0
Iterations of
Randomisation 1 2 3 4 5 6
Chart 2.3 Improvement in Lifetime with PDS-WSN(72)
200
190 Lifetime with
Lifetime
Cooperative
180
routing
170
Lifetime with
160 PDS-WSN(72)
150
140
1 2 3 4 5 6
Iterations of Randomisation
Chart 2.4 - Improvement in Lifetime of PDS-WSN(72) with

Classification
300 Lifetime
PDS-WSN(72) with
250 out
classification
200
Lifetime
150 Lifetime
Life Time PDS-WSN(72) with
100 classification
50
0
1 2 3 4 5 6
Chart 2.5 -Comparison of Lifetime - Cooperative Routing with PDS-WSN

(48), PDS-WSN (72), without classification
210
200 Lifetime
Cooperative
190 routing
180 Lifetime
PDS-
Lifetime
170 WSN(48)
160 Lifetime
PDS-
150 WSN(72)
140
1 2 3 4 5 6
The above chart presents the comparison of lifetime in case of cooperative routing with
PDS-WSN (48) and PDS-WSN (72) without using classification techniques.
2.10. Limitations of PDS-WSN
1. It can be used only for fixed (non-mobile) deployment strategy

2. Number of nodes (n=2+ +1) that can be deployed in each cluster must follow PDS
definition i.e.  must be prime number. Hence as shown in table-9.1, the number of
nodes can be 7, 13, 21,…., 253. To suit the application, few idle routing nodes may
be deployed to satisfy the number of nodes criteria.
3. It cannot be used for bigger clusters with greater diameters. Energy consumption
increase with bigger clusters as the transmission range between nodes increases.
Hence, we suggest carrying out optimisation to decide the maximum number of
nodes and the maximum possible diameter of the PDS-WSN cluster.
2.11. Conclusion
PDS-WSN uses the mathematical notion of Prefect Difference Set for the network
interconnections in a peculiar geometrical form. Hence, PDS-WSN can be effectively used
for fixed (non mobile) deployment of wireless sensor nodes.
PDS–Networks provide the simplest possible interconnections amongst nodes, which
simplify the routing technique. All the nodes in the cluster know their relative as well
absolute location in the network. Hence, such sensor nodes do not require a location-finding
system, which is usually the part of commercially available nodes. The absence of a location
finding system also contributed to the low power consumption. The simple shift permutation
routing has been used in PDS-WSN, which can be efficiently used for VLSI implementation
of low power – low cost sensor node. The use of two radios - normal data transmission radio
and MAC radio, also contribute to the reduction of energy consumption during
communication.
Two cases of PDS-WSN are implemented. In the first case, four clusters, each of 12
nodes, and in another case, six clusters each of 12 nodes, were implemented. Improvement in
the lifetime of WSN as shown in Chart 2.1 is achieved because of the simplest routing
technique and reduced number of hop counts. Improvement in the lifetime of WSN is about
5% with WSN (48), and with WSN (72) the life time of the WSN is improved by about 4.5%.
This is achieved because of the simple routing technique and reduced number of hop counts.
Thus, this implementation reduces the consumption of energy by computational unit. Many
energy-aware protocol designs are striving for reducing the energy consumption on
communication. The combination of PDS-WSN cluster and energy aware routing for a large
population of sensor nodes can effectively reduce the energy consumption of the whole
network and thereby add to the lifespan of WSN. Cooperative routing is normally popular for
equitable energy consumption over the network and thereby improving average life of the
network.
Classification techniques using ART1 neural network model provides dynamic
classification of sensor data, which reduces the data traffic to be handled by communication
radio. This reduces the energy spent on the radio unit and improves the lifespan of the
network. With WSN (48), the lifetime is improved by about 41.08%, and for WSN, (72) the
lifetime improvement is about 40.82% as mentioned in the comparative charts 2.1 – 2.5.
PDS-WSN facilitates clustering architecture for WSN. Two tire PDS networks can be
implemented with ART1 classifiers at each tire, which adds to great dimensionality reduction
in data traffic of communication of the whole sensor network and achieves improvement in
the lifetime of WSN.
REFERENCES
[1] Daniele Puccinelli and Martin Haenggi,”Wireless Sensor Networks: Applications and
Challenges of Ubiquitous Sensing”, IEEE circuits and systems magazine, third quarter,
2005, 19-29.
[2] Akyildiz, IF; Su, W; Sankarasubramaniam, Y; Cayirci, E. ”Wireless sensor networks: a
survey”, Computer Networks, 2002, 38, 393-422.
[3] Carlos, F; García-Hernández†, Pablo, H. Ibargüengoytia-González†, Joaquín García-
Hernández†, and Jesús A. Pérez-Díaz.,”Wireless Sensor Networks and Applications: a
Survey”, IJCSNS International Journal of Computer Science and Network Security,
2007, VOL.7 No.3, March 264-273.
[4] Lewis, FL. “Wireless Sensor Networks, Smart Environments: Technologies, Protocols,
and Applications”, DJ; Cook, SK. Das, John Wiley, New York, 2004.
[5] Chee-Yee Chong, and Srikanta P. Kumar, ”Sensor Networks: Evolution, Opportunities,
and Challenges”, Proceedings of the IEEE, VOL. 91, NO. 8, AUGUST 2003, pages:
1247-1256.
[6] Vijay Raghunathan, Curt Schurgers, Sung Park, and Mani B. Srivastava, “Energy-
Aware Wireless Microsensor Networks”, page(s): 40-50, IEEE signal processing
magazine, March 2002.
[7] Matthew, J; Miller, Nitin, Vaidya, H. “A MAC Protocol to Reduce Sensor Network
Energy Consumption Using a Wakeup Radio”, IEEE Transactions On Mobile
Computing, VOL. 4, NO. 3, MAY/JUNE 2005, pages, 228-242.

[8] Gracanin, D; Eltoweissy, M; Olariu, S. Wadaa, ”On modelling wireless sensor
networks,” A. Parallel and Distributed Processing Symposium, 2004.
[9] Xiong, Y. “An Extensible Type System for Component-Based Design,” Technical
memorandum UCB/ERL Mo2/13, University of California, Berkely, CA94720, May 1,
2002.
[10] Baldwin, P; Kohli, S; Lee, EA; Liu, X; Zhao, Y. "Modelling of Sensor Nets in Ptolemy-
II,” In Proc. of Information Processing in Sensor Networks, (IPSN), April 26-27, 2004,
359.
[11] Saket Sakunia, Shantanu Bhalerao, Abhishek Chaudhary, Mukund Jyotishi, Mandar
Dixit, Raghav Jumade, Rajendra Patrikar, “UbiSens: Achieving Low Power Wireless
Sensor Nodes,” IFIP International Conference on Wireless and Optical
Communications Networks, (WOCN 2007), page(s), 1-6.
[12] Amrutur, BS. Joshi, R. Karmarkar, NK. AT&T Bell Labs., Murray Hill, NJ, “A
projective geometry architecture for scientific computation”, Proceedings of the
International Conference on Application Specific array processors, page(s): 64-80.
[13] Pingyi Fan, Member, IEEE, Guansheng Li, Kai Cai, and K. B. Letaief, Fellow, IEEE,
“On the Geometrical Characteristic of Wireless Ad-Hoc Networks and its Application
in Network Performance Analysis”, IEEE Transactions On Wireless Communications,

VOL. 6, NO. 4, APRIL 2007, 1256-1264.
[14] Muhammad, Rashid Bin. “A Distributed Geometric Routing Algorithm for Ad Hoc
Wireless Networks”, Fourth International Conference on Information Technology,
2007, ITNG apos; 07, Volume , Issue , 2-4 April 2007 Page(s), 961-963.
[15] Behrooz Parhami, Fellow, IEEE, and Mikhail Rakov, ”Perfect Difference Networks and
Related Interconnection Structures for Parallel and Distributed Systems”, IEEE
transactions on parallel and distributed systems, vol. 16, no. 8, august 2005, 714-724.
[16] Behrooz Parhami, Fellow, IEEE, and Mikhail Rakov, “Performance, Algorithmic, and
Robustness Attributes of Perfect Difference Networks”, IEEE transactions on parallel
and distributed systems, vol. 16, no. 8, august 2005, 725-736.
[17] Rey Casse, “Projective Geometry an Introduction”, Oxford University press, 70-81.
[18] Akojwar Sudhir, G. and Patrikar, Rajendra, M. Perfect Difference Set - Networks for
Wireless Sensor Networks, 4th IEEE International Conference on Communications and
Sensor Networks, IIIT Allahabad, INDIA, Dec 27-29, 2008. 95-100.
[19] Patrikar, Rajendra, M and Akojwar, Sudhir, G. Neural Network Based Classification
Techniques For Wireless Sensor Network with Cooperative Routing, 12th WSEAS
International Conference on COMMUNICATIONS, Heraklion, Greece, July 23-25,
2008, page(s) 433-438.
[20] Akojwar, Sudhir, G and Patrikar, Rajendra. M. Improving Life Time of Wireless Sensor
Networks Using Neural Network Based Classification Techniques with Cooperative
Routing, International Journal of Communications, issue- 1 vol -2 2008, 75-86.
[21] Neural networks Algorithms, Applications, and Programming Techniques,
J.A.Freeman, David, M. Skapura, Pearson Education Asia, 291-339, 2001.
In: Wireless Sensor Networks ISBN: 978-1-61728-125-9
Editor: Liam I. Farrugia © 2011 Nova Science Publishers, Inc.
Chapter 2
KEY MANAGEMENT PROTOCOL FOR

HETEROGENEOUS SENSOR NETWORKS
Ma Chun-Guang1, Wang Jiu-Ru1, Geng Gui-Ning2,

Shang Zhi-Guo1 and Zhang Bing-Zheng1
1
College of Computer Science and Technology, Harbin Engineering University,
Harbin, 150001, China
2
School of Computer Science and Technology, Beijing University of Posts
and Telecommunications, Beijing, 100876, China
1.1 INTRODUCTION
In this chapter, we will talk about Heterogeneous Sensor Networks key Management
Protocol.
At the first of this part, we will talk about what is Heterogeneous Sensor Networks
(HSN), and how to divide key management protocol by different method.
The second part of this chapter, based on different network threats and assumptions, we
will introduce one key pre-distributed protocol, two key management protocols. About the
key pre-distributed protocol, the key ring of nodes is based on divisible core pair-wise
balanced design. We design two key pre-distributed protocol, Divisible Core Pair-wise
Balanced Design Scheme shorted by DCPBD, and variant DCPBD shorted by VDCPBD.
One of the key management protocols, we propose a key management protocol for
heterogeneous sensor networks to improve the random key pre-distribution scheme using
deployment knowledge of nodes and the prior area deployment information. Another key
management protocol, we propose a Location-aware and secret share based dynamic key
management protocol to effectively replace the compromised central node and enhance the
security level of the network.
At the last of this chapter, we will summarize this chapter, and predict the future
development of key management.
28 Ma Chun-Guang, Wang Jiu-Ru, Geng Gui-Ning et al.
1.2 HETEROGENEOUS SENSOR NETWORKS KEY MANAGEMENT

Early research in wireless sensor networks (WSN) focused on homogeneous networks in
which all nodes possess identical software and hardware. However, due to the constant
characteristics of WSN, most existing key management schemes developed for the traditional
wire-line and wireless networks may not be feasible for WSN (for example, PKI-based
solutions). Recently, because of their potential to increase network security, lifetime and
reliability without significantly increasing the cost, it has made the heterogeneous sensor
networks (HSN) popular [1].
In HSN, typically, consists of a base station, a small number of powerful cluster head
nodes (H-sensors) and a large number of data sensing nodes (D-sensors). H-sensors have
sufficient energy resources, great computation ability and wild communication range. D-
sensors have limited computation ability, energy resources and communication range, and are
used to sense data in the detecting region. D-sensors are grouped into clusters. Each cluster is
controlled by an H-sensor. H-sensors provide data collecting, fusion and transport and can
broadcast messages to all sensors in the cluster. The three layer HSN is depicted in Fig. 1.[2]
Base Station
Cluster Head
Key Generation Node
Sensor Node
Figure 1. Heterogeneous Sensor Networks
According to the different key distribution methods, the key management protocol can be
divided into random key management scheme and deterministic key management scheme. In
random key management scheme, the key ring of node is obtained randomly. In the
deterministic key management scheme, the key ring is obtained in deterministic manner,
avoiding the blindness of random key management scheme, and gaining higher key
connectivity rate and take full advantage of the node storage space.
According if the network can dynamically add or evict nodes and refreshing the key(s) of
its nodes periodically or requested, the key management can be divided into static key
management and dynamic key management. Static key management schemes perform key
management functions statically prior to or shortly after the deployment of the network
deployment. Accordingly, lost keys due to node capture and/or failure are not compensated.
The main feature of dynamic key management schemes is repeating the key management
process either periodically or on demand to respond to node capture.
Key Management Protocol for Heterogeneous Sensor Networks 29
1.3 A NOVEL KEY PRE-DISTRIBUTION PROTOCOL BASED ON PAIR-

WISE BALANCED DESIGN
How to construct the key ring is the most important issue in the deterministic key
distribution schemes. There is a problem in DCPBD scheme that sub-key-ring used in the
secure communication among cluster head nodes is fixed, which can not change according to
the scale of nets and has a low anti-captured ability. Then, through the construction of the
basic set of core  in two-dimensional SBIBD and the extension of the special block design
method of DCPBD, we propose a variant protocol Variant Divisible Core Pair-wise Balanced
Design Scheme (VDCPBD), which is more conformed to the application of HSN. VDCPBD
could be better applied to large-scale HSN, and improved the safety performance of cluster
head nodes and common nodes connectivity significantly.
1.3.1 Basic Knowledge
Definition 1[3]: It is supposed that   {B1 , B2 , Bb } is a block design based on v -

set S  {s1 , s2 ,  sv } . If  meets the following conditions:
(1) (1) B j is a constant independent of j 1  j  b  ;

(2) For any element s of S , the number of subsets in  including s is a constant r
independent of s ;
(3) For any binary subset {s i , s j } , the number of subsets in  including it is a
constant  independent of {s i , s j } .
Then the block design  is (b, v, r , k ,  ) - BIBD， (b, v, r , k ,  ) -design for short,
which is required to satisfy  (v  1)  r (k  1) , bk  vr ,in particular, it is called SBIBD
when b  v .
Definition 2[3]: It is supposed that   {B1 , B2 , Bb } is a block design based on v -
set S  {s1 , s 2 ,  s v } and K  {k1 , k 2 ,  k m } .If  meets the following conditions:
(1) (1) B j  K 1  j  b ；
(2) For any binary subset of S , the number of blocks in  including it is a concrete
constant with independent of the binary subset.
Then  is a pair-wise balanced design (PBD) of set S . If the constant in (2) is  , then
the PBD is defined as PBD( K ;  ; v) or PBD({k1 , k2 , km };  ; v) .The PBD is defined as
PBD(k ; ; v) when K  {k} .
In BIBD, when the condition " B j is a constant independent of j 1  j  b  " is deleted,

PBD is obtained.
Camtepe first uses BIBD to WSN distributed key management protocol, and proposed
SD protocol based on combination design. Moreover, he applies BIBD to Homogeneous
WSN key pre-distribution scheme, which better solves the exist problem of random scheme.
But it could not be directly used in HSN. So we propose the key pre-distribution scheme
based on PBD, which can be directly applied to HSN, and the structure is simpler.
1.3.2 Pair-wise Balanced Design Based Key Pre-distribution Scheme
1.3.2.1 DCPBD Scheme

A divisible core special block design on the pair-wise balanced block design is the
theoretical basis in this paper.
Definition 3 [4]: It is supposed that there is PBD( K ;  ; v) in the basis set S , and
v 1
| S | v , m  K ,   S and m  1 | v  1 ,  , S \ { }  S1  S 2    S  ,
m 1
S i  S j   (i  j ) , | S i | m  1(1  i   ) .
If Si {}(1i  ) appears  times as a block exactly, then the design is called a
divisible core design.  is called the core, Si { }(1  i  ) is called the special block.
Specific parameters are shown in table 1.
Table 1. the mapping between Key Pre-distribution Scheme and PBD parameters
PBD
Key Pre-Distribution
Element Comment
S Base set Key pool
v Size of base set Number of keys
Block Key chain
B Particular block Cluster key chain
k Size of block Size of key chain
r Time of element used Time of key used
b Number of blocks Number of nodes
Number of shared keys in two key

 Size of a intersection of two blocks
chains
Algorithm 1: DCPBD Algorithm

InputⅠ: The number of common nodes N c .
InputⅡ: The number of cluster head nodes N h . (If take it directly, or we take the
smallest prime which is bigger than the number).
Order: Common node storage space is N c cluster head node storage space is N h .
(1) The maximum of q is taken to meet q 2  N c , and q  N h is taken, if N h  q ,
otherwise q is unchanged.
2
(2) The (q , q,1) -design is generated by mode difference method and a parallel
class {S1 , S q } is taken. [5]
(3) (3)The special block S j   1  j  q  is obtained.
Output: PBD({K c , K h };  ; v) that is PBD ({q, q  1};1; q  1) -design.

2
For a given (q , q,1) -design,   {Bij' | i  Z q*1 , j  Z q*1} ,
Bij  {q1i , qt 1, j (i 1)(t 1) | t  Zq*} is first determined, and then the block is gotten. The detailed
results of PBD({3, 4};1;10) is shown in table 2, K1 ,  K 9 are in the basis set, and K 0 is
the core  , and the special block is B10  B12 .
1.3.2.2 VDCPBD Scheme

Definition 4: It is supposed that   {B1 , B2 , Bb } is a block design based on v -set
S and K  {k1 , k 2 , k m } .If  meets the following conditions:
(1) (1) B j  K 1  j  b ;
(2) For any binary subset of S , the number of blocks with the same length in 
including it is a concrete constant with independent of the binary subset.
Then  is called a variant pair-wise balanced design of set S . If the order of the
constants in (2) is {1 , 2  k } according to K , then the variant pair-wise balanced design
is defined as VPBD({k1 , k2 , ki };{1 , 2 , i }; v) -design.
Algorithm 2: VDCPBD Algorithm
InputⅠ: The number of common nodes N c , the storage space of common nodes K c .
InputⅡ: The number of cluster head nodes K h , the storage space of cluster head nodes
K h .(The number of actual cluster head is taken directly if it is a prime, or we take the
smallest prime which is bigger than it).
Table 2. PBD({3, 4};1;10) Matrix
K1 K2 K3 K4 K5 K6 K7 K8 K9 K0
B1 1 0 0 1 0 0 1 0 0 0
B2 1 0 0 0 1 0 0 1 0 0
B3 1 0 0 0 0 1 0 0 1 0
B4 0 1 0 1 0 0 0 1 0 0
B5 0 1 0 0 1 0 0 0 1 0
B6 0 1 0 0 0 1 1 0 0 0
B7 0 0 1 1 0 0 0 0 1 0
B8 0 0 1 0 1 0 1 0 0 0
B9 0 0 1 0 0 1 0 1 0 0
B10 1 1 1 0 0 0 0 0 0 1
B11 0 0 0 1 1 1 0 0 0 1
B12 0 0 0 0 0 0 1 1 1 1
(1) Algorithm 1 (1).

(2) Algorithm 1 (2).
(3) The number of core  is computed as K h  K c  k .
(4) The (q, k ,  ) -SBIBD design is generated [3], q  q .The q blocks non-
repeated is taken as  j ,1  j  q .
(5) The particular block S j   j , 1  j  q  is gotten.
Output: VPBD ({K c , K h };{1,    1}; q 2  q) -design.

The detailed results of VPBD ({3, 6}; 1, 2 ;16) is shown in table 3. The  in the
special blocks B10 , B11 , B12 is  1 ,  2 ,  3 separately, which are any three blocks taken from
(7,3,1) -SBIBD with the basis set K 01 , K 07  .
1.4 AN IMPROVED KEY MANAGEMENT PROTOCOL FOR HSN

In this part, we present an improved key management protocol for HSN based on the
random key pre-distribution protocol. The protocol exploits deployment knowledge of nodes
and the prior region deployment information. We divide the network sensing region into
several sub regions; divide the key pool into several key pools, and divide nodes into several
groups. Nodes in a certain group randomly select some keys from the corresponding key
pools are deployed at the corresponding sub regions. This protocol can improve the
probability that neighboring nodes share common keys at a certain extent. It also has great
scalability, high connectivity, low memory requirement and a stronger resilience toward node
capture. An example of the network model is shown in Fig 2, where suppose the number of
network deployment sub regions is 9.
Figure 2. The model of HSN
Table 3. VPBD({3,6};{1, 2};16) Matrix
K1 K2 K3 K4 K5 K6 K7 K8 K9 K01 K02 K03 K04 K05 K06 K07

B1 1 0 0 1 0 0 1 0 0 0 0 0 0 0 0 0
B2 1 0 0 0 1 0 0 1 0 0 0 0 0 0 0 0
B3 1 0 0 0 0 1 0 0 1 0 0 0 0 0 0 0
B4 0 1 0 1 0 0 0 1 0 0 0 0 0 0 0 0
B5 0 1 0 0 1 0 0 0 1 0 0 0 0 0 0 0
B6 0 1 0 0 0 1 1 0 0 0 0 0 0 0 0 0
B7 0 0 1 1 0 0 0 0 1 0 0 0 0 0 0 0
B8 0 0 1 0 1 0 1 0 0 0 0 0 0 0 0 0
B9 0 0 1 0 0 1 0 1 0 0 0 0 0 0 0 0
B10 1 1 1 0 0 0 0 0 0 1 1 1 0 0 0 0
B11 0 0 0 1 1 1 0 0 0 1 0 0 1 0 1 0
B12 0 0 0 0 0 0 1 1 1 1 0 0 0 1 0 1
1.4.1 Assumption
(1) H-sensors are equipped with tamper-resistant hardware. It is reasonable to

assume that a H-sensor can not expose all the information on stored it when it
was compromised.
(2) Assume that if an adversary compromises a D-sensor, he can extract all the key
material, data, and code on stored it.
1.4.2 The Scheme Description
Our scheme consists of three phases: key pre-distribution, shared-key discovery, and
path-key establishment. Because we adopt the pre-deployment knowledge, prior area
information and some powerful nodes, all phases are considerably different from the random
key pre-distribution protocol [6].
Step 1: key pre-distribution phase.
This phase is performed offline and before the deployment of sensor nodes. First we need
to divide the network sensing area D into u  v sub regions Di j , 1  i  u , 1  j  v with
Di j corresponding to the key pools S i j and deployment group Gi j . After completion of key
pools setup, for each D-sensor and H-sensor in Gi j , we randomly select m and M keys from
its corresponding key pool S i j , and load these keys and its identifiers into the memory of the
node.
Step 2: shared-key discovery phase.
After deployment, HSN is divided into multiple clusters according to the clustering
algorithm.[7] After completion of clusters forming, each node needs to discover whether it
shares any keys with its neighbors. To do this, each D-sensor could broadcast message:
a, K i j ( , ID ),1  i  m , where a is a random number. The decryption of a, K i j ( , ID )
with the proper key by a recipient would reveal  , ID and establish a shared key with the
broadcasting node.
After the above step, most D-sensor and H-sensor can establish direct shared-key. These
D-sensors are referred to as several 1 hop neighbor nodes of H-sensors, which is defined as
follows:
Definition 5 . A D-sensor that shares at least one key with H-sensor and its
communication range covers the H-sensor is referred to as a 1-hop neighbor node of the H-
sensor [8].
Step 3: the path-key establishment
After shared-key discovery phase, there is a direct shared-key graph in the cluster and
between clusters. This graph consists of the nodes that establish the direct shared-key and
their security link. The rest of nodes that can not establish the direct shared-key could find the
path key through this graph.
After this phase, D-sensors that haven’t established the direct shared-key with H-sensor
need to find (k+1)-hops path-key (for k=1, 2,…). These D-sensors are referred to as several
(k+1)-hops neighbor nodes of H-sensors, which is defined as follows:
Definition 6. A D-sensor that hasn’t established k-hops path-key with H-sensor, but
shares at least one key with some k-hops neighbor nodes and its communication range covers
the k-hops neighbor nodes is referred to as a (k+1)-hops neighbor node of the H-sensor [8].
1.4.3 Setting Up Si,j
In this subsection, we show how to assign keys to each key pools S i j , such that S i j and
neighbors of S i j have a certain number of common keys. Based on Du et al.’s protocol [9],
we adopt a new method of setting up key pools and determining | SG | in our scheme, it is
different from Du et al.’s scheme. In our protocol, we have:
(1) Two horizontally, vertically and diagonally neighboring key pools share exactly
a | SG | keys, where 1/ 8  a  1/ 4 . This property is illustrated in Fig.3 (a).
(2) To the group Gi j and its corresponding the key pool S i j , a | Si1, j1 | and a | Si1, j | ,
a | Si1, j1 | and a | Si, j1 | , a | Si1, j1 | and a | Si1, j | , a | Si1, j1 | and a | Si1, j1 | , share
exactly b | SG | keys, where a | Si1, j1 | Si, j ,…, a | Si1, j | Si, j 0  a 1/ 4 , 8a  4b  1.
This property is illustrated in Fig.3 (b).
Given the global key pool S and the overlapping factor a and b, we now describe how we
can select keys for each key pool S i j for i=1,…,u and j=1,…,v. The following procedure
describes how we choose keys for each key pool:
(a)
(b)
Figure 3. The relationship of overlapping factors and key pools.
(1) For group S1,1 , select | SG | keys from S, then remove these | SG | keys from S.
(2) For group S1, j , for j=2, 3, ... , v, select a | SG | keys from S1, j 1 , then select
  (1 a)| SG | keys from S, and remove σ keys from S.
(3) (3)For group S i j , for i=2, 3,…u, and j=1,2,…v, select a | SG | keys from each of
Si1, j and Si, j1 , Si1, j 1 and Si1, j 1 , if they exist; then select σ keys from S ,and
remove σ keys from S. According to the difference between the deployment sub
regions of nodes in each group, and the difference between neighboring key pools, σ
may have the following several values:
 1 2a  b  | SG |, (for j 1)


  1 4a  2b  | SG |, (for 2  j  v 1) (1)
 13a  b  | S |, (for j  v)
 G
Now we can calculate the size of the key pool | SG | for each group. Since keys selected
from the other groups are all distinct, the sum of all the numbers of keys should be equal to
| S | .Therefore, we have the following equation:
|S|
| SG | (2)
uv  (4uv  3u  3v  2)a  2(uv  u  v  1)b
This protocol, which uses deployment knowledge and prior region information, shows
many advantages in improving the random key management. The probability that two nodes
share common key rises with the increasing number of deployment sub regions, which can
efficiently reduce communication overhead for finding path-keys. Furthermore, connectivity
of neighboring regions could rise with increment of overlapping factor. Simulation results
show that the scheme has great scalability, high connectivity, low memory requirement and a
stronger resilience toward node capture.
1.5 A LOCATION-AWARE AND SECRET-SHARE BASED DYNAMIC

KEY MANAGEMENT PROTOCOL
In this part, we proposed a location-aware and secret share based dynamic key
management protocol (LASSB), which bases on EBS [10] and SEHLL [11], to effectively
enhance the security level of the network and replace the compromised central node.
1.5.1 Basic knowledge
Key generate Node: Every cluster choose two cluster heads which have larger energy and
share communication keys with it in order to generate the administrative keys for it.
Gateway nodes: All cluster heads also can be act as Key Generating Node or Key Storage
Node, we called all the cluster heads the gateway nodes.
Original cluster head: When parts the keys of a cluster head are kept in the other clusters,
the former cluster head is called original cluster head.
Key storage nodes: The cluster head node chooses two or more gateway nodes to store its
own part keys, with the principle of nearby physical location and larger energy from all
gateway nodes of communication nearby. We call these two or more gateways as key storage
nodes.
Each sensor node stores: K sg , KSCH , KS Key , administrative keys, and one way hash
function. Each cluster head stores: K gc , K sg , inter-gateway keys, EBS table and key storage
table, KSCH . The base station stores: All the valid IDs, EBS table, one way hash function, all
the preloaded keys to sensors: K sg , KSCH , KS Key .
Table 4. The list of notation
Notation Description
CN Base station
Cj Cluster j
GWi Gateway i
N The total number of sensors in the network

n The total number of sensors in a cluster j
Kgc Key between gateway and base station
Ksg Sensor discovery node
KSCH[Sj] Keys shared between sensor Sj and its cluster head
CK[j] Session for the cluster j
F The counter
1.5.2 Assumption
 All the keys’ IDs and locations are known; and secretly stored the central node’s
keys into two H-Sensors, so that the keys can be effectively protect from the
adversary.
 The adversary can not launch a coordinated attack.
Another random document with
no related content on Scribd:
The Project Gutenberg eBook of Kantopään
lukuhuoneella
This ebook is for the use of anyone anywhere in the United States
and most other parts of the world at no cost and with almost no
restrictions whatsoever. You may copy it, give it away or re-use it
under the terms of the Project Gutenberg License included with this
ebook or online at www.gutenberg.org. If you are not located in the
United States, you will have to check the laws of the country where
you are located before using this eBook.
Title: Kantopään lukuhuoneella

Kansannäytelmä kahdessa näytöksessä
Author: Väinö Kataja
Release date: June 15, 2024 [eBook #73837]
Language: Finnish
Original publication: Hämeenlinna: Boman ja Karlsson, 1901
Credits: Tapio Riikonen
*** START OF THE PROJECT GUTENBERG EBOOK KANTOPÄÄN

LUKUHUONEELLA ***
KANTOPÄÄN LUKUHUONEELLA
Kansannäytelmä kahdessa näytöksessä
Kirj.
VÄINÖ KATAJA
Hämeenlinnassa, Boman ja Karlssonin kustannuksella, 1901.
Henkilöt:
Heikki, Kantopään isäntä. Saara, hänen vaimonsa. Sanna, | Iikka,

| heidän lapsiansa. Eno-Matti, naapuritalon miehiä. Hanhiojan
Salkko, vasta Amerikasta palannut siirtolainen. Olli Kytökangas,
etelästä tullut tukkilainen. Poliisi. Kylännuorisoa.
Tapaus on Perä-Pohjolassa helmikuussa 1900.
Ensimäinen näytös.
Näyttämö: Kantopään kookas pirtti tavallisine talonpoikaisine
huonekaluineen. Perällä ja vasemmalla ovi. Oikealla pitkä penkki,
jonka luona pöytä, jossa on sanomalehtiä sikin sokin. Seinällä näkyy
koukuissaan lehtiä riippumassa. Lähellä vasemmalla olevaa ovea
uuni, jossa pystytuli palaa. Heikki tekee pärekoppaa takkatulen
valossa. Sanna kutoo sukkaa. Iikka selailee sanomalehtiä pöydällä.
—
Sanna: Mitä sinä haet?
Iikka: Katselen sitä Kalevan numeroa, jossa on niistä buuerien

voittamista härjistä. Saisin tuolle Hanhiojan Salkolle näyttää, hän kun
ei näy uskovan, että buuerit voitolla ovat. —
Sanna: Luuletko hänen uskovan, vaikka kultakirjaimilla näkisi

voitot painettuina. Väittelee vaan kömpelösti vastaan ja pitää
engelsmannien puolta, höyhenkenkä.
Iikka: Vaan annapa, että kerran joutuu väittelemään Ollin kanssa,

niin jo vesille joutuisi. Eivätköhän satu yhteen. Ollihan aikoi täksi
illaksi metsästä palata. Väleen hän on täällä. Ja postin tulo-ilta on
myöskin. Aikoiko muita kyläläisiä postiin lähteä kuin Pörhölän
Hanna?
Sanna: Aikoi niitä paljonkin. Meille asti kuitenkin tulevat lehtiä

lukemaan.
Iikka (Yhä hakien): No, mihinkä lempoon se lehti tästä katosi!

Viikon vihollinen kuitenkin.
Heikki: Jos lie Eno-Matti ottanut, kun päivällä kävi. Hän niitä näkyi
siinä selailevan.
Iikka: Eno-Matti sen sitten on ottanutkin.
Heikki: Se on Mattikin niin innostunut noihin lehtiin, että kaikki

pitäisi saada tietoonsa. Vaan parhaiten Matti sotauutiset tietääkin.
Kuului mennä sunnuntaiaamuna ajaneen rovastinkin häpeään, niin
oli tarkka tiedoilleen ollut.
Iikka: Mitä, kun on aikaa päiväkaudet lueskella. Helppohan nyt on

tietoja saada kun näin paljon lehtiä tulee.
Sanna: Mutta ilman Ollin tointa, ei täänkylän nuoriso olisi saanut

lukuhuonetta toimeen. Eikö totta isä ja Iikka?
Heikki: Ollin tointahan se parhaiten oli. Soma mies on muutenkin.

Ja lukumies verraton.
Iikka: Enempi hän tietää kuin muut etelästä tulleet jätkät, mutta
hänhän onkin käynyt kansanopistossa.
Sanna: Eikä ole juoppo niinkuin kaikki muut mieronkulkijat.
Iikka: Se on tosi. Mutta minun tässä täytyy lähteä talliin ennen

postiin lähtöä. (Hakee takin naulasta ja panee ylleen). Elkää vaan
minua jättäkö. (Menee).
Sanna: Kyllä odotamme.
Saara (Ilmestyy vasemmalle ovelle): Sanna tule tänne! (Sanna

menee ovenpieleen. Saara näyttää hänelle puhuvan innoissaan.
Sanna kuuntelee allapäin ja loittonee hetken kuluttua keskilaattialle.
Saara sulkee oven.) Se on päätetty asia!
Heikki: Mikä se nyt oli? Mitä se äiti taas toimittaa?

Sanna: Mitä lie ollut, en tuota oikein kuunnellutkaan. (Keikauttaa
itseään ylpeästi.) Pyh! (Perältä tulee kylän nuorisoa ja asettuvat mikä
minnekin istumaan. Olli tulee jälkimäisenä, laukku seljässä).
Olli: Terve talohon ja terveisiä tukinajosta! (Kättelee Sannaa ja

Heikkiä).
Heikki: Terve vaan. Mitäpä sinne kuuluu?
Olli: Eipä liikoja. Hevoseni jalka loukkaantui, täytyy käydä

apteekissa vähä lääkkeitä.
Sanna: Laske laukkusi pois. Hiihtäenkö tulit?
Olli: Hiihtäen tästä suoraan enkä kauvan taipaleella viipynytkään.

(Laskee laukun penkille. Iikka tulee).
Iikka: Kas sitä! Jopa karttui kansaa. Ja Ollikin. Lähdemme miehissä

postiin.
Sanna: Ja naisissa myöskin.
Iikka: Kaikki. Onpa nyt kuutamo-ilta. Tie on hyvä eikä ole

tulipalopakkanenkaan.
Olli: Tässä postissa minä odotan Helsingistä kirjoja, jotka on

luvattu lähettää tälle lukuhuoneelle.
Sanna: Niinkö? Sellaisia hauskoja kertomuskirjojako?
Olli: Niin luulen. Meidän täytyy yhteisesti vielä koettaa saada

lainakirjasto tänä talvena tähän kylään. Sitten vasta on työmme
alulla. Ja sitten vasta koetamme seuraksi liittyä.
Sanna: Tarkoitatko sellaista nuorisoseuraa kuin kuuluu muualla
olevan melkein joka pitäjäässä?
Olli: Sitähän tarkoitan.
Sanna: Sepä olisi hauskaa.
(Kaikki muut toistavat: Sepä olisi hauskaa).
Iikka: Mutta nyt lähdemme. Ilta kuluu ja posti on jo aikaa ajaa

helistänyt sivu. (Kaikki nousevat lähteäkseen. Sanna ottaa lisää
vaatetta.) Joko ollaan valmiit?
Sanna: Jo minun puolestani.
(Kaikki muut menevät paitsi Olli).
Olli: Jopahan isäntä on alkanut valmistamaan sen mallisia koppia

kun minä neuvoin. (Katselee koppaa).
Heikki: Koetin tässä. Ensimäinen tämä toki on, vaan sievä tämä
malli on.
Olli: Onhan se. (Hiljaa) Onko Hanhiojan Salkko tuolla kamarissa,

koska sieltä kuuluu puhelua?
Heikki: Taitaa olla. Iikka hänet oli tavannut. Miksi sitä kysyt?
Olli: Ilman aikojani. — Olisi minulla muutakin kysyttävää kun

olemme näin kahden kesken.
Heikki: Niinpä kysy sitten.

Olli: Tahdon puhua suoraan ja lyhyesti. Olisiko teillä mitään
vastaan jos pyytäisin tytärtänne Sannaa vaimokseni. Köyhä olen,
mutta ilman muiden apua kyllä vaimostani huolen pidän, jos Jumala
terveyttä antaa.
Heikki: Mikä sinun nyt semmoista pani kysymään yhtäkkiä?
Olli: Sannan kanssa olemme asiasta jo viime kesänä sopineet.
Heikki: Silloinko jo? No mitäpä siitä. Parasta kuitenkin että annatte

tuumanne olla tuonnemmaksi.
Olli: Mutta ette kiellä suostumustanne?
Heikki: Jos niin on Sannalle sallittu, niin mitäpä minä siihen.
Olli: En muuta tahdokaan tällä kertaa tietää. Lopusta kyllä pidän

huolen.
Heikki: Luulen minä, että saat kylläkin toimittaa ennenkun

äitimuorin saat puolellesi. Vaan saanet koettaa.
Olli: Akkaväkeä minä kyllä osaan miellyttää. Eikä emäntäkään niin

paha ole kuin luulette.
Heikki: Entäpä Hanhiojan Salkko? Luulen että hän täällä kulkee

juuri Sannan vuoksi ja äitimuori tuntuu olevan hänen puolellaan.
Olli: Semmoisen puolihupsun kehujan minä kyllä uitan. Muuten

minä tiedän hänen Ameriikka-elämästään yhtä ja toista, josta kyllä
kerran ilmoitan. Eikä hänellä niin paljon rahaa ole kuin kuuluu
kehuvan.
Sanna (Palaa takasin): Joudu nyt Olli!

Olli (Ottaen Sannaa vyötäisiltä): Minä tulen kultakäpyni (He
menevät, Olli laulaen jotakin lempilaulua.)
Heikki (Yksin): Kunnon mieheksi hänet uskon ja kaikkihan häntä

kehuvat vireäksi työmieheksi. Kuuluu tänäkin talvena ansainneen
kovasti tukinajossa. Juomaton ja siivo mies. Mitäpä siitä jos hupsut
häntä etelän varekseksi kutsuvat. Mieluisampi vävy minusta olisi kuin
tuo rokonarpinen ja hassahtava Hanhiojan Salkko. Mutta toista
mieltä taitaa Saara olla.
Saara (Tulee ovelle vasemmalle): Mihinkä ne tästä taas lähtivät?

Iikkakin taas raitilla.
Heikki: Postikonttoriin kuuluivat aikovan. Sinne meni muitakin

kylän nuoria. Nythän on postipäivä tiedämmä. — Saapi tässä taas
kuulla uutisia sodastakin.
Saara: Lupasit noita lehtiä tähän kantaa. Eivät nyt muuta jouda
tekemään kuin niitä kaiket illat lukemaan.
Heikki: Mielelläsipä tuota kuuntelet sinäkin sotauutisia he, he.
Saara: Mihinkäpä minä korvani panen, kun ääneensä lukevat, Eno-

Matti varsinkin. Veihän se täältä yhden lehden päivällä matkassaan.
Heikki: Jos Matti viekin, niin Matti tuopi takaisin. Luulen hänen
pian tulevan, koska tietää postipäivän olevan. (Saara kävelee
laattialla, huomaa Ollin laukun penkillä.) Kenenkä tuo laukku on
tuossa penkillä?
Heikki: Eikö lie sen sinun vieraasi, joka kuuluu kamarissa olevan.
Saara: No Salkko nyt ei noin kelottonutta laukkua viitsi kantaa.
(Tarkastelee laukkua) Mitä tuossa lieneekään, ei se vaan kovin
pullollaankaan ole. Ketä täällä on käynyt? Vastaa.
Heikki: Mistä minä tiedän ketä on käynyt sillä aikaa, kun tallissa
olen ollut.
Saara: Eihän liene vaan sen Suomen-Ollin laukku? Täälläpäinhän

se kuuluu olevan. —
Heikki: Hanhiojan Salkon laukuksi luulen. Mitä sinä sitä käsittelet.

Anna laukun olla, olipa kenen tahansa.
Saara: Heittäisin tämän porrasten päähän, jos Ollin laukuksi

tietäisin.
Heikki: Minkä vuoksi?
Saara: Siksi, etten kärsi sitä miestä.
Heikki: Mutta laukku ei ole mies he, he.
Saara: Mitä meinaat?
Heikki: Mitä minä meinaisin. En mitään.
Saara: Tiedätkö missä tarkotuksessa Salkko täällä on kulkenut.
Heikki: Kai tahtoo lukea sanomalehtiä. Ei hänellä minulle ole ollut

asiaa.
Saara: Mutta sitä enemmän minulle. Salkko kysyy meidän Sannaa

vaimokseen! Etkö riemastu ukko? Hänellä on parikymmentä tuhatta
rahaa, että pystyy hyvän talon ostamaan. Oikein siunailin, että
otettiin noita lehtiä, sillä niiden vuoksi Salkko ensiksi meille oitostui
ja tutustui Sannaan. Mikä ilo meille ja mikä onni Sannalle! Koko
kylän tytöt häntä kadehtivat. Kihlat on Salkolla taskussaan. Etkö
ilosta hyppele ukko? (Tarkastelee yhä laukkua).
Heikki: Mitäpä siitä hyppelisin. Mitä sinä siitä laukusta nypläät,

anna laukun olla penkillä.
Saara (Erikseen): Kyllä tämä taitaa olla Salkon laukku. — Mutta

minä sanon, että tiedätkin, että kohtelet Salkkoa kuin kultaa
kämmenellä. Semmoisia kosijoita harvoin on. Kaksikymmentä
tuhatta! Siihen kultakellot ja arkullinen vaatteita. Mitä arvelet?
Heikki: Arvelen, että paras kun annat asian olla sinänsä. Nuorihan
tuo on Sannakin vielä.
Saara: Minä aavistan jotain, sinä kuhnuri. Sinä suojelet sitä

Suomen Ollia, mokomaa etelän varesta. Mutta varokoon vaan, etten
aja ulos, jos meille vielä tulee Sannan kanssa livertelemään. Ollihan
se tämän huhun meille toimitti…
Heikki: Vastahan sinä sanoit siunaavasi, että lukuhuone meille

saatiin…
Saara: Niin, niin, mutta…
Heikki: Niin, niin, mutta Ollin toimesta se tapahtui. Saisit kiittää

Ollia.
Saara: Elä viisastele! Tiedä, että kohtelet Salkkoa niinkuin vävyä

tulee.
Heikki: Sinulla on aina sinun tuumasi. Olet varmaan itse Sannaa
Salkolle esitellyt, niinkuin mennä talvena Herralan Yrjölle… ja teit
naurun alaiseksi koko talon, hupsu.
Saara (Kiivaasti): Mitä sitten! Kuka sinua minulle toimitti. Oma

äitisi. Koko talo oli velkaan uppoamassa, mutta minun rahoillani on
velka maksettu. Tämä on minun taloni ja minä tässä tyttärenikin
naitan kenelle tahdon.
Heikki: Pidä suus' kiinni ja häpeä.
Saara: Minun tässä valta on. Luuletko, että se olisi ollut sinun
luvattavasi tämä lukuhomma? Elä uskokaan. Se oli juoni, jolla
tahdoin saada Salkon meillä käymään, kun kuulin hänen Amerikasta
palanneen. Ja jahka tässä on kihlat annettu ja asiasta sovittu, niin
ulos pihalle joka lehti. Tässä joka päivä kärsimään kyläläisiä niinkuin
markkina paikalla.
Heikki (Pilkallisesti): Tee tahtosi, sinun on valta!
Saara: Elä sitten murjota. Suomen Ollista, tuosta etelän vareksesta

sinä aina saarnaat. Mikä on Olli Hanhiojan Salkon rinnalla? Tyhjä
säkki.
Heikki: Kas sitä! Jopahan sainkin oikein sievän mallin tähän

koppaan.
Saara: Tyhjä säkki ja alaston mies. Ei muuta tee kuin lukee loruja
ja tyhjää remmaa. Anna, että se alkaisi hyväilemään Sannaa, niin jo
tästä saisi kuumat tuhkat niskaansa. Niin kaksinenkin kosija!
Heikki: Tuli juuri semmoinen koppa kuin kuvassa näyttää, jonka

Olli minulle antoi.
Saara: Ja muista mitä sanoin, että pidät Salkkoa kuin kultaa
kämmenellä. (On aukaissut laukun nuorat ja vetää salaa kaksi poron
juustoa esille, kätkien esiliinansa alle. Poliisi tulee perältä. Erikseen.)
Salkon laukku, aivan varmaan. Hän on jostain kuullut, että minä
tykkään poron juustoista, niin on vasite niitä lappalaisilta
neuvottanut. Se kunnon poika!
Heikki (Pidellen varovasti koppaa) Niinkuin kultaa kämmenellä.
Saara (Menee vasemmalle käsi esiliinan alla): Minun on tässä valta

ja minun tavarani.
Poliisi: Hyvää iltaa.
Heikki: Iltaa. (Erikseen.) Onpa tuon eukon kanssa toisenkin kerran

saanut makeaa mahan täydeltä. Hohoi. (Pyyhkii hikeä.)
Poliisi (Asettaa jalattoman kynttilän pöydän nurkalle palamaan ja

alkaa selailemaan lehtiä): Ei tään päivän postia ole vielä tuotukaan?
Heikki: Noutamaan on menty.
Poliisi: Minulla onkin viime postin lehtiä vähä keskissä.
(Eno-Matti tulee).
Eno-Matti: Iltaa taas. Pakkanen tuntuu kiihtyvän. (Menee takan

luo ja istuu Heikin viereen.) Hauska lehti tämä Kaleva lehtikin on.
(Ottaa lehden taskustaan).
Heikki: Taisi olla sotauutisiakin?
Eno-Matti: Oli kyllä. Sitä ei ole muissa lehdissä ollutkaan.

Heikki: Vai niin. Mikä uutinen se oli. Luehan minullekin. Sinä kun
luet, niin minä ymmärrän oikein hyvin, vaan nuo nuoret ne panee
kuin vellipata, ettei selvää saa. (Kohentelee tulta.)
Eno-Matti (Asettuu likemmäksi tulta lukemaan): Lontoo 20 p. tätä

kuuta. Pretoriasta mainitaan kiivasta taistelua taisteltavan
Kimberleyn luona. Gronje pitää asemansa, Buuerit ovat ottaneet
useita vankeja, vaunuja ja neljätuhatta härkää.
Heikki: Voi lemppari; Niinkuin minun kädestäni! Onpas ne urheita

poikia ne buuerit.
Eno-Matti: Voiton ne ottavat engelsmannilta, saadaan nähdä.
Heikki: Tietenkin. Eihän engelsmannit ole kertaakaan voittaneet.
Eno-Matti: Selkäänsä saaneet joka kerta. Minä olen jokaisen

sotauutisen lukenut.
Heikki: Saivatpa kerran härkiä kelpolailla — vai neljätuhatta. Liekö

hyvinkin isot navetat siellä Transvaalissa?
Eno-Matti: Siellä on kesä jo. Laskevat nurmelle härkänsä syömään

ja siitä penkkiin lyövät, kun liha loppuu.
Heikki: Hehe. Ja engelsmannit saavat kaluta pettua kuin pukit.

Kuka käski mennä härkiä viemään toisen maalle! Kestää sinne
engelsmannin väkeä ja ruokaa kuljettaa. Buuerit ottavat paikalla
pois. On ne poikia.
Eno-Matti: No on ne. Ja niin lempparin tarkkoja ampumaan. Ei

yhtään laukausta hukkaan. Ja viisaita tappelemaan. Muistatko kun
luin siitä vanhasta miehestä, joka kultakuulilla ampui?
Heikki: Sen minä kaiken ikäni muistan. Menneen viikon
lehdissähän siitä olikin.
Eno-Matti: Ja aivan upseereja tähtäilee. Joka päivä kaksi ampuu

kuin metsoa ansaan.
Heikki: Näkyykö muuta erinomaista? (Kohentelee tulta) Näetkö

lukea?
Eno-Matti: Näkee tässä. Vaan ei tässä näy muuta erinomaisempaa

olevankaan. (Hanhiojan Salkko tulee vasemmalta, puettuna
Ameriikkalaisiin vaatteisiin, kaulukseen ja kalosseihin. Lakki päässä.)
Poliisi (Erikseen): Siinäpä mies on, jota etsin.
Salkko: Täällä ukot yhdessä tutkivat sotauutisia, luulen ma.

Buuerit tietenkin alkavat perääntymään.
Eno-Matti: Buueritko? Voittajina ovat niinkuin ennenkin.
Heikki: Neljätuhatta härkää nytkin saaneet.
Salkko: Valheellisia tietoja. Eivät kestä buuerit engelsmannille.

Minä, joka olen kymmenen vuotta ollut engelsmannin parissa, minä
sen tiedän.
(Katselee tuolia istuakseen, käy sitten penkille ja istuu

poliisin kynttilää lähelle.)
Eno-Matti: Vaikka olisit ollut viisikymmentä vuotta. Pakoon on nyt

engelsmannit ajetut, että santa soi.
Salkko: Valetta. Siksi paljon minä olen Englantin maata ajanut,

että tiedän jotakin sanoa.
Eno-Matti: Mutta sinä et ole käynyt sotatantereella, iso vieras.
Salkko: Buuerit ovat yhtä kuin tässä maassa mustilaiset. Mutta

englantilainen on punaveristä ja rikasta kansaa.
Eno-Matti: Vieläkö sinä nytkin pidät englantilaisen puolta.
Salkko: Tämän maan sanomalehdet eivät tiedä sodan oikeasta

menosta. Mutta Amerikan lehdet ne ne tietävät.
Eno-Matti: No sanoppa sitten mitä ne tietävät, tietomies.
Salkko: Sen ne tietävät, että buuerien olisi parasta ajoissa käydä

rukouskirjaan.
Eno-Matti: Sitä he eivät tee.
Salkko: Vaan minä olen siksi paljon lukenut sotahistoriaa ja

näinhän minä Calumethissa kun poikia Kuuban sotaan lähti.
Eno-Matti: Siltipä sinä et mitään tiedäkään, kielimies.
Salkko: Ja sitä paitse ei tässä maassa moni ymmärrä

sanomalehdistä paljon mitään.
(Poliisi silmäilee Salkkoa salaa tarkoin.)
Eno-Matti: Vai niin. Mutta sinä yksin. Tiedätkö mille oma kiitos
haisee, nokikolari.
Salkko: Mitä sinä ukko lisäilet puheesi perään? Sanoppas

kovemmin, että minäkin kuulen.
Eno-Matti: Pistepä tietenkin tässä maassa on lauseen lopussa,
mustilainen.
Salkko: Eikä tämä maa ole kuin isojen valtojen sylkilaatikko; ei

kannata kerskua tämän maan hyvyyksille.
Eno-Matti (Kiivastuen ja nousten seisomaan): Sinäpä taidat olla

toinen Koskela, isänmaan pettäjä, vääryyden puolustaja ja
engelsmannin puolustaja. Tässä maassa… ja tämä maa… Eikö tämä
ole sinunkin isänmaasi, kelvoton!
Heikki: Matti, Matti!
Salkko: Äijä kiivastuu, kun tietää hyvin, etteivät buuerit kuitenkaan

kestä engelsmannille. Aivan sama ja yhtä hullua olisi suomalaisilla
alkaa vastustamaan Venäjän valtaa kuin nyt buuerilla Englantia.
Eno-Matti (Raivoissaan): Oikein arvasin. Laukku-ryssä ja Koskela!

Tiedätkö sinä lurjus mitä sinusta sanottaisiin, jos kirjoittaisin sinusta
sanomalehden toimitukseen!
Salkko (Kiivastuen, nousten penkiltä, telmäten poliisin asettaman

kynttilän sammuksiin laattialle): Minäkö lurjus? Huuti ukko! Olisit nyt
Amerikassa, niin saisit mustat silmät aivan äkkiä. Tuommoinen
suukko ukko!
Eno-Matti: Päällekö tulet. Hyppää hänksesille. Vielä minä yhden

mustilaisen kuritan. (Lyö kämmeniä yhteen.)
Heikki (Koettaen tyynnyttää Mattia): Matti, Matti mitä sinä noin

kiivastut.
Eno-Matti: En kärsi tuota miestä.

Salkko: Sanoitko minua mustilaiseksi? Häh?
Eno-Matti: Luuletko sitten olevasi punaverinen engelsmanni,

pimeämies.
Poliisi: Mikä tämä elämä on? Kuka minun kynttiläni uskaltaa

viskata sammuksiin. Minä en kärsi tätä.
Salkko: Yks' kynttilä. Se ei ole minunlaiselle miehelle iso asia. Minä

ostan tuommoisen miehen tyskyssä. Paljonko tahdot?
Poliisi: Tietääkö vieras, että minä olen kruunun palvelija?
Salkko: Teidän kruununne! (Potkasee kynttilän oven suuhun).
Poliisi: Tästä tulee isompi asia.
Heikki: Helkkarissa nyt alkako riitelemään, täydet miehet.

(Erikseen.) Parhaat enkelit tähän sattuivatkin vastakkain.
Eno-Matti: Olisi minun huoneeni niin ulos ajasin tuon Koskelan

ensiksi.
Salkko (Mennen Matin nokan alle): Tiedätkö sinä ukko, että voin
tämmöisen talon ostaa milloin hyvänsä. Jos tämä olisikin minun
taloni nyt, niin kumpikohan tästä ulos lähtisi minä vai sinä,
nokkaviisas ukonkännyri.
Eno-Matti: Sinulle ei myy taloaan kukaan. Elä luulekaan.
Poliisi: Maksaako vieras sovinnolla minun kynttiläni, vai täytyykö

haastaa käräjiin?
Salkko: Kuka käski sinun panna kynttiläsi aivan pöydän nurkalle!

Poliisi: Kuka käski sinun tulla minun kynttiläni valoon huutamaan.
Eno-Matti: Kas niin. Siinäpä se on. Luuletko, ettei tässä maassa

lakia ole. Hähä?
Saara (Ovella): Mikä meteli täällä on? Onkin Matti täällä

uutisineen. (Viittaa Salkolle) Herra Salkko on hyvä ja tulee tänne.
(Salkko menee vasempaan mutta pukkaa ohimennessään Mattia.
Matti tuivaa vuorostaan, että Salkko menee vauhdissa Saaran syliin).
Eno-Matti: Kas siinä punaverinen engelsmanni!
Heikki: Matti, Matti helkkarissa…
Saara (Ottaen luudan loukosta ja alkaen hätyytellä Mattia):

Paikalla pois! Joka ilta käy täällä tinkaamassa. (Matti pakenee ovelle
päin.) Ulos ja hopusti.
Eno-Matti (Ovella): Tuota rokonarpista mustilaista vähä härnäilin.
Saara (Yhä hätyytellen): Pois! Sinä vanha kälmi.
(Matti katoaa).
Heikki: No tämä on kristillistä elämää.
Saara (Kääntyen poliisiin): Onko poliisi ulosmittauksilla?
Poliisi: Muita asioita on, vaikk'ei tähän taloon. Aijoin samalla

silmätä sanomalehtiä. Onhan täällä mukamas lukuhuone.
Saara: Mutta minä en kärsi kruununmiehiä.
Heikki: Mitä sinä rupeat ihmisiä haukkumaan.

Saara: Pidä suus kiinni siellä, lammas. Että poliisi saapi mennä.
Poliisi: Kyllä, mutta minä toimitan asiani ensin.
Saara: Mitä on asiaa?
Poliisi: Herra Salkko Hanhiojalle. Käskekää hänet tänne.
Saara: Mitä on asiaa?
Poliisi: Sen sanon asianomaiselle. (Aikoo lähteä.) Mutta minä

palaan tänne lautamiehen matkassa. Ei tässä muutoin näy asiat
oikenevan. (Menee ovelle.) Nyt on iltaa vielä toimitella. Hyvästi.
Heikki: Siinä sait. Oletkin valmis jokaiselle suutasi soittamaan.
Saara: Mitä asiaa on hänellä Salkolle? Olisipa kumma tietää.
Heikki: Ehkä saat sen tietää.
Saara: Tiedätkö sinä? Sano heti. Muutoin…
Heikki: Mistä minä tiedän.
Saara: Ei niin mitään. Luuli minua säikyttävänsä. Mutta mitä hittoa

ne viipyvät niin kauvan postikonttorissa, Sannakin?
Heikki: Eiväthän ne vielä kauvaa ole viipyneet.
(Eno-Matti tulee.)
Eno-Matti: Tästä tuli semmoinen lähtö, etten asiaani muistanut.

(Saaralle.) Antaahan emäntä taas anteeksi.
Saara: Selkääsi tarvitsisit, vanha konna.

Eno-Matti: Jo vainkin että luut laulaisivat. (Katselee Ollin laukkua.)
Tässäpä se taitaa olla Ollin laukku. Hän käski tämän täältä ottaa ja
viedä meille. Hyvä laukku.
Saara (Hämillään): Eihän se Ollin laukku ole — tuota — mitä sitä

kenenkään laukusta. Eihän Olli ole täällä käynytkään.
Eno-Matti: Laukkunsa käski ottaa äsken, kun postiin meni ja aikoi

apteekiinkin. (Ottaa laukun) Keveällepä tämä tuntuu.
Saara: Anna laukun olla siinä! Se ei ole Ollin laukku, se on…
Eno-Matti: Onhan tässä Ollin nimikin pikilangalla neulottu.

Katsokaa: Olli Kytökangas. Ollin tämä on, ei auta.
Saara (Heikille): Onko se riivattu käynyt meillä tänä iltana, sano

heti!
Heikki: Käynyt on.
Saara: Miks'et sitä ole heti sanonut?
Heikki: Ethän sinä ole kysynytkään.
(Saara on hämillään).
Eno-Matti: Hyvä porontalja laukku.
(Ottaa laukun selkäänsä ja menee).
Saara (Hätäisesti): Elä vie sitä laukkua. (Erikseen.) Olisiko poliisi

nähnyt, että otin juustot? (Menee kiireesti vasempaan).
Heikki: Mitähän se tuosta laukusta noin? Kummaa oli Saaraksi?
Mikähän asia sillä mahtoi poliisilla olla Salkolle?
(Olli tulee).
Olli: Ennempä ehdin, näemmä. Kävin apteekissa karboolivettä.

(Vetää pullon povestaan.) Onko ketään käynyt lehtiä kyselemässä?
Heikki: Etkö sinä postissa käynytkään?
Olli: En. Hiihdin apteekkiin sillä aikaa kun toiset menivät postiin.
(Hakee laukkuaan penkiltä ja penkin alta).
Heikki: Eno-Matti on laukkusi äsken ottanut. — Sitäkö haet?
Olli: Niin. Mitä Eno-Matti minun laukustani tahtoo?
Heikki: Sanoi sinun käskeneen hänet noutamaan.
Olli: Tapasin kyllä hänet kujalla postiin mennessä vaan laukustani

en virkkanut mitään. Mitähän konnanjuonia sillä ukon kälmillä taas
on mielessään?
(Kylän nuoria palaa postista, eräällä tukko sanomalehtiä

kainalossa).
Ääni joukosta: Antakaa minulle Päivälehti.
Toinen ääni: Minä tahdon Suomettaren.
Sanna (Alkaa sytyttämään kattolamppuun tulta. Olli nostaa erään

pojan kanssa pöydän lampun alle keskelle laattiaa. Sanomalehtikäärö
hajoitetaan ja kukin valitsee lehden. Heikkikin käy pöydän luo).
Olli: Eikö ollut minulle kirjettä?
Sanna: On, on täällä montakin. Odotahan, että saan tämän

lampun reilaan. Pianhan sinä ehditkin. Luulimme sinun olevan
jälessä. —
Olli: Pianhan minä lylylläni lykkelen. (Sanna antaa Ollille kirjeitä.)

Ah! jopa on meidän pitäjän rovastilta ja Helsingistä…
Sanna: Eikö lie lupa niistä kirjoista.
Olli (Pistäen kirjeet taskuun): Mutta yhtä asiaa tahtoisin teille

muistuttaa nuoret ystäväni. — — — — — — — — — Muistattehan
kaikki sen ilman pitempiä selityksiä. (Kaikki laskevat lehtensä ja
kuiskivat toisilleen.) Ehdotan että tapauksen muistoksi yhteisesti
laulaisimme jonkun isänmaallisen laulun.
(Syntyy puhelua nuorten kesken. Esitellään puoli ääneen

yhtä ja toista).
Olli (Alkaa laulaa): Oi Herra siunaa Suomen kansa j.n.e.
Kaikki: (Nousevat seisaalleen ja yhtyvät lauluun. — Saara

ilmaantuu ovelle vasemmalle. Kuuntelee tuokion ja näyttää
miettivältä, poistuu sitten. Esiripun laskiessa kuuluu vielä laulu).
Toinen näytös.
Näyttämö: Sama pirtti samana iltana.

Saara (Yksin kävellen kahakäteen levottomana): Mihin hiiden
säkkiin jouduinkaan noiden kirottujen juustojen takia! Tämä on Eno-
Matin toimia, minä arvaan. Se konna hoksasi tämän. Sillä selvää on,
että poliisi havaitsi minun laukusta jotakin ottavan, kertoi näkemänsä
Matille ja Matti heti tahtoi saattaa minut välikäteen. — Olisikin se
laukku vielä tuossa! Konstikos minun olisi pistää juustot siihen ja
sanoa poliisille, että häpeä valehtelemasta. — Mutta mitä nyt
hoksaisin tehdä? Niin mitä, mitä?
Poliisi (Tulee): Tulen toisen kerran, ajaakseni asiani Salkko

Hanhiojalle. Onko hän vielä talossa?
Saara (Kohteliaasti): Ei ole. Hän läksi pikimältään kylällä käymään.

Palanen aikaa sitten kun meni. Poliisi on hyvä ja istuu. Tässä ovat
pöydällä tämän päivän sanomalehdetkin.
Poliisi: Ei ole aikaa.
(Heikki ja Iikka tulevat sisälle perältä).
Saara: Mitä olisi ollut Salkolle asiaa?
Poliisi: Häntä haastetaan arvannostosta poissa olosta ja taitaa

hänellä olla muitakin syntejä. Ei olisi maksanut vaivaa lähteä
Suomeen miehen, jolla on jos jonkinlaisia haasteita. Ei se niistä
vähillä selviäkään. Luulen, että kukkaro hoikkenee…
Saara: Mutta hän on rikas.
Poliisi: Rikas! Kuka takaa hänen rikkautensa. Ja sitä paitse — näin

talonväen kesken puhuen — kaipasi Olli Kytökangas kahta poron
juustoa, jotka hänen laukustaan olivat kadonneet. Hm. Minä
ajattelen jotakin.
Iikka: Samoin sanoi Olli meillekin vasta. Mutta olisiko se
mahdollista? Isähän sanoo olleensa pirtissä.
Heikki: En minä vaan ole nähnyt.
(Saara peräytyy vasemmalle ja pujahtaa kamariin).
Saara (Ovella): Kai ne on Salkko ottanut.
Poliisi: No niinpä niin. Minä en joudakaan nyt sanomalehtiä

lukemaan. Toivotan vaan hyvää vävyä. Hyvästi! (Menee).
Heikki: Tämähän on nyt harmillista. Kerrassaan. Laukkuun en ole

muiden nähnyt koskevan kun äitisi ja 'Eno-Matin, jompikumpi on
juustot ottanut.
Iikka: Mutta tämähän on hävitöntä, että tässä varkaita on. Mitä

tässä oikeen ajatteleekaan.
Heikki: Eikö lie Eno-Matti taas ollut kujeineen.
Iikka (Kiivaasti): Tästä täytyy tulla selvä, vaikka mätänis.
(Menee ulos).
Heikki (Yksin): Minulle alkaa asia selvenemään. Sen näen jo

Saaran käytöksestä, miten asia on. Saara otti juustot juuri silloin kun
poliisi tuli. Poliisi havaitsi sen, ja kertoi Matille ja Matti nouti laukun
pois saattaakseen Saaran naurettavaan asemaan. Niin luulen
käyneen. (Rupeaa toimittamaan jotakin, ollen selin Saaraan, joka
hiipien tulee vasemmalta ja vie vikkelästi juustot penkille, heittäen
sanomalehden niiden päälle).
Saara: Mitä sinä haet?

Heikki: Mistä sinä tähän ilmausit, vastahan sinä menit kamariin?
Onpa se hullua niiden juustojen… Eihän siitä tiedä Hanhiojan
Salkostakaan…
Saara: On hyvinkin voinut viedä. Miks'ei. Syytön minä olen.
Heikki (Erikseen): Kas kas, kun osaakin pelata. Syytön tietenkin

sinä olet, vaikka persohan sinäkin olet poronjuustolle ha, ha.
Saara: Syö sieniä, ne on tammikuussa pieniä ja sieviä.
Heikki: Olisipa se ikävää, jos tässä vävypoikasi olisi…
Saara: Liekö ollut koko juustoja laukussa. Olli voipi valehdella.
Heikki (Erikseen): Mielimpä vähän koetella eukkoa. (Ääneen.)

Mutta poliisi ei valehtele.
Saara: Poliisi on samanlainen etelän vares kuin Ollikin.
Heikki: Kyllä tästä asiasta selvä tulee. Kumma mies kun kehtaakin
varastaa toisen juustot. Kai hänellä oli tarkoitus ne sinulle lahjoittaa,
kun tietää, että olet perso poronjuustolle. Antoiko Salkko ne sinulle?
Saara: Pidä suus kiinni! Tästäpä vasta pyry nousi.
(Menee kamariin).
Heikki (Huutaen perään): Elä mene. Selvä tästä täytyy tulla.
(Eno-Matti ja Iikka tulevat).
Iikka: Tämä on kumma asia ja saattaa häpeään koko talon. Kun

Salkko palaa otetaan mies lujalle.
Eno-Matti: Tiukalle mies ellei tunnusta ottaneensa.
(Saara ilmaupi ovelle).
Heikki: Juuri niin. Opetetaan näpistelemään toisen laukkua.
Iikka: Hävitöntä ja sikamaista. Ja Olli parka, joka äidille juustot toi

tuliaiseksi. Lappalaisilta oli ostanut.
Heikki: Oliko Olli naapurissa?
Iikka: Menivät naapuri talon isännän kanssa navettaan. Kuului

muuan lehmä olevan kipeänä ja Ollihan on vähäinen eläinlääkäri.
Saara: Pitäisi meidänkin navetassa käydä. "Omenakukka", kun on

herennyt märehtimästä.
Iikka: Miks'ei. Olli kyllä neuvot tietää, kun häntä pyydätte.
Eno-Matti (Selaillen sanomalehtiä): Minulta ne jäivät lehdet

lukematta tältä illalta. Vaan ompa edessäpäin pitkiä pimeitä.
Saara: Eikä päivilläkään kiirettä.
Eno-Matti: Emäntä olisi saanut makeat juustot, mutta kuuluisa

Hanhiojan Salkko taisi…
Saara: Elä leveile ja naureskele, vanha kälmi.
(Olli tulee).
Olli: Ei niistä juustoista maksa vaivaa pitää semmoista mellakkaa.

Ottipa nuo kuka tahansa. Olisin minä mielelläni kuitenkin emännälle
lahjoittanut kun tiesin emännän pitävän semmoisista. Vaan
saanenhan toisella kerralla paremmalla onnella emännän käteen. Ei
niistä isoa vahinkoa tullut, veipä ne kuka tahansa. (Lähestyen
Saaraa.) Muuten minulla on emännälle paljon terveisiä sisareltanne
Kerolan emännältä.
Saara: Kiitoksia niistä. Oletko käynyt Kerolassa?
Olli: Siinähän talossa majailen tänä talvena. Hyvä on talo ja

rauhalliset ihmiset.
(Hanhiojan Salkko tulee, hätäisesti hengittäen.

Saara pujahtaa kamariin).
Salkko: Onko poliisi täällä käynyt? Kuului kylällä hakeneen.
Eno-Matti: On käynyt. Ja lautamieskin sinua on hakenut,

hunsvotti.
Salkko: Taas tuo ukko lurjus on vastuksenani. Missä on emäntä ja

Hanna. Tästä täytyy tulla tosi.
Iikka: Tosi tässä tuleekin. Et pääsekään leikillä. Saat uskoa sen.
Eno-Matti: Juu—juupasenjuu. Jaa—jaapasenjaa. Kuinkahan käypi

engelsmanni?
Salkko: Mitä te murisette kaikin? Tuoko se nyt on se teidän

kuuluisa Suomen Ollinne? (Osottaa Ollia pilkallisesti).
Eno-Matti: Se se on, vaan pysyhän erilläsi, uusi Koskela. (Tehden

itsensä konstikkaan näköiseksi, pilkallisella äänellä-) Juu—
Juupasenjuu, jaa—jaapasenjaa. Oletko käynyt arvannostossa, sinä
Kuuban sotilas?

(FREE PDF Sample) Wireless Sensor Networks 1st Edition Liam I. Farrugia Ebooks

Uploaded by

Copyright:

Available Formats

(FREE PDF Sample) Wireless Sensor Networks 1st Edition Liam I. Farrugia Ebooks

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

(FREE PDF Sample) Wireless Sensor Networks 1st Edition Liam I. Farrugia Ebooks

Uploaded by

Copyright:

Available Formats

Full download ebook at ebookgate.

Wireless Sensor Networks 1st Edition Liam I.

Download more ebook from https://ebookgate.com

Protocols and architectures for wireless sensor

Cooperating Embedded Systems and Wireless Sensor

Security for Wireless Sensor Networks using Identity

Mobile Wireless and Sensor Networks Technology

Synchronization in Wireless Sensor Networks Parameter

Co Operative and Energy Efficient Body Area and

Deploying wireless networks Wilton A.

Technological breakthroughs in modern wireless sensor

WIRELESS SENSOR NETWORKS

Additional books in this series can be found on Nova’s website

Additional E-books in this series can be found on Nova’s website

WIRELESS SENSOR NETWORKS

Nova Science Publishers, Inc.

NOTICE TO THE READER

LIBRARY OF CONGRESS CATALOGING-IN-PUBLICATION DATA

Wireless sensor networks / editors, Liam I. Farrugia.

Published by Nova Science Publishers, Inc. † New York

A wireless sensor network (WSN) consists of spatially distributed autonomous sensors to

humidity, seismic, and position sensors. Computational unit is usually a microcontroller

structure-based category to create and maintain an organization of the sensor nodes.

IMPROVEMENT IN LIFETIME OF WIRELESS SENSOR

G. Sudhir Akojwar*1 and M. Rajendra Patrikar**2

1. WIRELESS SENSOR NETWORKS

1.1. Review and Introduction

* Corresponding author: [email protected], [email protected].

Figure 1. Architecture of a typical Sensor Node.

1.2. Characteristics and Requirements of Wireless Sensor Networks

nodes. It is possible to employ renewable energy resources within the nodes.

9. Scalability: Wireless sensor networks consist of thousands of sensor nodes, while

13. Application-specific: WSN are deployed in a wide-range of scenarios. Hence, they

1.3. Fixed (Non Mobile) Deployment of Sensor Nodes

1.4. Research Approach

1.5. UbiSens: Project at VLSI Lab VNIT Nagpur

microcontroller, a power supply, sensor, ADC, memory and RF transceiver to achieve

1. Central controller: This performs generic computation and is responsible for

Control and Processing Unit

Power Controller Central Controller Compression

Sensors ADC MEMORY TRANSCEIVER

Figure 2. UbiSens: Wireless Sensor Node architecture.

Control and ANN Based Classifier

Sensors ADC MEMORY TRANSCEIVER

Power Source Power Storage

Figure 3.Wireless Sensor Node Architecture with ANN Based classifier.

2. PERFECT DIFFERENCE SET NETWORKS

2.1. Projective Geometry in WSN

The parallel computer architecture needs efficient interconnection techniques. Such

2.2. Fixed Geometrical Deployment of Sensor Nodes

2.3. Perfect Difference Set

Projective geometry gives many mathematical approaches for arranging nodes in a

Links PDS(0,1,3,9) PDS(0,1,4,6)

Property 3. Generation: A PDS of order δ = ph, where p is a prime number, represents a

mod-n differences si+ 1 - si : 1, 2, 6, 4.

2.4. Perfect Difference Networks

Table 1. Formation of Various PDS

 n PDS of order  in normal form

Table 2. Summary of parameters for PDS-Network