Advanced Computing and Systems for Security

Volume Nine Rituparna Chaki download

https://textbookfull.com/product/advanced-computing-and-systems-
for-security-volume-nine-rituparna-chaki/

Download more ebook from https://textbookfull.com

We believe these products will be a great fit for you. Click
the link to download now, or visit textbookfull.com
to discover even more!

Advanced Computing and Systems for Security Volume

Seven Rituparna Chaki

https://textbookfull.com/product/advanced-computing-and-systems-
for-security-volume-seven-rituparna-chaki/

Advanced Computing and Systems for Security Volume Ten

Rituparna Chaki

https://textbookfull.com/product/advanced-computing-and-systems-
for-security-volume-ten-rituparna-chaki/

Advanced Computing and Systems for Security Volume

Eleven Rituparna Chaki

https://textbookfull.com/product/advanced-computing-and-systems-
for-security-volume-eleven-rituparna-chaki/

Advanced Computing and Systems for Security Volume Four

1st Edition Rituparna Chaki

https://textbookfull.com/product/advanced-computing-and-systems-
for-security-volume-four-1st-edition-rituparna-chaki/
Advanced Computing and Systems for Security Volume
Three 1st Edition Rituparna Chaki

https://textbookfull.com/product/advanced-computing-and-systems-
for-security-volume-three-1st-edition-rituparna-chaki/

Advanced Intelligent Systems for Sustainable

Development AI2SD 2018 Volume 5 Advanced Intelligent
Systems for Computing Sciences Mostafa Ezziyyani

https://textbookfull.com/product/advanced-intelligent-systems-
for-sustainable-development-ai2sd-2018-volume-5-advanced-
intelligent-systems-for-computing-sciences-mostafa-ezziyyani/

Advanced Intelligent Systems for Sustainable

Development AI2SD 2019 Volume 6 Advanced Intelligent
Systems for Networks and Systems Mostafa Ezziyyani

https://textbookfull.com/product/advanced-intelligent-systems-
for-sustainable-development-ai2sd-2019-volume-6-advanced-
intelligent-systems-for-networks-and-systems-mostafa-ezziyyani/

Advanced Intelligent Systems for Sustainable

Development AI2SD 2019 Volume 2 Advanced Intelligent
Systems for Sustainable Development Applied to
Agriculture and Health Mostafa Ezziyyani
https://textbookfull.com/product/advanced-intelligent-systems-
for-sustainable-development-ai2sd-2019-volume-2-advanced-
intelligent-systems-for-sustainable-development-applied-to-
agriculture-and-health-mostafa-ezziyyani/

Fog/Edge Computing For Security, Privacy, And

Applications Wei Chang

https://textbookfull.com/product/fog-edge-computing-for-security-
privacy-and-applications-wei-chang/
Advances in Intelligent Systems and Computing 995

Rituparna Chaki
Agostino Cortesi
Khalid Saeed
Nabendu Chaki Editors

Advanced
Computing
and Systems
for Security
Volume Nine
Advances in Intelligent Systems and Computing

Volume 995

Series Editor
Janusz Kacprzyk, Systems Research Institute, Polish Academy of Sciences,
Warsaw, Poland

Advisory Editors
Nikhil R. Pal, Indian Statistical Institute, Kolkata, India
Rafael Bello Perez, Faculty of Mathematics, Physics and Computing,
Universidad Central de Las Villas, Santa Clara, Cuba
Emilio S. Corchado, University of Salamanca, Salamanca, Spain
Hani Hagras, School of Computer Science & Electronic Engineering,
University of Essex, Colchester, UK
László T. Kóczy, Department of Automation, Széchenyi István University,
Gyor, Hungary
Vladik Kreinovich, Department of Computer Science, University of Texas
at El Paso, El Paso, TX, USA
Chin-Teng Lin, Department of Electrical Engineering, National Chiao
Tung University, Hsinchu, Taiwan
Jie Lu, Faculty of Engineering and Information Technology,
University of Technology Sydney, Sydney, NSW, Australia
Patricia Melin, Graduate Program of Computer Science, Tijuana Institute
of Technology, Tijuana, Mexico
Nadia Nedjah, Department of Electronics Engineering, University of Rio de Janeiro,
Rio de Janeiro, Brazil
Ngoc Thanh Nguyen, Faculty of Computer Science and Management,
Wrocław University of Technology, Wrocław, Poland
Jun Wang, Department of Mechanical and Automation Engineering,
The Chinese University of Hong Kong, Shatin, Hong Kong
The series “Advances in Intelligent Systems and Computing” contains publications
on theory, applications, and design methods of Intelligent Systems and Intelligent
Computing. Virtually all disciplines such as engineering, natural sciences, computer
and information science, ICT, economics, business, e-commerce, environment,
healthcare, life science are covered. The list of topics spans all the areas of modern
intelligent systems and computing such as: computational intelligence, soft comput-
ing including neural networks, fuzzy systems, evolutionary computing and the fusion
of these paradigms, social intelligence, ambient intelligence, computational neuro-
science, artificial life, virtual worlds and society, cognitive science and systems,
Perception and Vision, DNA and immune based systems, self-organizing and
adaptive systems, e-Learning and teaching, human-centered and human-centric
computing, recommender systems, intelligent control, robotics and mechatronics
including human-machine teaming, knowledge-based paradigms, learning para-
digms, machine ethics, intelligent data analysis, knowledge management, intelligent
agents, intelligent decision making and support, intelligent network security, trust
management, interactive entertainment, Web intelligence and multimedia.
The publications within “Advances in Intelligent Systems and Computing” are
primarily proceedings of important conferences, symposia and congresses. They
cover significant recent developments in the field, both of a foundational and
applicable character. An important characteristic feature of the series is the short
publication time and world-wide distribution. This permits a rapid and broad
dissemination of research results.
** Indexing: The books of this series are submitted to ISI Proceedings,
EI-Compendex, DBLP, SCOPUS, Google Scholar and Springerlink **

More information about this series at http://www.springer.com/series/11156

Rituparna Chaki Agostino Cortesi
• •

Khalid Saeed Nabendu Chaki

•

Editors

Advanced Computing
and Systems for Security
Volume Nine

123
Editors
Rituparna Chaki Agostino Cortesi
A. K. Choudhury School Full Professor of Computer Science
of Information Technology DAIS—Università Ca’ Foscari
University of Calcutta Venice, Venezia, Italy
Kolkata, West Bengal, India
Nabendu Chaki
Khalid Saeed Department of Computer Science
Faculty of Computer Science and Engineering
Bialystok University of Technology University of Calcutta
Bialystok, Poland Kolkata, West Bengal, India

ISSN 2194-5357 ISSN 2194-5365 (electronic)

Advances in Intelligent Systems and Computing
ISBN 978-981-13-8961-0 ISBN 978-981-13-8962-7 (eBook)
https://doi.org/10.1007/978-981-13-8962-7
© Springer Nature Singapore Pte Ltd. 2020
This work is subject to copyright. All rights are reserved by the Publisher, whether the whole or part
of the material is concerned, specifically the rights of translation, reprinting, reuse of illustrations,
recitation, broadcasting, reproduction on microfilms or in any other physical way, and transmission
or information storage and retrieval, electronic adaptation, computer software, or by similar or dissimilar
methodology now known or hereafter developed.
The use of general descriptive names, registered names, trademarks, service marks, etc. in this
publication does not imply, even in the absence of a specific statement, that such names are exempt from
the relevant protective laws and regulations and therefore free for general use.
The publisher, the authors and the editors are safe to assume that the advice and information in this
book are believed to be true and accurate at the date of publication. Neither the publisher nor the
authors or the editors give a warranty, expressed or implied, with respect to the material contained
herein or for any errors or omissions that may have been made. The publisher remains neutral with regard
to jurisdictional claims in published maps and institutional affiliations.

This Springer imprint is published by the registered company Springer Nature Singapore Pte Ltd.
The registered company address is: 152 Beach Road, #21-01/04 Gateway East, Singapore 189721,
Singapore
Preface

This volume contains the revised and improved version of papers presented at the
6th International Doctoral Symposium on Applied Computation and Security
Systems (ACSS 2019) which took place in Kolkata, India, during 12–13 March
2019. The University of Calcutta in collaboration with Ca’ Foscari University of
Venice, Italy, and Bialystok University of Technology, Poland, organized the
symposium. This symposium is unique in its characteristic of providing Ph.D.
scholars an opportunity to share the preliminary results of their work in an inter-
national context and be actively supported towards their first publication in a sci-
entific volume.
In our pursuit of continuous excellence, we aim to include the emergent research
domains in the scope of the symposium each year. This helps ACSS to stay in tune
with the evolving research trends. The sixth year of the symposium was marked
with a significant improvement in overall quality of papers, besides some very
interesting papers in the domain of security and software engineering. We are
grateful to the Programme Committee members for sharing their expertise and
taking time off from their busy schedule to complete the review of the papers with
utmost sincerity. The reviewers have pointed out the improvement areas for each
paper they reviewed, and we believe that these suggestions would go a long way in
improving the overall quality of research among the scholars. We have invited
eminent researchers from academia and industry to chair the sessions which mat-
ched their research interests. As in previous years, the session chairs for each
session had a prior go-through of each paper to be presented during the respective
sessions. This is done to make it more interesting as we found deep involvement
of the session chairs in mentoring the young scholars during their presentations.
The evolution of ACSS is an interesting process. We have noticed the emer-
gence of security as a very important aspect of research, due to the overwhelming
presence of IoT in every aspect of life.
The indexing initiatives from Springer have drawn a large number of
high-quality submissions from scholars in India and abroad. ACSS continues with
the tradition of the double-blind review process by the PC members and by external
reviewers. The reviewers mainly considered the technical aspect and novelty of

v
vi Preface

each paper, besides the validation of each work. This being a doctoral symposium,
the clarity of the presentation was also given importance.
The Technical Programme Committee for the symposium selected only 18
papers for publication out of 42 submissions.
We would like to take this opportunity to thank all the members of the
Programme Committee and the external reviewers for their excellent and
time-bound review works.
We thank the members of the Organizing Committee, whose sincere efforts
before and during the symposium have resulted in a friendly and engaging event,
where the discussions and suggestions during and after the paper presentations
create a sense of community that is so important for supporting the growth of young
researchers.
We thank Springer for sponsoring the Best Paper Award. We would also like to
thank ACM for the continuous support towards the success of the symposium. We
appreciate the initiative and support from Mr. Aninda Bose and his colleagues in
Springer Nature for their strong support towards publishing this post-symposium
book in the series “Advances in Intelligent Systems and Computing”. Last but not
least, we thank all the authors without whom the symposium would not have
reached up to this standard.
On behalf of the editorial team of ACSS 2019, we sincerely hope that ACSS
2019 and the works discussed in the symposium will be beneficial to all its readers
and motivate them towards even better works.

Kolkata, India Nabendu Chaki

Bialystok, Poland Khalid Saeed
Kolkata, India Rituparna Chaki
Venice, Italy Agostino Cortesi
Contents

WSN and IoT Applications

Fuzzy Logic-Based Range-Free Localization for Wireless Sensor
Networks in Agriculture . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3
Arindam Giri, Subrata Dutta and Sarmistha Neogy
End-User Position-Driven Small Base Station Placement
for Indoor Communication . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13
Anindita Kundu, Shaunak Mukherjee, Ashmi Banerjee
and Subhashis Majumder
ZoBe: Zone-Oriented Bandwidth Estimator for Efficient IoT
Networks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27
Raghunath Maji, Souvick Das and Rituparna Chaki

Software Engineering and Formal Specification

for Secured Software Systems
Extracting Business Compliant Finite State Models
from I Models . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 39
Novarun Deb, Nabendu Chaki, Mandira Roy, Surochita Pal
and Ankita Bhaumick
Behavioral Analysis of Service Composition Patterns in ECBS
Using Petri-Net-Based Approach . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 53
Gitosree Khan, Anirban Sarkar and Sabnam Sengupta

VLSI and Graph Algorithms

Generation of Simple, Connected, Non-isomorphic
Random Graphs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 69
Maumita Chakraborty, Sumon Chowdhury and Rajat Kumar Pal

vii
viii Contents

Bottleneck Crosstalk Minimization in Three-Layer

Channel Routing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 79
Tarak Nath Mandal, Kaushik Dey, Ankita Dutta Banik, Ranjan Mehera
and Rajat Kumar Pal
Arithmetic Circuits Using Reversible Logic: A Survey Report . . . . . . . . 99
Arindam Banerjee and Debesh Kumar Das
Author Index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 111
About the Editors

Rituparna Chaki is Professor of Information Technology in the University of

Calcutta, India. She received her Ph.D. Degree from Jadavpur University in India in
2003. Before this she completed B.Tech. and M.Tech. in Computer Science &
Engineering from the University of Calcutta in 1995 and 1997 respectively. She has
served as a System Executive in the Ministry of Steel, Government of India for nine
years, before joining the academics in 2005 as a Reader of Computer Science &
Engineering in the West Bengal University of Technology, India. She is with the
University of Calcutta since 2013. Her area of research includes Optical networks,
Sensor networks, Mobile ad hoc networks, Internet of Things, Data Mining, etc.
She has nearly 100 publications to her credit. Dr. Chaki has also served in the
program committees of different international conferences. She has been a regular
Visiting Professor in the AGH University of Science & Technology, Poland for the
last few years. Dr. Chaki has co-authored a couple of books published by CRC
Press, USA.

Agostino Cortesi, Ph.D., is a Full Professor of Computer Science at Ca’ Foscari

University, Venice, Italy. He served as Dean of the Computer Science studies, as
Department Chair, and as Vice-Rector for quality assessment and institutional
affairs. His main research interests concern programming languages theory, soft-
ware engineering, and static analysis techniques, with particular emphasis on
security applications. He published more than 110 papers in high-level international
journals and proceedings of international conferences. His h-index is 16 according
to Scopus, and 24 according to Google Scholar. Agostino served several times as
member (or chair) of program committees of international conferences (e.g., SAS,
VMCAI, CSF, CISIM, ACM SAC) and he is in the editorial boards of the journals
“Computer Languages, Systems and Structures” and “Journal of Universal
Computer Science”. Currently, he holds the chairs of “Software Engineering”,
“Program Analysis and Verification”, “Computer Networks and Information
Systems” and “Data Programming”.

ix
x About the Editors

Khalid Saeed is a Full Professor in the Faculty of Computer Science, Bialystok

University of Technology, Bialystok, Poland. He received the B.Sc. Degree in
Electrical and Electronics Engineering in 1976 from Baghdad University in 1976,
the M.Sc. and Ph.D. Degrees from Wroclaw University of Technology, in Poland in
1978 and 1981, respectively. He received his D.Sc. Degree (Habilitation) in
Computer Science from the Polish Academy of Sciences in Warsaw in 2007. He
was a Visiting Professor of Computer Science with the Bialystok University of
Technology, where he is now working as a Full Professor He was with AGH
University of Science and Technology in the years 2008–2014. He is also working
as a Professor with the Faculty of Mathematics and Information Sciences at Warsaw
University of Technology. His areas of interest are Biometrics, Image Analysis and
Processing and Computer Information Systems. He has published more than 220
publications, edited 28 books, Journals and Conference Proceedings, 10 text and
reference books. He supervised more than 130 M.Sc. and 16 Ph.D. theses. He gave
more than 40 invited lectures and keynotes in different conferences and universities
in Europe, China, India, South Korea and Japan on Biometrics, Image Analysis and
Processing. He received more than 20 academic awards. Khalid Saeed is a member
of more than 20 editorial boards of international journals and conferences. He is an
IEEE Senior Member and has been selected as IEEE Distinguished Speaker for
2011–2016. Khalid Saeed is the Editor-in-Chief of International Journal of
Biometrics with Inderscience Publishers.

Nabendu Chaki is a Professor in the Department of Computer Science &

Engineering, University of Calcutta, Kolkata, India. Dr. Chaki did his first gradu-
ation in Physics from the legendary Presidency College in Kolkata and then in
Computer Science & Engineering from the University of Calcutta. He has com-
pleted Ph.D. in 2000 from Jadavpur University, India. He is sharing 6 international
patents including 4 U.S. patents with his students. Prof. Chaki has been quite active
in developing international standards for Software Engineering and Cloud
Computing as a member of Global Directory (GD) for ISO-IEC. Besides editing
more than 25 book volumes, Nabendu has authored 6 text and research books and
has more than 150 Scopus Indexed research papers in Journals and International
conferences. His areas of research interests include distributed systems, image
processing and software engineering. Dr. Chaki has served as a Research Faculty in
the Ph.D. program in Software Engineering in U.S. Naval Postgraduate School,
Monterey, CA. He is a visiting faculty member for many Universities in India and
abroad. Besides being in the editorial board for several international journals, he has
also served in the committees of over 50 international conferences. Prof. Chaki is
the founder Chair of ACM Professional Chapter in Kolkata.
 Part I
WSN and IoT Applications
Fuzzy Logic-Based Range-Free
Localization for Wireless Sensor
Networks in Agriculture

Arindam Giri, Subrata Dutta and Sarmistha Neogy

Abstract Knowing the accurate location of nodes in wireless sensor networks

(WSN) is a major concern in many location-dependent applications. Especially in
agricultural monitoring, accurate location of the affected crop is required for effective
use of fertilizers or pesticides. Low-cost and accurate localization algorithms need to
be developed for such applications. Range-free localization like DV-Hop has become
popular over the years due to its simplicity and that too without complicated hard-
ware. This paper proposes a modified DV-Hop algorithm that improves the accuracy
of localization, based on weights of anchors. In this algorithm, fuzzy logic is used
to determine the weights of anchors. Simulation results also show that the proposed
algorithm improves localization accuracy with respect to DV-Hop algorithm and its
variants.

Keywords Sensor network · Localization algorithms · DV-Hop · Fuzzy logic

1 Introduction

Rapid technological advances of low-cost, low-power, and multifunctional sensor

devices make them used in every area of life. Generally, deployed wireless sensor
networks (WSNs) are used to gather spatiotemporal characteristics of the physi-
cal world. They are being successfully used in environmental monitoring, preci-
sion agriculture, patient monitoring, habitat monitoring, and object tracking [1].
For instance, WSN used in agriculture [2] can monitor field characteristics like

A. Giri (B)
Haldia Institute of Technology, Haldia 721657, West Bengal, India
e-mail: [email protected]
S. Dutta
National Institute of Technology, Jamshedpur 831014, Jharkhand, India
e-mail: [email protected]
S. Neogy
Jadavpur University, Kolkata 700032, West Bengal, India
e-mail: [email protected]

© Springer Nature Singapore Pte Ltd. 2020 3

R. Chaki et al. (eds.), Advanced Computing and Systems for Security,
Advances in Intelligent Systems and Computing 995,
https://doi.org/10.1007/978-981-13-8962-7_1
4 A. Giri et al.

temperature, moisture, nutrients, etc. Location-dependent applications like object

tracking need to know the physical location of sensors. Determining the physical
position of sensors in WSN is known as localization. Localization plays a major
role in such WSN applications. Researchers have contributed much in the field of
localization [3]. However, it remains a challenging task due to the limited capacity
of tiny sensors like energy and computing power.
Although global positioning system (GPS) provides position information to sen-
sors, it is not feasible to be incorporated in every sensor in terms of cost, hardware,
and computational capacity requirements of GPS devices. Again, GPS is not suitable
in hazardous environment like industrial plant monitoring as well in indoor applica-
tions like habitat monitoring. Generally, in localization algorithms, only a few nodes
are enabled with GPS to assess their positions. These nodes are called anchor nodes.
The problem of localization is to estimate the locations of unknown nodes in WSN
with the help of anchor nodes.
There are many localization algorithms proposed in WSN. The existing local-
ization algorithms can be classified as range-based and range-free. Range-based
algorithms estimate locations of sensors based on range measurements like distance
[4] or angle [5]. Although range-based localization technique provides accurate loca-
tion information, they require additional costly hardware for range measurements.
Range-free algorithms do not rely on absolute range information, and hence not
require additional hardware [6, 7]. Range-free localization algorithms are widely
used in large-scale WSN applications like agriculture to avail economic benefit over
the other. DV-Hop [8] is a popular range-free localization algorithm. It assumes that
anchors closer to an unknown node can provide more accurate hop count than a
farther one. However, two anchors closely placed in a randomly distributed WSN
consider others in calculating hop count between them. So, a closest anchor not
always offers better hop count estimation than far away anchors with accurate hop
count. Efforts have been made to DV-Hop algorithm to enhance the accuracy of
localization. In weighted DV-Hop, weights of anchors are used to enhance localiza-
tion accuracy without additional hardware device. The weight of an anchor node is
calculated based on minimal hop count between an unknown node and an anchor
node. In all DV-Hop variants, weights are assigned to anchors such that the more the
hop count, the less is the weight of an anchor. However, this fact is not always true;
an anchor farther from an unknown node may give better estimation of hop count
possessing more weights than a closely placed anchor. Inaccurate hop count results
in inaccurate anchor weights. Hence, deciding weights of anchors with erroneous
hop count is a challenge in range-free localization. Uncertainty in deciding weights
of anchors can introduce error in localization.
In this paper, we propose a fuzzy-based weighted DV-Hop (called, FWDV-Hop)
algorithm where localization accuracy is enhanced dealing with uncertainty in anchor
weights. Here, weights are estimated based on erroneous hop count through fuzzy
logic [9].
The remainder of this paper is organized as follows. In Sect. 2, DV-Hop algorithm
is illustrated along with its variants. We present an overview of the proposed FWDV-
Hop algorithm along with an introduction to fuzzy logic in Sect. 3. Simulation of the
Fuzzy Logic-Based Range-Free Localization … 5

proposed algorithm and result analysis are presented in Sect. 4, while the conclusion
is drawn in Sect. 5.

2 Literature Review

2.1 DV-Hop Algorithm

In this section, DH-Hop algorithm is explained. It is a classical localization algorithm

used in WSN which uses a few anchor nodes. The positions of unknown nodes are
estimated after getting position information from nearby anchor nodes. DV-Hop is
based on the concept of distance vector routing. It works in three phases. In the
first phase, all nodes in the network get distances to anchor in hops by exchanging
distance vector. In the second phase, each node calculates the average hop size. Then
unknown nodes estimate their positions with the help of known anchor positions in
the last phase.
In the first phase, each anchor node broadcasts a beacon message providing its
position information along with an initial hop count of one. Each receiving node
maintains a minimum hop count per anchor for all beacons. Beacons with higher hop
count values (called, stale information) from anchors are simply discarded. Non-stale
beacons are flooded outward after incrementing hop count at every intermediate hop
until all the shortest paths are found. This is how all nodes including the anchors
know the shortest path distance to all anchor nodes in terms of minimum hop count
information.
In the second phase, each anchor node calculates its average hop size of one hop
using the minimum hop count to other anchors. The average hop size of ith anchor
is calculated as
 
j=i (xi − x j )2 + (yi − y j )2
H opsi zei =  (1)
j=i h i j

 
where (xi , yi ), x j , y j are the coordinates of anchors i and j and h i j denotes the
minimum hop count between them. After calculating average hop size, an anchor
node broadcasts that value throughout the network. Once an unknown node receives
hop size, it finds the distance to the beacon node using hop count and hop size using
Eq. (2).

di j = H opsi zei ∗ h i j . (2)

In the last step, an unknown node, P estimates its position, X = (x, y) using
position of n number of anchor nodes and the distance, di , i = 1, 2, ..n, to P as
follows:
6 A. Giri et al.
⎧
⎪
⎪ (x − x1 ) + (y − y1 ) = d1
2 2 2
⎪
⎪
⎪
⎨ (x − x2 ) + (y − y2 ) = d22
2 2

. . (3)
⎪
⎪
⎪
⎪ .
⎪
⎩ (x − x )2 + (y − y )2 = d 2
n n n

Now, Eq. (3) can be expanded as

⎧
⎪
⎪ 2(x1 − xn )x + 2(y1 − yn )y = x12 − xn2 + y12 − yn2 − d12 + dn2
⎪
⎪
⎪
⎨ 2(x2 − xn )x + 2(y2 − yn )y = x22 − xn2 + y22 − yn2 − d22 + dn2
· . (4)
⎪
⎪
⎪
⎪ ·
⎪
⎩ 2(x
n−1 − x n )x + 2(yn−1 − yn )y = x n−1 − x n + yn−1 − yn − dn−1 + dn
2 2 2 2 2 2

Now, Eq. (4) can be represented by linear system as AX = B, where

⎡ ⎤
x1 − xn y1 − yn
⎢ x −x y −y ⎥
⎢ 2 n 2 n ⎥
⎢ ⎥
A = −2 × ⎢ . ⎥,
⎢ ⎥
⎣ . ⎦
xn−1 − xn yn−1 − yn
⎡ ⎤
d12 − dn2 − x12 + xn2 − y12 + yn2
⎢ d 2 − d 2 − x 2 + x 2 − y2 + y2 ⎥
⎢ ⎥
⎢ 2 n 2 n 2 n
⎥
B=⎢ . ⎥, and
⎢ ⎥
⎣ . ⎦
dn−1 − dn − xn−1 + xn − yn−1 + yn
2 2 2 2 2 2
 
x
X=
y

Hence, X can be calculated as

 −1
X = A T A A T B. (5)

The limitation of DV-Hop algorithm is that it assumes that the anchor node closest
to unknown node can contribute accurate hop size estimation than others. However,
in a randomly deployed WSN, anchor nodes situated very close to each other add
up nearby anchors in their hop count calculation. In this situation, an anchor closest
to a sensor node does not necessarily give accurate hop size estimation. Instead, an
anchor far away from an unknown node with appropriate hop size may offer better
estimation of coordinates. Hence, location estimation of WSN nodes is affected by
this approach. To get rid of the problem, the anchors are assigned weights reflecting
their impact on location estimation for an unknown node.
Fuzzy Logic-Based Range-Free Localization … 7

2.2 Variants of DV-Hop Algorithm

In weighted DV-Hop algorithm [10], the weight of an anchor is calculated based on

the hop count as below:

1
wi = . (6)
hi j

Here, wi ’s might not be accurately calculated if hop count is inaccurate. Unlike

DV-Hop, the weighted DV-Hop algorithm uses weights of anchors in a least square
method in Eq. (7) as

 
n
 2  2
f (x, y) = min wi2 ( xi − x j + yi − y j − di )2 , (7)
i=1

where wi is the weight of anchor node i and n is number of nodes. In weighted DV-
Hop, the value of wi is set using Eq. (6). Now, Eq. (7) can be expressed as A X = B  ,
where
⎡ ⎤
w12 wn2 (x1 − xn ) w12 wn2 (y1 − yn )1
⎢ w2 w2 (x − x ) w22 wn2 (y2 − yn ) ⎥
⎢ 2 n 2 n ⎥
 ⎢ ⎥
A = −2 × ⎢ · · ⎥, and
⎢ ⎥
⎣ · · ⎦
2
wn−1 wn2 (xn−1 − xn ) wn−1
2
wn2 (yn−1 − yn )
⎡   ⎤
w12 wn2 d12 − dn2 − x12 + xn2 − y12 + yn2
⎢ w22 wn2 (d22 − dn2 − x22 + xn2 − y22 + yn2 ) ⎥
⎢ ⎥
 ⎢ ⎥
B =⎢ . ⎥
⎢ ⎥
⎣ . ⎦
2
wn−1 wn2 (dn−1
2
− dn2 − xn−1
2
+ xn2 − yn−12
+ yn2 )

Then, the position of unknown node is estimated as

  −1 
X = A T A A T B. (8)

Improved weighted DV-Hop (IWDV-Hop), proposed by Guadane et al. [10], is

an improved version of DV-Hop algorithm which uses inversely weighted hop count
in location estimation. It considers not only a closer anchor to estimate unknown
location but also it uses a far away anchor with accurate hop count.
1
hi j
wi = n 1
, k = j. (9)
k=1 h k j
8 A. Giri et al.

The root mean square error (RMSE) in location estimation of a node i is calculated
in Eq. (10), assuming (x, y) and (xi , yi ) as the actual and estimated coordinates of
i, respectively.

1 
n
e= (x − xi )2 + (y − yi )2 . (10)
n − 1 i=1

In [11], each anchor node calculates its own hop size as the weighted sum of
all average one-hop distances (HWDV-Hop) to all anchors. The proposed algorithm
estimates the average hop size for the network by average hop size weighted mean.
HWDV-Hop demands that it outperforms DV-Hop in terms of localization accuracy.

3 Overview of FWDV-Hop Algorithm

This section gives an overview of fuzzy-based weighted DV-Hop (FWDV-Hop) algo-

rithm. In DV-Hop, an unknown node estimates its location using least square method
as in Eq. (3). Weights are assigned to anchors to impose the relative importance
over other anchors. Weights calculated directly based on inaccurate/noisy hop count
provides inaccurate average hop size which decreases the accuracy of localization.
In the proposed FWDV-Hop, we assign weights of anchors using fuzzy logic. This
approach can provide appropriate weight assignment so as to localize nodes dealing
with uncertainty in hop count. This algorithm follows DV-Hop except the weight
calculation for anchors through fuzzy inference system. The fuzzy input hop count
is feed to FIS in Fig. 1 and mapped to weight via inference engine using fuzzy rules.
The membership function of fuzzy variables hop count and weight is divided into
five triangular membership functions such as very low (VL), low (L), moderate (M),
high (H), and very high (VH) as shown in Fig. 2.

Fig. 1 Fuzzy model for proposed algorithm with hop count as input and weight as output
Fuzzy Logic-Based Range-Free Localization … 9

Fig. 2 Membership functions for input/output variable

3.1 Introduction to Fuzzy Logic

Fuzzy logic [9], proposed by Zadeh, is mainly used in linguistic reasoning. It resem-
bles the logic of human thought in a flexible computation than computers. It demands
to be used in application environments dealing with imprecise, vague, and erroneous
information. Fuzzy logic can better represent the vague data like “high” with fuzzy
set and fuzzy membership function. Fuzzy sets can prescribe a range of values,
called domain characterized by the membership function. A fuzzy logic system,
called fuzzy inference system (FIS), consists of three components: fuzzifier, infer-
ence engine, and defuzzifier. A fuzzifier maps each crisp input to a fuzzy set and
assigns a value equal to the degree of membership. Inference engine is built upon
IF-ELSE rules and inference methods. Given a number of fuzzy inputs, the inference
engine calculates fuzzy output based on fired rule. Finally, the defuzzifier maps the
fuzzy output to crisp output. The fuzzy inference system is depicted in Fig. 3.

R1: If x is small then y is low

R2: If x is moderate then y is high
R3: If x is high then y is high
Crisp inputs Crisp output
x y
Fuzzifier Defuzzifier

Inference Engine

Fig. 3 A fuzzy inference system showing three components

10 A. Giri et al.

4 Simulation Results and Analysis

For implementing our algorithm, we have used MATLAB. A wireless sensor network
is created with randomly deployed sensors and anchors over an area of 50 × 50 m2 .
All nodes are assumed to have 2 J initial energy and 15 m of communication radius.
The number of nodes is varied from 100 to 500, while radio communication radius of
each sensor is changed from 10 to 20 m. We plot the localization error with varying
anchor nodes to sensor node ratio from 5 to 40%. In the same MATLAB environment,
other algorithms are evaluated. Performance of the proposed algorithm is measured
through normalized localization error (RMSE) using Eq. (10) as the average deviation
of estimated location to actual location of all unknown nodes. The simulation results
are used to plot localization error varying number of nodes, communication radius,
and ration of anchor nodes. Proposed algorithm is compared with other algorithms
in Fig. 4 through Fig. 6.
Figure 4 depicts localization error with respect to total number of sensor nodes,
with 10% of anchor nodes. It is observed that localization error decreases as number
of increases. The error becomes steady when the number of sensors increases to
300 and more. With the same number of nodes and 10% of anchors, the proposed
algorithm outperforms others as the weights are fixed with fuzzy logic.
In Fig. 5, localization error is plotted with varying anchor nodes from 5 to 40%
of total nodes in the network. The more is the anchor ratio, the less is the local-
ization error. Dealing uncertainties through fuzzy logic, our algorithm gives better
localization accuracy than others.
In Fig. 6, localization error is depicted with respect to communication radius of
sensors varying from 10 to 20 m. Here, 100 nodes are randomly deployed with 20%
of anchor nodes in order to evaluate algorithms. With increasing radius of sensors,
localization accuracy is improved. Again, proposed algorithm outperforms others in
terms of localization accuracy due to appropriate fixing of anchor weights in location
estimation.

0.5
Normalized RMSE

0.4
0.3
DV-Hop
0.2
0.1 HWDV-
Hop
0
100 150 200 250 300 350 400 450 500
Number of nodes

Fig. 4 Normalized localization error versus number of sensor nodes, with 10% of anchor nodes
Fuzzy Logic-Based Range-Free Localization … 11

0.7
0.6

Normalized RMSE
0.5 DV-Hop
0.4
HWDV-Hop
0.3
0.2 IWDV-Hop
0.1 FWDV-Hop
0
5 10 15 20 25 30 35 40
Anchor node %

Fig. 5 Normalized localization error versus anchor nodes ratio (in percentage of anchor nodes)
with 100 sensor nodes

0.6
Normalized RMSE

0.5
0.4 DV-Hop
0.3 HWDV-Hop
0.2
IWDV-Hop
0.1
FWDV-Hop
0
10 11 12 13 14 15 16 17 18 19 20
Communication radius(m)

Fig. 6 Normalized localization error versus radio range of sensors, with 100 sensors and 10% of
anchor nodes

5 Conclusion

Localization is essential to location-dependent applications like agriculture monitor-

ing. Range-free localization becomes popular because of its no additional hardware
requirements. In this paper, a range-free localization method is presented using fuzzy
logic. In weighted DV-Hop algorithm, weights are calculated as the inverse of hop
count. With closely placed anchor nodes in randomly distributed sensor network,
hop count estimation is not correct, so weights are affected. In this research, fuzzy
logic is utilized to find weights of anchors based on erroneous hop count. Local-
ization error is calculated for algorithms with varying number of sensors, anchors,
and communication radius of sensors. Simulation results reveal that the proposed
algorithm improves others in terms of localization accuracy.
12 A. Giri et al.

References

1. Akyildiz, I.F., Su, W., Sankarasubramaniam, Y., Cayirci, E.: Wireless sensor networks: a survey.
Comput. Netw. 38(4), 393–422 (2002)
2. Giri, A., Dutta, S., Neogy, S., Dahal, K., Pervez, Z.: Internet of things (IoT): a survey on
architecture, enabling technologies, applications and challenges. In: Proceedings of the 1st
International Conference on Internet of Things and Machine Learning, pp. 7:1–7:12 (2017)
3. Patwari, N., Ash, J.N., Kyperountas, S., Hero, A.O., Moses, R.L., Correal, N.S.: Locating the
nodes: cooperative localization in wireless sensor networks. IEEE Signal Process. Mag. 22(4),
54–69 (2005)
4. Shanshan, W., Jianping, Y., Zhiping, C., Guomin, Z.: A RSSI-based self-localization algorithm
for wireless sensor networks. J. Comput. Res. Dev. 45(1), 385–388 (2008)
5. Niculescu, D., Nath, B.: Ad hoc positioning system (APS) using AOA. In: INFOCOM 2003.
Twenty-Second Annual Joint Conference of the IEEE Computer and Communications. IEEE
Societies, vol. 3, pp. 1734–1743 (2003)
6. Shang, Y., Ruml, W.: Improved MDS-based localization. IEEE Infocom 4, 2640–2651 (2004)
7. He, T., Huang, C., Blum, B.M., Stankovic, J.A., Abdelzaher, T.: Range-free localization
schemes for large scale sensor networks. In: Proceedings of the 9th annual international con-
ference on Mobile computing and networking, pp. 81–95 (2003)
8. Niculescu, D., Nath, B.: Ad hoc positioning system (APS). In: Global Telecommunications
Conference, 2001. GLOBECOM’01. IEEE, vol. 5, pp. 2926–2931 (2001)
9. Zadeh, L.A.: Fuzzy logic = computing with words. IEEE Trans. Fuzzy Syst. 4(2), 103–111
(1996)
10. Guadane, M., Bchimi, W., Samet, A., Affes, S.: Enhanced range-free localization in wireless
sensor networks using a new weighted hop-size estimation technique. In: 2017 IEEE 28th
Annual International Symposium on Personal, Indoor, and Mobile Radio Communications
(PIMRC), pp. 1–5 (2017)
11. Hadir, A., Zine-Dine, K., Bakhouya, M., El Kafi, J.: An optimized DV-hop localization algo-
rithm using average hop weighted mean in WSNs. In: 2014 5th Workshop on Codes, Cryptog-
raphy and Communication Systems (WCCCS), pp. 25–29 (2014)
End-User Position-Driven Small Base
Station Placement for Indoor
Communication

Anindita Kundu , Shaunak Mukherjee, Ashmi Banerjee

and Subhashis Majumder

Abstract With the increase in the number of wireless end users, the demand for
high-speed wireless network has increased by multiple folds. Moreover, most of
the data traffic is observed to be generated from the indoor environment. Hence,
researchers have come up with the solution of deploying Small Base Stations (SBSs)
in the indoor environment, which proves to be immensely effective in providing the
last mile connectivity. However, since the deployment of the SBSs is unplanned, there
exists a high chance of co-tier interference, which might be mitigated by transmission
power control but at the cost of degraded quality of service for some of the end users.
Hence, in this work, we propose the concept of mobile SBSs which are connected
to power grids located in the ceiling of the deployment region. The mobile SBSs
position themselves at the received signal strength-based centroids of the end user
clusters, thereby mitigating the co-tier interference as well as conserving power of the
handsets by minimizing the distance between the SBS and the end-user handsets. To
enhance the security of the system, the SBSs are considered to form a closed group.
The proposed system requires about 13% lesser number of SBSs while enhancing the
end-user coverage by 9.3%. Moreover, about 10% improvement has been observed
in terms of cumulative throughput for mobile SBSs compared to fixed SBSs. Thus,
the deployment of mobile SBSs proves to be more effective compared to the fixed
SBSs for indoor communication.

A. Kundu (B) · S. Majumder

Heritage Institute of Technology, Kolkata, India
e-mail: [email protected]
S. Majumder
e-mail: [email protected]
S. Mukherjee
Whiting School of Engineering, Johns Hopkins University, Baltimore, USA
e-mail: [email protected]
A. Banerjee
Technische Universität München, Munich, Germany
e-mail: [email protected]

© Springer Nature Singapore Pte Ltd. 2020 13

R. Chaki et al. (eds.), Advanced Computing and Systems for Security,
Advances in Intelligent Systems and Computing 995,
https://doi.org/10.1007/978-981-13-8962-7_2
14 A. Kundu et al.

Keywords 5G network · Small cells · Mobile base station placement · K-means

clustering algorithm · Shortest path avoiding obstacles

1 Introduction

As the number of wireless end users are increasing, the demand for high-speed
communication is also increasing. However, recent research exhibits that most of the
data traffic is usually generated in the indoor environment [18]. 5G assures to satisfy
this demand and provide a high-speed communication platform using millimeter
waves, small cells, MIMO, etc. [21, 23]. Small cells have gained immense popularity
in recent years in providing satisfactory network coverage in the indoor environment
[13, 18]. They also participate in energy conservation by transmitting at a very low
Transmission Power (TP) [5].
Usually, the small cells are considered to be deployed in the end-user premises by
the end-user themselves and mostly in an unplanned fashion [10, 24]. Thus, unknow-
ingly two families residing in two adjacent apartments may place their Small Base
Stations (SBSs) on either side of the same wall which will lead to significant amount
of co-tier interference between the two, leading to a much degraded performance.
Also, the fixed, unplanned deployment of the small cells may lead to scenarios where
the end users may be located in a cluster at the cell edge which in turn will demand the
highest TP of the serving SBS. The transmission of the SBSs at their highest TP may
again lead to enhanced co-tier interference. Thus, as a solution to this problem, we
propose the deployment of mobile SBSs. This solution is applicable to both residen-
tial and commercial spaces. To provide enhanced security, the SBSs are considered
to form a closed group such that they can serve only verified end users [7].
A power grid is assumed to exist on the ceiling of the buildings where the coverage
is to be provided. It is considered that the SBSs can be mounted on the grid and are
also movable along the grid. The initial positions of deployment of the SBSs are
identified using the PSOM algorithm [14]. On identifying the positions, the SBSs
are mounted on the power grid at the positions identified by the PSOM algorithm.
The SBSs within a specific deployment region are all considered to be connected to
a special SBS which also acts as the Central Intelligent Controller (CIC). The CIC
may be located anywhere in the deployment region and is connected to the rest of
the SBSs via the wired power grid located at the ceiling. It monitors the position of
the other SBSs as well as the active end users via the SBSs. Initially, the SBSs are
all in low-powered sleep mode but monitor the end-user requests. Whenever an end
user requests for service, while the nearest SBSs is in sleep mode, the SBS counts
the number of times the request is being sent from the end user (Request Count-RC).
If the RC is below the desired threshold (RC T h ) of 15 [15], the SBS continues in
the sleep mode assuming that it may be some other SBS which is in the active state
and will take care of the end user. However, once RC = RC T h , the SBS turns itself
on and passes on the information to the CIC. The CIC then considers the cluster
End-User Position-Driven Small Base Station Placement for Indoor Communication 15

of the active end users and identifies the Cluster Center (CC) using the RSS-based
K-means clustering algorithm [1].
On finding the CC, the CIC identifies an A* algorithm-based shortest path, while
avoiding the obstacles [8] to identify the nearest SBS to the CC. It then instructs the
concerned SBS to follow the obstacle-free shortest path identified and place itself at
the desired CC. To evaluate the performance of the network after the deployment of
the SBS, the coverage and throughput of the system are next calculated. If either cov-
erage or throughput is observed to be below the desired threshold, the CIC partitions
the end-user cluster into further subclusters and thus identifies new CCs. The SBSs
located nearest (in terms of obstacle-free path) to these new CCs are then instructed
to move to CCs. More SBSs are activated if required. This process continues iter-
atively until all the end users have been accommodated or as long as the number
of available SBSs do not get exhausted. On positioning the SBSs, a distributed TP
control algorithm [15] is used by the SBSs to identify the minimum TP required to
satisfy the end users.
Results exhibit that mobile SBSs require about 13% lesser number of SBSs while
enhancing the end users’ coverage by 9.3% compared to the fixed SBSs. Moreover,
about 10% improvement has been observed in terms of cumulative throughput for
mobile SBSs. Thus, the mobile SBSs prove to be more effective compared to fixed
SBSs for indoor communication.
The rest of this paper is organized as follows. Section 2 discusses the existing
works in the literature, while Sect. 3 describes the scenario considered. Section 4
illustrates the methodology adopted. The proposed algorithm is illustrated in Sect. 5
followed by the discussion of the results obtained in Sect. 6. Finally, Sect. 7 concludes
the work.

2 Related Work

Small cells have been of great interest to the researchers for quite some time now
[2, 9, 11, 13, 22]. Ultradense networks comprising of small cells co-existing with
traditional macro-cells is one of the major features of the evolving 5G networks. In
spite of the various advantages of using small cells in terms of capacity enhancement,
cost-effectiveness, and energy efficiency, there are some major challenges that need
to be addressed. Maintaining an economic and ubiquitous connectivity among the
macro-cells and the small cells is one of them. Other significant challenges include
co-tier and cross-tier interference, energy efficiency, reliability, and maintenance of
the SBSs and security.
In this work, we have taken up the problem of placement of SBSs. The pro-
posed system ensures that the co-tier and cross-tier interference is minimized while
improving the energy efficiency of the system. Similar works have been taken up by
researchers in recent literature [3, 12, 20]. A Mixed Integer Nonlinear Programing
(MINLP) approach has been taken by authors of [12] to understand the various inter-
ference and multiplexing patterns at sub-6 GHz and microwave band. It is then solved
16 A. Kundu et al.

using linear relaxation and branch-and-bound algorithm and is applied in a wireless

backhaul network. The experiment conducted by the authors of [25] exhibits that
densely deployed dedicated indoor femtocells are more efficient in terms of spec-
trum and energy and provides a solution to the huge indoor capacity demands. Hence,
we have taken up the problem of placement of SBSs.
However, all of the aforementioned algorithms have considered the deployment
of fixed SBSs. Fixed SBSs may be sufficient for the traffic load at the time of its
deployment but may not satisfy the increasing demand with chores of time. It may also
prove to be inefficient in cases where the end users form nonuniform clusters. Hence,
we have proposed the concept of mobile SBSs which places itself at the centroid of
the clusters formed by the end users. The movement of the SBSs is controlled by
the CIC. There are various algorithms in the literature that deal with finding the
shortest path between two specific locations while avoiding the obstacles between
the nodes [4, 17]. However, traditional breadth-first search, Dijkstra’s algorithm, and
A* algorithms also provide the desired solutions [6]. Each of these algorithms has
its pros and cons. Since we are dealing with a graph almost similar to a grid, we
consider the A* algorithm to identify the shortest obstacle-free path between the old
and new positions of the SBSs.

3 Scenario Description

To model the scenario, the floor plan is system generated based on a random binary
matrix called GRID whose dimensions indicate the plan of the campus. Each cell
of the matrix represents the presence/absence of an obstacle in a unit area of the
deployment region considered. If a cell has value 1, it indicates the presence of an
obstacle, while a 0 indicates a free area. A cell of the matrix corresponds to 1 m ×
1 m region of the deployment area. N number of SBSs are initially considered to
be placed in a random fashion in the deployment region. Initially, these SBSs are
considered to be in the sleep mode. Now, n number of active calls are considered to be
generated from the randomly placed end users within the system. Figure 1 illustrates
such a randomly generated scenario.

4 Methodology Adopted

4.1 Forming the Graph of the Deployment Region

For the proposed algorithm to work in the CIC, the first job that needs to be done
is to model the deployment region. In order to model the deployment region, first a
random binary matrix GRID is generated. Each cell of the GRID corresponds to 1
sq. m. area of the actual deployment region. If the entry in the cell is 1, it implies
End-User Position-Driven Small Base Station Placement for Indoor Communication 17

Fig. 1 Initial scenario of the deployment region where the black squares indicate the obstacles, the
blue dots indicate the end users, while the big magenta dots indicate the SBSs

Fig. 2 Modeling the deployment region. a Plan of the deployment region. b The GRID matrix
corresponding to the plan of the deployment region. c Undirected graph corresponding to the plan
of the deployment region

that an obstacle is present in the actual deployment region corresponding to it while

a 0 implies free space.
For example, Fig. 2b exhibits the content of the GRID matrix corresponding to
a sample region given in Fig. 2a. Next, in order to find the shortest obstacle-free
path, the layout of the deployment region is converted into an undirected graph as
shown in Fig. 2c. In order to proceed with it, each section of the deployment region is
represented by a node, and two nodes are connected by an edge if their corresponding
sections are vertically or horizontally adjacent.
18 A. Kundu et al.

4.2 Identifying the Initial Positions of the SBSs and the End
Users

After the deployment region has been modeled, the N SBSs are deployed randomly.
Initially, all of them are considered to be in the sleep mode. Their Cartesian coordinate
positions with respect to the GRID matrix are updated in the CIC in the N × 2 matrix
called SBS_POS. Similarly, the end users are deployed randomly and their Cartesian
coordinate positions with respect to the GRID matrix are stored in the n × 2 matrix
called the USER_POS.

4.3 Identifying the Cluster Centers (CC) Positions

Once the position of the end users has been identified, the CIC employs a modified
version of the K-means clustering algorithm to divide them into K clusters based on
their Received Signal Strength (RSS) [1]. The K -value initially starts with 20 and
increases with every iteration. The coverage of the end users (C) and their cumulative
throughput (T ) are calculated at every iteration. If either C or T fails to achieve the
desired threshold, then K is incremented exponentially (with increasing powers of
2) and the RSS-based K-means clustering algorithm is executed to identify the next
set of CCs. The process continues till both the constraints of C and T are satisfied.
Once the satisfaction level is achieved for C and T , a binary search like method
is executed within the K-values ranging from K /2 to K to identify the minimum
K-value for which the constraints of C and T are satisfied. The CCs thus identified
form the set of desired SBS positions for the iteration. The CIC then instructs the
K -nearest SBSs (in terms of obstacle-free shortest path) to move to the CCs.

4.4 Calculating the RSS

The RSS of each end user is a function of their distance (d) from the CC and is
calculated using Eq. (1) where the expression for path loss is given by Eq. (2) [16].

RSS[d B] = T P[d B] − P L(d)[d B] (1)

d
P L(d)[d B] = P L(d0 ) + 10β.log10 ( ) + Xσ (2)
d0

In Eq. (2), d0 indicates the free space reference distance of 1 m, while β indicates
the optimized pathloss exponent (PLE) for a particular frequency band or environ-
ment and X σ indicates a zero mean Gaussian random variable with standard deviation
σ in dB. The height of the SBSs is considered to remain fixed at 7 m, and their oper-
ating center frequency is considered to be 28.0 GHz.
End-User Position-Driven Small Base Station Placement for Indoor Communication 19

4.5 Calculating the Coverage (C)

An end user is considered to be covered only if its RSS requirement, interference

requirement as well as its channel requirement are satisfied. Considering that K
SBSs are required to be deployed out of which j is the serving SBS, the RSS of
each end user is calculated using Eq. (1). The interference is considered to be the
cumulative sum of the RSS from the K − 1 neighboring non-serving SBSs and is
given in Eq. (3).
K
I nter f er ence[d B] = RSSi (3)
i=1,i= j

Each SBS is considered to have a fixed number of channels (c) of equal bandwidth
(b). If the cumulative channel requirement of a group of end users to be served by
an SBS is more than c, then the end users whose channel requirement could not be
satisfied is considered to be uncovered in spite of their RSS and interference constraint
being satisfied. In such cases, the RSS-based K-means clustering algorithm executes
again with increased K-value until the desired constraints are satisfied. Thus, the
coverage of the system is calculated as the percentage of end users whose RSS,
interference as well as the channel constraints are satisfied.

4.6 Calculating the Throughput (T )

T of the system indicates the cumulative throughput achieved by the end users in
the system. The individual throughput of the end users is calculated using Shannon–
Hartley theorem [19]. Thus, the T of the system is as given by Eq. (4) where B
indicates the bandwidth, S N R indicates the signal-to-noise ratio where noise refers
to Additive White Gaussian Noise (AWGN).


n
T = (Blog2 (1 + S N R))i (4)
i=1

4.7 Identifying the Obstacle-Free Shortest Path

The obstacle-free shortest path from the actual position of an existing SBS to its new
position identified by the CIC is calculated by traversing the graph given in Fig. 2c
using the A* algorithm. For each active SBS, if the current Cartesian position is given
by the coordinates (x, y), the node (x, y) in the graph (refer Fig. 2c) is considered to
be the source node. If the destination Cartesian position is (p, q), then the node (p, q)
in the graph is identified by the CIC as the destination node. Any intermediate node
Discovering Diverse Content Through
Random Scribd Documents
deposit, the red wretch may have repented his bargain and taken
steps to prevent our profiting by it."
"Look for tracks!" exclaimed the Corsican, on sudden inspiration,
but Britton shook his head.
"No use," he lamented, pointing to the pine-banked curve of the
river, shining like glass, "the ice is too clean!"
"Curse him! Curse him!" exploded Lessari, again, growing more
violent of speech.
"There's no use in cursing, either," Britton said seriously. "We're
facing death, Lessari, but we must keep alive as long as possible.
We have a tent and some food, and we'll make a strong fight."
The Corsican studied his dubious expression. "Go back?" he
asked.
"It can't be done," said Rex. "Our provisions will not last half the
time required to make the journey on foot, and there is nothing to
shoot over those barren stretches."
"Go on where gold is, then?" Lessari inquired dismally.
"Yes," Britton answered, "our path lies over those five hills. We
have only two chances, Lessari, and they are mighty slim! There is
the chance of stumbling on the encampment of these Thron-Diuck
Indians–they have retired somewhere in these mountains–and the
possibility of finding game in the pine forests. The way lies yonder,
and, if we find gold there, we'll stake it in case a miracle should
bring us out of this trap."
Rex stirred the nose of his dead leader with the toe of his
shoepack as he finished speaking, and Lessari saw him bend quickly.
"See that!" Britton exclaimed in quivering anger. He held out
something between his fingers, and the Corsican recognized a piece
of frozen whitefish covered with reddish powder.
"Poisoned!" he ejaculated with renewed horror.
"Yes, someone has fed them poisoned whitefish," said Rex,
vehemently. "Gaucho had this in his teeth!"
Lessari broke out in a flood of denunciation. Britton quelled his
own indignation and began untying the tent-ropes.
They thawed their canvas shelter from the banked ice and snow
by means of several brush fires and loaded the sled. Any articles
which could be dispensed with and which unnecessarily impeded
them were cast away. The outfit was reduced to a minimum, and
Rex packed all the remaining provisions carefully in one large sack.
He preserved, too, the food intended for the dogs, for he thought
they might easily find themselves in such straits as to be glad of it.
When all was securely lashed on the heavy Yukon sleigh, the
two men harnessed themselves in the traces and started laboriously
toward the circle of hills six miles away. For Lessari, they were six
long and excruciating miles. He was weak and unfit, and though
Britton took the heavier portion of the toil, the tramp told rapidly on
his companion.
The river curved with such a sweep that they struck overland to
shorten the distance. They bridged wide gullies full of blistered ice
and swerved erratically with the loaded sled among rugged rocks
and slippery hummocks that barred their path. Lessari continued to
mutter and complain during the whole six miles, his mumblings
toward the end becoming somewhat incoherent.
When they slipped down a long ravine which opened on the
river right in the middle of the circling hills, the Corsican was
staggering along with protruding tongue.
 "You're fagged!" Rex exclaimed, noticing his plight. "Better rest
here a minute!"
Lessari's answer was a vicious pull on the sleigh rope that nearly
took Britton off his feet. They moved on because the Corsican would
accept no delay, and Rex saw that the other's motive power was a
sort of delirium which instilled unlimited feverish energy.
The pair of toilers emerged at last from the black rift and
climbed an ice-capped ridge which fell like a sloping watershed in a
southward direction. Around them the five beaver-house mountains
rose strangely dome-like, the great river apparently losing itself in
the bowels of the thousand ice chasms which furrowed the base of
the valley-beds.
"This is the Klondike's source," Rex murmured as he
contemplated the scene, "and it looks cold enough to kill you."
"Yes," sighed Lessari, "you have it right. But the gold–the gold is
warm. Here I feel it!" He put his hand to his breast, and smiled
contentedly.
"It's all that's keeping you warm," Rex gruffly commented. The
observation quickly altered Lessari's expression, and he glared with a
wild impenetrable look as they proceeded to skirt the fringing line of
gravelled granite which was the shore of the now glacier-like stream.
Here the detached ice lay scattered about in huge blocks, an
impediment to their feet, where it had glided with the shining rubble
from the farther plateaus. In the shallow cup that the five hills
formed, they met with a long, treacherous crevasse whose yawning
depth of three hundred feet effectually cut off any further progress
in a direct line. The great abyss seemed to possess a fascination for
Lessari, and he trod dangerously near the edge to peer over.
 "Don't do that!" Britton sharply cautioned, pulling him back. "A
slip of your moccasin would put you at the bottom. We'll have to
leave the sled here and see if there is any way round!"
The immense crevasse dipped from an overhanging glacier on
one of the five mountains and slanted across the granite ridge they
had been skirting. The two men left the Yukon sleigh standing,
blocked, above the deep split and followed along the edge,
searching for a place to cross. The slant of the ravine became more,
acute, and, where the sides were jagged and shelved, they
clambered down lower and lower till the whole formation suddenly
broke upon a vast cavern that nosed into the river-bed and opened
on the other side where the way was passable though extremely
hard.
"It's rough going, but we must get across," Rex said, turning
round to Lessari.
The latter was handling some rusty-looking pebbles which he
had kicked out of the black cavern floorway.
"Ironstone!" he grunted scornfully, gazing at the cave side
where similar fragments with glacier-worn edges stuck out.
"Let me see," cried Britton, hastily jumping forward. Lessari
dropped the stones in his hand, and Britton's heart leaped at the
weight of them.
"Ironstone!" he exclaimed, his voice all trembling. "My God,
Lessari, it's gold!"
"Santa Virgin!" the Corsican screamed–"Gold!" He snatched
frantically at the precious pebbles, chattering madly.
"I'm positive it is," Rex said excitedly, "but the flame-test will
soon tell."
 He produced a bit of candle from his coat and lit it with
unsteady fingers. While Lessari held the specimens, he applied the
flame to them. The heat singed the Corsican's hands, but he did not
seem to feel any pain. Presently the rusty red covering of the
pebbles disappeared as fine dust in the blaze, and Lessari gripped
pure alluvial gold.
"Santa Virgin!" he screamed again. "We're rich! We're rich!"
Rex was off immediately, running about the cavern walls,
making a hasty survey with his candle end. The walls, like the floor,
were studded here and there with peeping corners of the precious
ore for which he had endured two thousand miles of pitiless Yukon
trails. Unbounded wealth lay within his grasp, and, with the triumph
of the moment, he forgot that he was a millionaire in a death-trap.
"Go up for a spade, Lessari," he cried. "It is a mighty
deposit–'big gold,' as the Thron-Diuck said."
The Corsican started up as a faint, rushing noise sounded
above, like ice sliding upon ice.
"What's that?" asked Britton anxiously.
They listened, but heard no further echo. Rex appeared ill at
ease.
"We're among glaciers, Lessari," he said, "and we must be
careful. An avalanche might easily bury us in a hole like this. Get
that shovel quickly!"
Lessari climbed up the lip of the ravine and disappeared, while
Britton pottered about, speculating, as well as exulting, over the
magnificent find. It was a showing that gave promise of surpassing
such far-famed creeks as the Eldorado and Bonanza, and Rex
gloated over his prospects. Standing in that deep cavern under the
Klondike's bed, his thoughts went back to the green Sussex lands,
Hyde Park in the London season, and the foaming Channel swells
under the Mottisfont's bows. He thought of the estates this buried
gold would buy, the power it would bring, the restoration to public
favor it would effect, and he laughed mirthlessly at the idea of
purchasing his way into quarters of society and diplomacy which had
closed their doors to him after his Algerian escapade.
A shrill cry from Lessari above interrupted his cogitations. He
scrambled out of the cavern and clawed his way up the slippery side
of the rift.
The Corsican was staring down into the abyss where they had
left the sled. On his face there rested a look of terrified
bewilderment, and he pointed into the gloomy depths.
"Gone!" he wailed–"gone down!"
Britton looked around for the sleigh, but it had vanished. A
sharp fear assailed him as he dashed to Lessari's side and saw the
mark of the runners on the powdered edge of the ravine where the
laden sled had taken the leap.
"That's what we heard slide," Rex groaned, "and it has all our
food!"
He went mechanically to the exact spot where the Yukon sleigh
had stood. There lay the piece of granite which had blocked the
runners, with the print of a husky's foot-pad in a minute snow-
pocket at its side. Rex showed it to the Corsican, a swift, ominous
wrath mantling his countenance.
"By heaven, Lessari, this is too much!" he cried. "It has been
done purposely like–like the poison! There's a hand in the dark
somewhere, and it means murder!"
 The Corsican's harrowed senses appeared incapable of
comprehending the statement.
"Starving–and rich!" he muttered wildly. "Rich–and starving!" He
walked without fear to the brink of the chasm and began to lower
himself over the rock with his hands.
"Here!" Rex roared in terror, rushing up. "What do you mean?"
"Stay back!" snarled the Corsican. "I go down to eat."
"The gold has turned your head!" Britton exclaimed. "You
couldn't get down there for all the food on earth. Why, man, it's
three hundred feet!" He sprang with a lithe movement and dragged
the Corsican from his perilous position.
Lessari gave an inhuman cry and closed with Britton. Rex saw
his eyes as they struggled and knew, with a feeling of chill horror,
that they were the eyes of a madman.
"Ha!" gasped the demented fellow. "This time you go!"
He strove to throw Britton into the gulf, for resistance had
resulted in giving his mania a different trend. The delirium gave him
the strength of six men, and Rex found himself being gradually
pushed into the crevasse. He strained and tugged with all the mighty
power of his shoulders and corded arms, but it was of no avail
against the frenzied Lessari. He tried another tack!
"Cool yourself, Lessari," he said soothingly, "and we'll get this
sled." They could never get it, but he hoped the artifice might serve!
Even that attempt at reason proved useless, for the Corsican
redoubled his efforts. The eternal cold, his illness, the death of the
dogs, the fever of the gold-finding, and the loss of their provisions
had all combined to drive him mad.
 "Devil!" he screamed, "you threw the food down!" And Rex
knew he was indeed demented.
Fighting every inch of the way, Britton was forced toward the
abyss. Three feet from it, he felt the necessity for desperate action.
Watching his opportunity, he tripped Lessari on the iced rock, and
they both fell heavily. Rex wound his arms about the Corsican,
putting forth the last ounce of strength; that grip of steel would have
held a giant, but it could not hold a madman. Lessari tore himself
free and gained the uppermost position, with hands on Britton's
throat.
Rex gazed into the rolling eyes, the wild, distorted visage of the
Corsican, and felt himself shoved to the very brink of the crevasse.
He wrenched violently at Lessari's wrists and arms, but they were as
iron rods, and the movement brought his head out over the rim of
the rock.
In one fleeting vision he saw the white, rising ice-fields cutting
into the blue sky, with glacier-capped peaks banking up behind; he
saw three of the five circling hills, their frozen gorges shining
emerald in the sun; then, as Lessari's wolfish face came closer to his
own and his arms were pressed down, the fingers felt the revolver
butt in his belt.
In sheer despair he grasped it as a drowning man snatches at
an oar. Its report cracked out and rattled in a hundred blatant
echoes down the gorge. Lessari uttered a gasping groan and lurched
to one side, his fingers lax and weak.
Britton wormed his shoulders back from the edge of the abyss,
shifting the Corsican's weight with his legs, and arose in safety. His
lungs were heaving with the tremendous strain like those of a spent
Channel-swimmer, and the cords of his throat were taut.
When he turned over the limp form at his feet, he looked into
Lessari's dead face.

CHAPTER XV.

Back in Dawson, on the evening of the same day when Britton stood
alone with the awful Klondike solitude at the edge of Five Mountain
Gulch–as it came to be named afterwards–when he faced at once
the icy phantom Cold, the grisly skeleton Starvation, and the devil-
faced thing Remorse, when he halted with death at his feet and its
dread power pervading the desolate snows about him, there
occurred, in the golden city, a strikingly different scene, a scene
palpitating with warmth and life.
A group of men, present at Grant Simpson's invitation, occupied
one of the ground-floor rooms of the Half Moon restaurant, engaged
ostensibly in doing justice to a very elegant and costly supper, but
really killing time in a luxurious way and waiting anxiously for the
bell-note of business which they knew their host intended to ring in
on them.
Simpson, with his accustomed lavish expenditure, had engaged
the room to the utter exclusion of other guests who might have
dined at two of the three tables which the chamber held; he had
ordered that the trio of tables be lined up and converted into one
long feasting board which could be covered with fine viands and
drinks–principally drinks! The catering was let to the hostess of the
Half Moon, Aline Giraud, who was a genius of management, all the
more so since Pierre's absence on the trails left every responsibility
in her hands. That night she expected him back from the completion
of his baggage-freighting contract with Laverdale, the big American
mine-owner who was bound for Dyea and the States, and Aline
wished to have everything right. She wished the supper that this
well-dressed, money-burning lawyer was giving to be a thing beyond
criticism, and her every effort was devoted to making it so.
And the bill! She told herself the bill would be the best of it all.
It would be a thing to cheer Pierre's heart and cause him to dance,
with his cap thrown among the ceiling festoons.
Simpson's was the dominating figure of the company present in
the room of the Half Moon where Aline Giraud served so assiduously
with her alert, graceful movements and her full, white arms. He
seemed to hold the key to some enterprise which claimed the
attention of all under their masks of good fellowship, but Simpson
did not yet consider the moment propitious for the unfolding of
hidden plans.
He sat at the head of his table, with his guests ranged in two
lines on either side, men well known in Dawson, the chief
characteristic of whom was money. That was why they were
present! If they had not had money to invest, they could have
entered into no proposition with Simpson.
Jarmand, the fat, wealthy broker with the currant-roll neck and
the oily insolence, was there; Fripps, the sour, thin, anæmic
promoter, maintained his usual unobtrusive but nevertheless certain
presence; a trio of capitalists of a somewhat similar stamp, keen-
visaged but rotund-bodied, quelled their impatience successfully,
while they secretly chafed at Simpson's dalliance, and awaited his
proposition. These men were inseparable in any business prospect;
they worked together, invested together, and stood or fell by a
triumvirate judgment; and since their names began with the same
letter–Cranwell, Crowdon, and Carr–they had been dubbed the three
C's.
Where the three C's went in, the financial project need not be
strictly legitimate. They had few scruples or qualms, and when they
took hold of a mining scheme or a real estate deal, wise men kept
out.
There were others present, probably a dozen in all, and among
them Jim Laurance, who had come with a great deal of misgiving
and scepticism on receipt of a letter from Simpson advising him of
an opportunity of getting in on the ground floor right under the
scoops of a dredging proposition.
And in preparation for his demonstration of ideas and plans,
Grant Simpson bade them all enjoy themselves, setting the example
himself with a free hand on the ladle of the punch bowl. Many
followed his example from appetite; the three C's imitated, thinking
of a relished business dessert as a sort of solace.
Famine might be threatening in the land of gold, but she had
certainly no embargo on liquors and cigars. Both were indulged in
without stint.
Blue, acrid wreaths of smoke filled the room, and the
atmosphere became very warm. No one would have guessed it was
forty below in the street, The two lines of guests at the table and the
host at its head emptied glasses and refilled, tossed them off and
ladled up again. Small talk hummed, and jests cracked out, more or
less coarse in the intervals when pretty Aline Giraud was absent from
the room during the different courses of the meal.
Jim Laurance, the only temperate one in the company, sipped
his simple glass of punch sparingly, refusing the bottled stuff and the
heavy wines. He felt disgusted and sorry that he had come, but he
had money to invest if Simpson's thing suited him, and he settled
himself to sit out the revel.
The roadhouse at Indian River had proved a good thing for
Laurance. He had struck his Klondike right on that creek, and he was
sane enough to know it. Instead of frittering away his coin on fool
stampedes in hopes of a mighty strike, he was satisfied to invest it in
sound mining securities and watch the dividends slowly grow. Such
an enterprise, he hoped, was in Simpson's mind.
Simpson's wine, however, was more in Simpson's thoughts than
the enterprise. He had unwisely glutted his taste for beverages with
a tang, and he lost control of his manners as well as his senses,
laughing boisterously and telling unsavory tales.
"Hi, there!" he would yell, skidding the empty punch-bowl down
the table to Jarmand. "Fill her up, Fatty. You're the doctor. Put in
something stiff–stiff enough to make your moustache stand!
Something d–d stiff, Fatty!"
"That's it, Jarmand," gurgled Bonneaves, a young profligate and
an especial chum of Simpson's. "Mix us a regular old hair-raiser.
We're out for fun! Who's holding us down?"
"No one! No one!" shouted three or four of the muddled men,
stamping on the floor and breaking into confused singing, which set
up rumbling echoes through the other parts of the restaurant and
went far to disturbing its customers.
"Tell us a story, Simp," said Jarmand. "Old Simp's the boy for
spicy ones. Eh, men? You bet your liver-colored notes he is. Rip one
off, Simp, there's a good fellow!"
Accordingly Simp ripped one off, a story that convulsed the
drinkers but which made Laurance's blood boil. The one-time
plainsman, now an Alaskan sourdough, sat very still, without the
shadow of a smile upon his face.
Aline Giraud, accompanied by a waitress, an ugly, angular
Danish woman, brought in the meats. These were bear steaks, slices
of moose flank, and grouse in pairs, a veritable feast which would
have fed a hundred poverty-pinched wretches in the outlying camps.
The thought came to Laurance as he poised his knife and fork over
the breast of a fat grouse dressed with sage dressing in a wonderful
brown gravy.
"Seems hard to waste this here," he said simply, "when there's
so many poor cusses starvin' round the Fields."
"To h–l with them!" cried Simpson, roughly. "What we have, we
got. Eh? We pay for it, and when you pay your way, the rest can go
and be d–d to 'em. How's that?"
"Right," nodded Bonneaves. "You're always right, Simp. You're a
wise old buck. Glad I've known you. You can show a fellow things.
Here's to you, Simp!"
The talk grew louder and looser. As the gravies were being
served, Simpson and Jarmand, exchanging winks, attempted a
double surprise. The lawyer made a bungling effort to kiss Aline
Giraud on the cheek, while at the same time the fat broker leaned
forward and pecked at the waitress. The result was a startling
surprise for Jarmand. The ham-like hand of the Danish woman
descended with a resounding smack on the currant-roll neck of the
broker.
The seated company roared at Jarmand. Jim Laurance frowned
at Simpson and half rose from his chair, but Aline had succeeded in
eluding the lawyer and fled through the doorway, the angry red
showing in her cheeks.
"That's one on you, Fatty," tittered his friends. "Beautiful throw-
down, that! Right place, too! Like another, Fatty? Better try again.
Ho! ho!"
"Cheer up, old man," laughed Simpson, accepting the joke.
"Better luck next time. Walk into the punch there, Fatty; you have a
weak heart."
They walked into the punch till the third bowl failed to withstand
the charges, and a fourth had to be mixed. Some of the men, unable
to restrain their vivacity, arose and capered about the laden table,
singing and playing the fool perfectly, and stopping only to refill
empty tumblers.
The Danish waitress, now secure in the triumph of her first
quick victory, held her ground undaunted, completing the serving of
the banquet in spite of the noise. Aline, no longer entering the room,
watched the progress of things through the doorway from the
farther chamber. Somehow, this fine supper over which she had
spent so much effort had not turned out as she had contemplated;
things were getting beyond her grasp; her eyes grew anxious wide,
and startled.
 After all, she thought, it might not please Pierre. Even the bill
would never compensate for the disgusting clamor and the
humiliation.
Laurance had finished his single glass of punch and was drawing
on his short, black pipe. He disdained the long, fat cigars of Jarmand
and the three C's, and cursed the ill-smelling, coronet-banded
cigarettes of Simpson and Bonneaves. The oddest figure in the
group himself, he felt nothing but contempt for the others. The only
thing about them he respected was the business instinct of their
sober moments, and there seemed but little chance for a display of
that now.
The Alaskan waited till the fourth bowl of punch ran low, hoping
that Simpson would open his mouth to speak sound sense, instead
of salacious nonsense, and tell them why he had invited them to
supper, but when the concoction of a fifth bowl was begun, amid
most uproarious hilarity, Laurance inwardly fumed, making up his
mind that he would not sit there much longer.
Unconsciously, he was frowning through the drifting haze of
smoke at his companions. There was no stern decorum present, nor
any nicety of attire. To be sure, Simpson, as host, and Bonneaves, to
imitate his model, wore dinner clothes, but the rest were dressed in
the ordinary dress which occupation demanded. The three C's were
in black broadcloth; Jarmand sported a suit of loud check pattern;
Fripps favored grey, as wrinkled and faded as his skin. The others of
the company were mostly mining men who had come in corduroys,
with trousers stuffed in knee-high cruisers, and had hung fur coats
and caps on the pegs behind their chairs. Laurance, travelling by
dog-train to Dawson, wore the musher's outfit of the trails.
 He looked rough and uncouth, but very much a man. His beard
was disreputable as ever; the iron-gray hair stood up stiffer and
stubbier, allowing his rat ears to be seen; his nose peeped out,
cherry-red and snub. He was lowering on the foolish antics of the
rest of the men, and his keen blue eyes were narrowed so much that
they did not flash.
"What's the matter with you, Laurance, old sport?" cried
Bonneaves, joyously. "Look as if you'd buried your best friend in the
punch-bowl!"
"Why," shouted Simpson, "if that's so, we'll resurrect him!
Resurrect's the word, boys. Eh? How's that?" He seized the bowl in
both arms and emptied it to the last drop in the array of glasses.
Then he turned the dish upside down on the table and hammered
upon its bottom, while the company roared as if he had done some
extremely witty thing.
"What say, Laurance?" asked young Bonneaves. "Feel any
better?"
"I feel like twistin' your cussed neck, young man," answered
Laurance, wrathfully. "What did I come here for? To eat a decent
meal an' talk business! I didn't come to swill meself–I'm certainly
certain of that! We're men anyhow, an' there's no call for us to act
like a lot of calf youngsters as can't pull the draw-string on their
gullets. I say we're here to talk business!"
"H–l, yes," grunted Bonneaves, with the air of sudden
recollection. "You're right, sport, now I come to remember. Simp did
bring us here for a purpose, and that's no lie. Give us your scheme,
Simp. Hot and heavy and fast–that's the way!"
 Because their tastes palled a little, the others added their
clamorous entreaties. Their exhortations made a confused babel:
"Hit it up, Simp! Uncork your oracle. Spread yourself quick, old
boy. What's the tune now? Time we talked, by gad!" And Bonneaves
nodded sagely at Laurance, muttering: "You're all right, sport. Simp's
a wise buck, but you're a wiser! See? Attention, you duffers!" He
secured order by pounding the board with the thick bottom of his
tumbler.
"Simp's going to spout," he announced authoritatively. Noticing
that the lawyer had engrossed himself with the opening of a
champagne bottle, Bonneaves hastily added: "Why, no! Rat me if he
isn't going to swallow! Here, Simp, that won't do. Put it away. Can't
you see your friends are waiting?"
"I'm busy," protested Simpson, struggling with the cork. "It's all
about that Yukon dredging business anyhow. I've taken it off Morris's
hands since he's played the fool and disappeared, d–n him! I need
backing. That's what I need. I can't go it alone!"
"What's the lay-out?" prompted Jarmand. "Put aside the bottle
and get down to business."
Simpson flung away the opener as a useless thing and grasped
a fresh one.
"Curse the bottle and curse the business," he fumed. "I'm busy,
I tell you. Here, I have the prospectus. Read it yourselves, and you'll
save my wind!" He drew some typewritten sheets from his breast-
pocket and flung them upon the cloth.
What he had called the prospectus passed down the line at one
side of the table, up again, and down the other side, greeted with
grunts of approval by those still clear-brained enough to understand
and with much head-wagging from such as were incapable of
comprehension.
"Bully!"
"Standard bred!"
"Up to snuff!"
"Neat as garters!"
These were some of the comments from the appreciative
assembly.
Last of all, the prospectus came to Jim Laurance. At the top of
the sheet, in large typing, was the name, "Yukon Dredging
Company." Underneath that reposed the list of directors, picked,
apparently, from the group invited to supper. Jarmand's name
appeared, and Fripps's, Bonneaves's, and the names of the three
C's.
Laurance quietly read the sheets through, with their significance
vitally impressing itself on him, and when he finished, he saw that
he held the kind of thing which is circulated by thousands through
the mails for the catching of suckers. It was the universally familiar,
folded sheet that expounded the virtues of the greatest dredging
proposition in the world.
"By gad," he cried, angrily shaking the prospectus in the air, "so
this is what you've hauled me over here to back up, eh? A cussed,
dirty, widow-an'-orphan robbin' swindle, if you ast me! An', gents, I
give it to you straight: you're a pack of low faro dealers, a bunch of
thimbleriggers, a handful of flimflammers if you put through that
there deal. You're a ring of thieves and d–d blacklegs, gents!"
"Hold on there, sport!" yelled Bonneaves. "You go it too strong.
We won't stand for all that."
 "I can go lots stronger yet, young cocky-neck," warned
Lawrence. "Why, I ain't half goin'. You should see me fizz some time,
me son, an' you'd run your feet off for fear of bein' blowed up." He
regarded the youthful profligate grimly, shaking his stubby scalp and
gray beard aggressively, but in the corners of his eyes there lurked a
humorous expression.
"Aren't you in on this?" asked Jarmand, rolling a wave of his oily
insolence down the table to Laurance. "Aren't you taking hold?
There's money in it!"
The Alaskan eyed him squarely.
"Not the kind of money I want," he said severely. "Not me own
kind, by a thousand yard shot! I don't want no widow's mites or
orphan's pennies; I don't steal no wimmen's savin's nor the hard-
earned dollars of some poor laborin' cuss as thinks the Yukon is one
whoppin' lump of gold an' all we got to do here is to file up our
finger-nails and claw it off in pieces. No, sir, count me out! An' I'll
see some law-sharp an' have you gents counted out, too. You don't
work this here game so easy. I'm certainly certain of that! You can't
rob people so d–d bare-faced. No, sir, you truly can't. Why, this here
would be wors'n jumpin' all the claims on Samson Creek!"
Laurance's glance rested full on Grant Simpson as he uttered his
bold words, and the lawyer looked up with suspicious, drink-steeped
eyes.
"What the devil's wrong with this thing?" he demanded angrily.
"What puts your back up?"
"Look here," snapped Laurance, pointing to the typewritten
sheet. "You claim to have one hundred miles river frontage, or 'bout
ten thousand acres, on Indian Creek. You bought it from the
Government! Pretty lie, if you ast me! Clear title from them, and all
the rest of the high-falutin's! Pah!–it turns me sick. For you haven't a
yard–not one d–d yard. I'm there, an' I know!"
The Alaskan's vehemence drew the attention of everyone, drunk
or sober.
"An' you have two dredges at work, expectin' a third," he went
on, continuing to read from the prospectus. "That's a crackin' good
Sunday paper joke. What does it mean?"
"Well," growled Simpson, "we will have. We intend to."
"The devil you do," said Laurance. "You'll put the money in your
pocket an' keep it there. To h–l with your prospectus!" He tore the
sheets in half and threw the fragments on the floor.
Simpson laughed. He viewed the whole affair with colossal
unconcern. In its time he could proceed with the venture at
immense gain to himself and the others. It must be postponed, in
spite of it being the reason for the assembly, because, just now,
wine was a much more important thing.
"You don't have to plunge," he commented. "Stay out if you
can't like it."
"Yes, but he doesn't need to give us extra work," interposed
Jarmand, expostulating about the torn prospectus.
"Have an ice, Laurance." advised young Bonneaves. "It'll cool
you down."
"I'll have nothin'," Laurance growled, reaching for his coat. "I
don't hanker after suppin' with them as I now know is thieves."
At the host's call, the Danish waitress brought in the ices on a
tray, while Jim Laurance muffled himself in his coat.
 "Where's Aline?" Simpson asked, assuming the privilege of
familiarity.
"My mistress?" said the waitress. "She will serve no more. She
will not enter."
"But she'll have to," cried Simpson, flushing with anger and
obstinacy. "Tell her to run in and serve immediately or I shall come
after her and kiss both her cheeks instead of one."
The Danish woman flounced out, and Jarmand involuntarily put
his fingers to his fat neck.
"You see," explained Simpson, "it isn't like as if I hadn't paid her
for the supper and for occupying her room. And, by the way, this
isn't the only room!" He nodded and laughed evilly, adding: "The
hubby's on the trails."
Laurance's coat went off his back with a reverse of the motion
which was putting it on. The garment flew into one corner, and the
owner's voice rang out across the room like the clank of good steel.
"By heaven, Simpson," he roared, "you can't throw one speck of
mud on Pierre's wife. You'll eat dirt for it. You're a d–d dago-hearted
liar!"
Laurance sprang along behind the row of chairs to reach
Simpson at the table's head, but a hand caught his elbow as he
passed the side door and whirled him about. With the suddenness of
an apparition, he saw Pierre, in musher's dress, fresh from the trails,
filling the entrance with his bulk, so that the white face of Aline had
to peer under the arm which held Laurance back.
"Dis for me, camarade," murmured Pierre, pushing the Alaskan
behind him.
 Giraud then walked quickly past the astonished men till he stood
in front of Simpson. Very deliberately he gazed at him.
"M'sieu'," he said, "you wan coward. You wan dam coward!"
And his open palms gave Simpson a stinging blow on either cheek.
The lawyer lashed out with both hands and feet, but Pierre
grasped him by the throat and shook him like a long rag. Bedlam
broke loose! Chairs and tables were overturned as the half-dazed
revellers jumped up. Aline's screams were mingled with the crash of
glass and chinaware. Jarmand, Bonneaves, and two or three more of
Simpson's friends rushed to his assistance, bent on violence toward
Pierre, but Jim Laurance swung on them sharply, with eight inches of
blued, cylindrical steel glittering in either hand.
"Back there," he yelled, "every man-jack of you, or I'll plug you
with these gas-pipes!"
The glinting light on the dull, ugly Colts daunted them no more
than the determined gleam in the eyes of the man behind. The
rescuers fell aside like gale-blown gravel and remained glued to the
wall.
Pierre Giraud set the lawyer on his feet. The voyageur's face
was pale and rigid.
"M'sieu'," he said, "you lak wan feather in my hand. Ah no be go
fight wit' you dat way, 'cause dat not be fair. Mais you geeve Aline
wan insult–de wors' insult dat man could geeve! An' Aline, she lak
wan leetle w'ite saint. M'sieu'," and he tapped Simpson's shoulder,
"wan of us be keel here. Ah keel you, fair, or you keel me. Tak' de
choice of dose!" He indicated Laurance's pistols.
It was no orthodox duel. There occurred no pacing, no
arrangement, no seconding, no counting! Laurance put one weapon
in Simpson's hand, whipped the other over to Giraud, and stepped
between the door-jambs, screening the thing from Aline.
Abruptly the shooting began, the revolvers spurting jets of
flame through the blue haze of the room, whose atmosphere
thickened into swirling wreaths with every report.
It was a scene of the wildest disorder, with the overturned
tables and chairs and shattered glass below; lights above, swaying
to the explosions of the pistols; at the sides the lines of awed yet
excited men flattened against the walls; the anxious Laurance and
the frantic, white-faced wife in the side entrance; guests fleeing
from the other parts of the establishment with shrieks and clamor;
and in the centre of it all the two combatants manoeuvring in the
mist of smoke to avoid being hit, advancing and firing swiftly as they
advanced.
Simpson shot the faster, with wild, deadly, malevolent hatred;
Giraud directed his weapon with slower deliberateness, ruled by one
earnest, avenging impulse. The room rocked to the deafening
reverberations of the pistols; the bullets went pang-panging on the
wainscoting; the jets of flame turned to crossed spears stabbing
through the smoke.
In ten seconds the men were within gun-reach in the centre of
the floor. Simpson's sixth ball broke the skin on his opponent's neck,
but Giraud's fifth went hurtling through the lawyer's brain.
Simpson sagged in a little heap of black tuxedo and white
starch, his brow stained with spurting red. Aline Giraud was sobbing
on Pierre's breast, but Laurance roused him roughly to an
acceptance of realities.
 "Hit it, an' hit it quick!" Jim urged vociferously. "The Mounted
will be here on the run in a minnit. Gad, that firin' must wake up the
whole town. Where's the dog-train? Is it unhitched?"
"Non," answered Pierre, speaking like a man in a dream, "she
be in de yard lak Ah left her."
"Come on, then," whispered Laurance, pulling him out.
Aline clung to him piteously, and Pierre embraced her with a
swift, despairing, passionate gesture. Then he put her from him with
an effort that was agony.
"He'll come back," consoled Laurance, "as soon as this blows
over. Come on, Pierre. I hear runnin'."
They were gone on the instant, leaving Aline Giraud with her
sweet, white face upturned in prayer and her hands clasped in an
attitude of fear, parting, and renunciation.
When the uniformed men of the Mounted Police filled the room
where Simpson lay dead, Pierre was galloping his dog-team at full
speed up the ice-trail of the Klondike.
"Hit it for the Thron-Diuck camps," Laurance had advised.
"They're somewhere in them mountains. An' lie low till I send you
word by an Indian."
That was how Pierre, heading for the Thron-Diuck
encampments near the Klondike's source, found Rex Britton four
days later, half dead from starvation and exposure, with his last
burned match in his pocket, ravings on his tongue and delirium in his
brain, about fifteen miles from Five Mountain Gulch.

CHAPTER XVI.
Welcome to our website – the ideal destination for book lovers and
knowledge seekers. With a mission to inspire endlessly, we offer a
vast collection of books, ranging from classic literary works to
specialized publications, self-development books, and children's
literature. Each book is a new journey of discovery, expanding
knowledge and enriching the soul of the reade

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

Let us accompany you on the journey of exploring knowledge and

personal growth!

textbookfull.com