Textbook Mathematical and Statistical Applications in Life Sciences and Engineering 1St Edition Avishek Adhikari Ebook All Chapter PDF

Mathematical and Statistical
Applications in Life Sciences and

Engineering 1st Edition Avishek
Adhikari
Visit to download the full and correct content document:
https://textbookfull.com/product/mathematical-and-statistical-applications-in-life-scien
ces-and-engineering-1st-edition-avishek-adhikari/
More products digital (pdf, epub, mobi) instant
download maybe you interests ...
Mathematical and Statistical Applications in Food

Engineering 1st Edition Surajbhan Sevda (Editor)
https://textbookfull.com/product/mathematical-and-statistical-
applications-in-food-engineering-1st-edition-surajbhan-sevda-
editor/
Mathematical Methods in Engineering and Applied

Sciences (Mathematics and its Applications) 1st Edition
Hemen Dutta (Editor)
https://textbookfull.com/product/mathematical-methods-in-
engineering-and-applied-sciences-mathematics-and-its-
applications-1st-edition-hemen-dutta-editor/
An Introduction to Statistical Analysis in Research:

With Applications in the Biological and Life Sciences
1st Edition Wiley
https://textbookfull.com/product/an-introduction-to-statistical-
analysis-in-research-with-applications-in-the-biological-and-
life-sciences-1st-edition-wiley/
Mathematical and Statistical Methods for Actuarial

Sciences and Finance MAF 2016 1st Edition Marco Corazza
https://textbookfull.com/product/mathematical-and-statistical-
methods-for-actuarial-sciences-and-finance-maf-2016-1st-edition-
marco-corazza/
Advanced Mathematical Techniques in Engineering
Sciences First Edition Davim
https://textbookfull.com/product/advanced-mathematical-
techniques-in-engineering-sciences-first-edition-davim/
Recent Advances in Mathematics for Engineering

(Mathematical Engineering, Manufacturing, and
Management Sciences) 1st Edition Mangey Ram (Editor)
https://textbookfull.com/product/recent-advances-in-mathematics-
for-engineering-mathematical-engineering-manufacturing-and-
management-sciences-1st-edition-mangey-ram-editor/
Physical Sciences and Engineering Advances in Life

Sciences and Oncology A WTEC Global Assessment 1st
Edition Paul Janmey
https://textbookfull.com/product/physical-sciences-and-
engineering-advances-in-life-sciences-and-oncology-a-wtec-global-
assessment-1st-edition-paul-janmey/
Handbook of Fractional Calculus with Applications Vol 7

Applications in Engineering Life and Social Sciences
Part A De Gruyter Reference 1st Edition Dumitru
Baleanu
https://textbookfull.com/product/handbook-of-fractional-calculus-
with-applications-vol-7-applications-in-engineering-life-and-
social-sciences-part-a-de-gruyter-reference-1st-edition-dumitru-
baleanu/
Handbook of Fractional Calculus with Applications Vol 8

Applications in Engineering Life and Social Sciences
Part B De Gruyter Reference 1st Edition Dumitru
B■leanu (Editor)
https://textbookfull.com/product/handbook-of-fractional-calculus-
with-applications-vol-8-applications-in-engineering-life-and-
social-sciences-part-b-de-gruyter-reference-1st-edition-dumitru-
Avishek Adhikari
Mahima Ranjan Adhikari
Yogendra Prasad Chaubey Editors
Mathematical
and Statistical
Applications in
Life Sciences and
Engineering
Mathematical and Statistical Applications in Life
Sciences and Engineering
Avishek Adhikari Mahima Ranjan Adhikari
•
Yogendra Prasad Chaubey

Editors
Mathematical and Statistical

Applications in Life Sciences
and Engineering
123
Editors
Avishek Adhikari Yogendra Prasad Chaubey
Department of Pure Mathematics Department of Mathematics and Statistics
University of Calcutta Concordia University
Kolkata, West Bengal Montreal, QC
India Canada
Mahima Ranjan Adhikari

IMBIC
Kolkata, West Bengal
India
ISBN 978-981-10-5369-6 ISBN 978-981-10-5370-2 (eBook)

https://doi.org/10.1007/978-981-10-5370-2
Library of Congress Control Number: 2017953827
© Springer Nature Singapore Pte Ltd. 2017

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, express 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.
Printed on acid-free paper
This Springer imprint is published by Springer Nature

The registered company is Springer Nature Singapore Pte Ltd.
The registered company address is: 152 Beach Road, #21-01/04 Gateway East, Singapore 189721, Singapore
Preface
This is the commemoration volume celebrating the 10th foundation anniversary

of the Institute for Mathematics, Bioinformatics, Information Technology and
Computer Science (IMBIC), a research institute that was founded in India in 2006
with branches in Sweden and Japan, and is dedicated to the scientific and technical
activities at the forefront of various areas across interdisciplinary Mathematics,
Statistics, and Computer Science.
To achieve its objectives, IMBIC organizes international interdisciplinary con-
ferences in every December in Kolkata on “Mathematical Sciences for
Advancement of Science and Technology (MSAST).” The conference MSAST
2016 was the 10th of the series of these conferences. Prominent mathematicians and
statisticians were invited to contribute articles to the celebration of this milestone.
The response was overwhelming. The present volume was conceived in order to
exhibit the beauty of mathematics and statistics and emphasize applications in allied
sciences. In order to achieve this goal, eminent scientists featuring such contribu-
tions were invited to contribute to the volume. As a culmination of this effort, the
present volume contains articles covering a wide range of topics of current
importance.
The articles on mathematical applications cover the topics of coloring problem,
control of stochastic structures, and information dynamics, whereas the articles on
statistical applications cover the areas on image denoising, life testing and relia-
bility, survival and frailty models, analysis of drought periods, prediction of
genomic profiles, competing risks, environmental applications, and chronic disease
control. The articles contained in this volume present state-of-the-art material along
with a detailed and lucid review of the relevant topics and issues concerned. This
volume reports on the newest developments in some of the interesting and
promising areas of mathematical and statistical research today, and is expected to
provide an important resource for researchers and practitioners in the relevant areas
mentioned above. The key features of this volume such as
v
vi Preface
• Focus on mathematical and statistical applications in a single volume,

• Accessible and lucid presentation,
• Detailed review of the cutting edge technologies,
• Articles from the worldwide experts in the field, and
• Wide range of topics on applications of modern relevance,
are expanded in 16 book chapters.
Mathematical applications span the first three chapters. Chapter 1 deals with the
problem of allocating radio frequencies to the base stations of the cellular network
such that interference between stations at a different distance can be avoided while
keeping the range of distinct frequencies to a minimum. Chapter 2 discusses the use
of vibration control techniques in order to control or suppress wild vibrations in
structures caused by resonance or flutter due to external sources along with a
numerical example with a recorded dataset from a California earthquake. Chapter 3
discusses the solutions of Hamilton ODE’s and Hamilton–Jacobi PDE’s involving
single-time and multi-time higher-order Lagrangians, and is expected to be of
interest to engineers and applied mathematicians.
Statistical applications span the next 13 chapters. Chapter 4 presents a com-
prehensive overview of the denoising problem and subsequent development in the
context of cDNA microarray images, in the context of genomic research, where
wavelet-based methods have been quite useful. Chapter 5 introduces a
transformation-based distribution family appropriate for hazard modeling where the
well-known Weibull model may fail, and Chap. 6 introduces yet another family
based on the so-called mode-centric Gaussian distribution as a fraternal twin of the
Gaussian distribution appropriate for life-data models.
Chapter 7 considers the use of stochastic volatility models for analyzing drought
periods with discussions of real examples. Chapter 8 gives a new estimator of the
mean residual life function and compares its performance with other nonparametric
estimators available in the literature while illustrating the relevance of the results on
a real dataset. Chapter 9 gives a detailed overview of techniques for predicting
outcomes such as colon cancer survival from diverse genomic profiles whereas
Chap. 10 discusses a bivariate frailty model with an application to competing risk
theory. Chapter 11 considers the estimation of the stress strength parameter under
the Bayesian paradigm with illustrations on a real dataset.
Chapter 12 reports on the spatiotemporal analysis of air pollution effects on clinic
visits using the data from Taiwan that is based on several Bayesian techniques, and
Chap. 13 gives a comprehensive review of statistical inference procedures in the
context of competing risks when the cause of failure may be missing or masked for
some units. Chapter 14 discusses and illustrates several environmental applications
based on the Birnbaum–Saunders model, and Chap. 15 considers multistate mod-
eling in the context of data on chronic conditions from individuals in disease reg-
istries. The final chapter, Chap. 16, presents estimation for lifetime characteristics in
connection with a time-constrained life-testing experiment.
We had invited more than 50 potential contributors, and 19 articles were sub-
mitted. After a careful review, 16 articles were selected that make the 16 chapters
Preface vii
of the present volume. We are thankful to all the contributors to this volume and
especially to the corresponding authors for their cooperation and timely submission
of the articles and revisions. We have derived substantial help in reviewing these
articles from the following reviewers to whom we owe our immense gratitude:
Jorge A. Achcar (Brazil), Koby Asubonteng (USA), Saria Awadalla (USA), M.
L. Lakhal-Chaieb (Canada), Ashish K. Chattopadhyay (India), Propser Donovon
(Canada), Sujit K. Ghosh (USA), David Hanagal (India), Tamanna Howlader
(Bangladesh), Abdulkadir Hussein (Canada), Nachimuthu Manickam (USA),
Manoel Santos-Neto (Brazil), Partha Sarathi Roy (Japan), Satyajit Roy (India),
Helton Saulo (Brazil), and Jingjing Wu (Canada).
We apologize for any omissions. We are also grateful to Springer for publishing
this volume and to all individuals who have extended their support and cooperation
in order to bring the project of IMBIC for publishing this commemorative volume
to fruition.
Kolkata, India Avishek Adhikari

Kolkata, India Mahima Ranjan Adhikari
Montreal, Canada Yogendra Prasad Chaubey
August, 2017
Contents
Part I Mathematics
1 Hole: An Emerging Character in the Story of Radio k-Coloring
Problem . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3
Ushnish Sarkar and Avishek Adhikari
2 Robust Control of Stochastic Structures Using Minimum Norm
Quadratic Partial Eigenvalue Assignment Technique . . . . . . . . . . . 43
Kundan Goswami, Sonjoy Das and Biswa Nath Datta
3 Single-Time and Multi-Time Hamilton–Jacobi Theory Based on
Higher Order Lagrangians . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 71
Savin Treanţă and Constantin Udrişte
Part II Statistics
4 On Wavelet-Based Methods for Noise Reduction of cDNA
Microarray Images . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 99
Tamanna Howlader, S. M. Mahbubur Rahman
and Yogendra Prasad Chaubey
5 A Transformation for the Analysis of Unimodal Hazard Rate
Lifetimes Data . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 121
Kobby Asubonteng, Govind S. Mudholkar and Alan Hutson
6 The Power M-Gaussian Distribution: An R-Symmetric Analog
of the Exponential-Power Distribution . . . . . . . . . . . . . . . . . . . . . . 141
Saria Salah Awadalla, Govind S. Mudholkar and Ziji Yu
7 Stochastic Volatility Models (SVM) in the Analysis of Drought
Periods . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 159
Jorge Alberto Achcar, Roberto Molina de Souza
and Emílio Augusto Coelho-Barros
ix
x Contents
8 Nonparametric Estimation of Mean Residual Life Function Using

Scale Mixtures . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 169
Sujit K. Ghosh and Shufang Liu
9 Something Borrowed, Something New: Precise Prediction of
Outcomes from Diverse Genomic Profiles . . . . . . . . . . . . . . . . . . . . 193
J. Sunil Rao, Jie Fan, Erin Kobetz and Daniel Sussman
10 Bivariate Frailty Model and Association Measure . . . . . . . . . . . . . . 209
Ramesh C. Gupta
11 On Bayesian Inference of R = P(Y < X)
for Weibull Distribution . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 225
Debasis Kundu
12 Air Pollution Effects on Clinic Visits in Small Areas of Taiwan: A
Review of Bayesian Spatio-Temporal Analysis . . . . . . . . . . . . . . . . 247
Atanu Biswas and Jean-Francois Angers
13 On Competing Risks with Masked Failures . . . . . . . . . . . . . . . . . . 257
Isha Dewan and Uttara Naik-Nimbalkar
14 Environmental Applications Based on Birnbaum–Saunders
Models . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 283
Víctor Leiva and Helton Saulo
15 Analysis of Chronic Disease Processes Based on Cohort and
Registry Data . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 305
Richard J. Cook and Jerald F. Lawless
16 Exact Likelihood-Based Point and Interval Estimation for
Lifetime Characteristics of Laplace Distribution Based on a
Time-Constrained Life-Testing Experiment . . . . . . . . . . . . . . . . . . 327
Xiaojun Zhu and N. Balakrishnan
Editors and Contributors
About the Editors

Avishek Adhikari, Ph.D. is Assistant Professor of Pure Mathematics at the University of
Calcutta. He is a recipient of the President of India Medal and Young Scientist Award. He was a
post-doctorate fellow at the Research Institute INRIA, Rocquencourt, France. He was a visiting
scientist at Indian Statistical Institute, Kolkata, and Linkoping University, Sweden. He visited
many institutions in India, Japan, Sweden, France, England, Switzerland, and South Korea on
invitation. His main interest lies in algebra, cryptology, discrete mathematics, theoretical computer
science, and their applications. He has published four textbooks on mathematics including one
book Basic Modern Algebra with Applications (Springer) and edited one research monograph. He
has published several papers in foreign journals of international repute, conference proceedings,
and book chapters. Four students have already been awarded Ph.D. degree under his guidance. He
successfully completed several projects funded by the Government of India and is a member of the
research teams from India for collaborative Indo-Japan (DST-JST and DST-JSPS) research pro-
jects. He is a member on editorial board of several journals and Founder Secretary (honorary)
of the Research Institute IMBIC and Treasurer (honorary) of the Cryptology Research Society of
India (CRSI).
Mahima Ranjan Adhikari, Ph.D. is the Founder President of the Institute for Mathematics,
Bioinformatics, Information Technology and Computer Science (IMBIC), Kolkata, and former
Professor of Pure Mathematics at the University of Calcutta. He has published a number of papers
in several Indian and foreign journals including Proceedings of American Mathematical Society
and eight textbooks including Basic Modern Algebra with Applications (Springer) and Basic
Algebraic Topology and its Applications (Springer). Twelve students have already been awarded
Ph.D. degree under his guidance on various topics such as algebra, algebraic topology, category
theory, geometry, analysis, graph theory, knot theory, and history of mathematics. He is a member
of American Mathematical Society and on the editorial board of several Indian and foreign
journals and research monographs. He was elected as the president of the Mathematical Science
Section (including Statistics) of the 95th Indian Science Congress, 2008. He visited several
institutions in India, USA, UK, China, Japan, France, Greece, Sweden, Switzerland, Italy, and
many other countries on invitation.
xi
xii Editors and Contributors
Yogendra Prasad Chaubey, Ph.D. is Professor of Statistics in Department of Mathematics and

Statistics at Concordia University, Montreal, Canada. His research interests include Sampling,
Linear Models, Distribution Theory, and Nonparametric Smoothing. His current research is
focused on nonparametric functional estimation that has been funded by discovery grant program
from Natural Sciences and Engineering Research Council of Canada. He has been active in
promoting Statistics through membership in various capacities of several statistical associations
and organization of scholarly conferences such as Statistics 2001 Canada, Statistics 2011 Canada,
and the workshop on Nonparametric Curve Smoothing during July 2001, July 2011, and
December 2013, respectively, that were held at Concordia University. He served and is currently
serving on the editorial board of several statistical journals and has trained approximately 40
masters and 6 doctoral students. He was elected as a member of the prestigious International
Statistical Institute in 2005 and was later inducted as a member of the Provost Circle at Concordia
University. He was awarded P. Stat. designation by the Statistical Society of Canada in 2014 for
his professional statistical qualifications and was given Life Achievement Award for his leadership
and contributions to Statistics by Forum for Interdisciplinary Studies in 2016. He has edited three
research monographs and published over 130 research articles in scholarly statistical journals of
international repute, conference proceedings, and book chapters.
Contributors
Jorge Alberto Achcar Medical School, University of São Paulo, Ribeirão Preto,
SP, Brazil; Department of Social Medicine, FMRP University of São Paulo, Monte
Alegre Ribeirão Preto, SP, Brazil
Jean-Francois Angers University of Montreal, Montreal, QC, Canada
Kobby Asubonteng AstraZeneca Pharmaceuticals, Gaithersburg, MD, USA
Saria Salah Awadalla Division of Epidemiology and Biostatistics, UIC School of
Public Health (SPH-PI), Chicago, IL, USA
N. Balakrishnan Department of Mathematics and Statistics, McMaster University
Hamilton, Hamilton, ON, Canada
Atanu Biswas Applied Statistics Unit, Indian Statistical Institute, Kolkata, India
Emílio Augusto Coelho-Barros Federal Technological University of Paraná,
Cornélio Procópio, PR, Brazil
Richard J. Cook Department of Statistics and Actuarial Science, University of
Waterloo, Waterloo, ON, Canada
Sonjoy Das Department of Mechanical and Aerospace Engineering, University at
Buffalo, Buffalo, NY, USA
Biswa Nath Datta Department of Mathematical Sciences, Northern Illinois
University, De Kalb, IL, USA
Isha Dewan Theoretical Statistics and Mathematics Unit, Indian Statistical
Institute, New Delhi, India
Jie Fan Division of Biostatistics, Department of Public Health Sciences, Miller
School of Medicine, University of Miami, Miami, FL, USA
Editors and Contributors xiii
Sujit K. Ghosh Department of Statistics, NC State University, Raleigh, NC, USA

Kundan Goswami Department of Mechanical and Aerospace Engineering,
University at Buffalo, Buffalo, NY, USA
Ramesh C. Gupta University of Maine, Orono, ME, USA
Tamanna Howlader Institute of Statistical Research and Training, University of
Dhaka, Dhaka, Bangladesh
Alan Hutson Department of Biostatistics, University at Buffalo, Buffalo, NY,
USA
Erin Kobetz Division of Biostatistics, Department of Public Health Sciences,
Miller School of Medicine, University of Miami, Miami, FL, USA
Debasis Kundu Department of Mathematics and Statistics, Indian Institute of
Technology Kanpur, Kanpur, India
Jerald F. Lawless Department of Statistics and Actuarial Science, University of
Waterloo, Waterloo, ON, Canada
Víctor Leiva Avenida Brasil 2241, Valparaíso, Chile
Shufang Liu Department of Statistics, NC State University, Raleigh, NC, USA
Roberto Molina de Souza Federal Technological University of Paraná, Cornélio
Procópio, PR, Brazil
Govind S. Mudholkar Department of Statistics and Biostatistics, University of
Rochester, Rochester, NY, USA
Uttara Naik-Nimbalkar Department of Mathematics, Indian Institute of Science
Education and Research (IISER), Pune, India
S. M. Mahbubur Rahman Department of Electrical and Electronic Engineering,
University of Liberal Arts Bangladesh, Dhaka, Bangladesh
Ushnish Sarkar Department of Pure Mathematics, University of Calcutta,
Kolkata, India
Helton Saulo Avenida Brasil 2241, Valparaíso, Chile; Department of Statistics,
University of Brasilia, Brasilia, Brazil
J. Sunil Rao Division of Biostatistics, Department of Public Health Sciences,
Daniel Sussman Division of Biostatistics, Department of Public Health Sciences,
Savin Treanţă Faculty of Applied Sciences, Department of Applied Mathematics,
University “Politehnica” of Bucharest, Bucharest, Romania
xiv Editors and Contributors
Constantin Udrişte Faculty of Applied Sciences, Department of

Mathematics-Informatics, University “Politehnica” of Bucharest, Bucharest,
Romania
Ziji Yu Biostatistics Department, Jazz Pharmaceuticals, Palo Alto, CA, USA
Xiaojun Zhu Department of Mathematics and Statistics, McMaster University
Hamilton, Hamilton, ON, Canada
Part I
Mathematics
Chapter 1
Hole: An Emerging Character in the Story
of Radio k-Coloring Problem
Ushnish Sarkar and Avishek Adhikari
Abstract Frequency assignment problem (FAP) of radio networks is a very active

area of research. In this chapter, radio networks have been studied in a graph-theoretic
approach where the base stations of a cellular network are vertices and two vertices
are adjacent if the corresponding stations transmit or receive broadcast of each other.
Here, we deal with the problem of allocating frequencies to the base stations of the
cellular network such that interference between stations at different distances can
be avoided and at the same time, and the range of distinct frequencies used can
be kept minimum. A certain variant of FAP on such network is radio k-coloring
problem (k ≥ 2 being a positive integer) where stations (vertices) are assigned
frequencies (colors) in such a way that the frequency difference increases (using k as
a parameter) with the growing proximity of the stations. The focus has been laid on
unused frequencies in the spectrum. These unused frequencies are referred as holes,
and they are found to heavily influence the structure of the network (graph). In this
chapter, we highlight many combinatorial aspects of holes in the context of radio
k-coloring problem and its applicability as well as importance in real life.
1.1 Introduction
1.1.1 Motivation
Channel assignment problem or frequency assignment problem is one of the most

important problems in the design of cellular radio networks. A geographical area
covered under a cellular network is divided into smaller service areas referred as cells.
Every cell has a base station and communications among all the wireless terminals or
U. Sarkar · A. Adhikari (B)

Department of Pure Mathematics, University of Calcutta, 35 Ballygunge Circular Road,
Kolkata 700019, India
e-mail: avishek.adh@gmail.com
U. Sarkar
e-mail: usn.prl@gmail.com
© Springer Nature Singapore Pte Ltd. 2017 3

A. Adhikari et al. (eds.), Mathematical and Statistical Applications in Life Sciences
and Engineering, https://doi.org/10.1007/978-981-10-5370-2_1
4 U. Sarkar and A. Adhikari
the users in these cells are made through their corresponding cell area base stations.
Now the base stations transmit or receive any communication in specified frequencies
which are obtained by dividing the radio spectrum uniformly into disjoint frequency
bands. Suppose a radio receiver is tuned to a signal on channel c0 , broadcast by its
local transmitter (i.e., the closest one). The quality of reception will be degraded if
there is excessive interference from other nearby transmitters. Also, due to the reuse of
the same channel c0 in the vicinity, there will be “co-channel” interference. Moreover,
since in reality, as the signal energy from one channel spills over into its nearby
channel, neither transmitters nor receivers operate exclusively within the frequencies
of their assigned channel. Hence, stations using channels close to c0 are also potential
source of interference. Imposing constraints to channel separations between pair of
transmitters, potentially capable of interference, is a way to maintain signal quality
of the broadcast. Let d(x, y) be the distance between transmitter sites x and y in
some suitable metric and δc (x, y) be the minimum allowed spectral separation of the
channels assigned to sites x and y. Then for any two pairs of transmitter sites (x1 , y1 )
and (x2 , y2 ), if d(x1 , y1 ) ≥ d(x2 , y2 ), then δc (x1 , y1 ) ≤ δc (x2 , y2 ). Thus there is a
monotonic trade-off between distance and spectral separation. Moreover, the physical
assumptions made above imply that the spectral constraints are determined only by
distances, in the sense that if d(x1 , y1 ) = d(x2 , y2 ), then δc (x1 , y1 ) = δc (x2 , y2 ).
In other words, if the distance between two stations is small, the difference in their
assigned channels must be relatively large, whereas two stations at a larger distance
may be assigned channels with a smaller difference. Thus, if the channels assigned
to the stations x and y are f (x) and f (y), respectively, then
| f (x) − f (y)| ≥ l x y , (1.1)
where l x y depends inversely on the distance between x and y. Also, the range of
frequencies, from which the assignment is to done, should be kept as narrow (and
hence economic) as possible keeping in mind the scarcity and dearness of the radio
spectrum.
Now the problem of channel assignment was first interpreted in a graph-theoretic
approach by Hale [19]. Here, we consider the base stations, i.e., the transceivers
in the network as vertices. Two vertices are adjacent if the corresponding stations
can broadcast or receive the transmission of each other, i.e., if their transmission
ranges intersect. Also, as the channels are uniformly spaced in the radio spectrum,
we can consider the channel assignment of networks as an integer coloring or integer
labeling problem of graphs.
Chartrand et al. [8, 9] have introduced the radio k-coloring problem of a simple
finite graph by taking l x y = k + 1 − d(x, y) in Eq. 1.1. Formally, for any positive
integer k ≥ 2, a radio k-coloring L of a finite simple graph G is a mapping L : V →
N ∪ {0} such that for any two vertices u, v in G,
|L(u) − L(v)| ≥ k + 1 − d(u, v) (1.2)

1 Hole: An Emerging Character in the Story of Radio k-Coloring Problem 5
While assigning colors (i.e., frequencies) to the vertices (i.e., transceivers) of a

graph (i.e., radio network) in an economic manner keeping in mind the distance
constraints of the radio k-coloring problem, some integers are found to be left unas-
signed to any vertex. Such integers represent unused frequencies in the radio spectrum
allotted for the network and they will be termed as holes in our subsequent discussion.
These holes are found to exhibit splendid properties. Studying them enables us to
understand various important structural properties of the graph representing the radio
network. Exploiting these properties, we may even construct a larger network whose
frequency assignment (radio k-coloring) can be done without changing frequencies
of the older stations and using the unassigned frequencies of the previous network
for the new stations. Thus, our new network shall not need to procure any additional
frequency. Besides, these holes possess some beautiful combinatorial features which
make them worth studying on their own.
1.1.2 Important Definitions, Notations, and a Brief Survey
First, we have to be acquainted with some important notions. Throughout this chapter,
graphs have been taken as simple and finite. The number of vertices of a graph is
its order. Let G be any graph. Then V (G) and E(G) denote the vertex set and
the edge set of G, respectively. Also, let L be any radio k-coloring of G. Then
(maxv∈V L(v) − min v∈V L(v)) is referred as the span of L, denoted by span(L),
and min L {span(L) : L is a radio k − coloring o f G} is referred as the radio k-
chromatic number of G, denoted by r ck (G). Interestingly, if L be a radio k-coloring
of G, then L + a is the same. Therefore, without loss of generality, we shall assume
min v∈V L(v) = 0, for any radio k-coloring L on G. Any radio k-coloring L on G
with span r ck (G) is referred as r ck (G)-coloring or simply r ck -coloring (when the
underlying graph is fixed).
So far, the radio k-coloring problem has been mostly studied for k = 2, k = 3,
k = diam(G) − 2, k = diam(G) − 1, k = diam(G). The radio k-coloring is
referred as radio coloring, antipodal coloring, and near-antipodal coloring, and the
corresponding radio k-chromatic numbers are known as radio number, antipodal
number, and near-antipodal number of G for k = diam(G), k = diam(G) − 1 and
k = diam(G) − 2, respectively, where diam(G) denotes the diameter of G, i.e., the
maximum distance between any two vertices of G. For k = 2, the problem becomes
the L(2, 1)-coloring problem introduced by Griggs and Yeh [18] and for k = 3,
the problem is often referred as L(3, 2, 1)-coloring. Note that the radio 2-chromatic
number r c2 (G) is also often denoted as λ2,1 (G) and termed as λ2,1 (G)-number or
simply λ2,1 -number. Also, r c2 -coloring of G is referred as λ2,1 -coloring of G. In
fact, in our subsequent discussion, we will mostly use λ2,1 (G) instead of r c2 (G).
For an r ck -coloring L on a graph G = (V, E), let L ik (G) = {v ∈ V |L(v) = i} and
li (G) = |L ik (G)|. We replace L ik (G) by L i and lik (G) by li if there is no confusion
k
regarding G and k. The vertices of L i are represented by vij , 1 ≤ j ≤ li , and if li = 1,

we replace vij by vi . In a r ck -coloring L on G, a color i is referred as a multiple color

if li ≥ 2. Sometimes colors are referred as labels too. If two vertices u and v of G
are adjacent, then we write u ∼ v; otherwise, we write u v. Let G be a graph and
e = uv be an edge of G. If the edge e is deleted, then the resulting graph, which has
same vertex set as that of G, is denoted by G − e. If S ⊂ V (G), then G − S is the
graph obtained by deleting the vertices in S and the edges incident to the vertices
in S.
Compared to other radio k-coloring problems, the L(2, 1)-coloring, the radio
coloring, and the antipodal coloring of graphs have received extensive attention.
A thorough survey of L(2, 1)-coloring may be found in [5]. For k = 2, the radio
k-coloring problem has been studied for comparatively fewer families of graphs
including paths, trees and cycles [7, 22, 25, 28], powers of paths and cycles [26, 29,
36], toroidal grids [37], etc. Saha et al. studied lower bound for radio k-chromatic
number in [38]. Readers may go through [35] for a detailed survey in radio k-coloring
problem.
1.1.3 Introduction to Holes
An interesting difference between general graph coloring and radio k-coloring is

that an optimal general graph coloring uses every color between maximum color
and minimum color, whereas an r ck -coloring may not do the same (see Figs. 1.2 and
1.3). If i is an integer, 0 < i < r ck (G), such that it is not assigned to any vertex of G
as a color by an r ck -coloring L of G, then i is called a hole of L. Any r ck -coloring
has at most (k − 1) consecutive holes [39]. The minimum number of occurrences
of (k − 1) consecutive holes in any r ck -coloring on G is said to be the (k − 1)-hole
index of G and denote by ρk (G) or simply ρk if there is no confusion regarding
the graph [40]. Also, exploiting the maximum number of consecutive holes in any
r ck -coloring on G, the notion of (k − 1)-hole has been defined in [40]. A (k − 1)-hole
in a r ck -coloring L on a graph G is a sequence of (k − 1) consecutive holes in L. If
L has (k − 1) consecutive holes i + 1, i + 2, . . . , i + k − 1, then the corresponding
(k − 1)-hole is denoted by {i + 1, i + 2, . . . , i + k − 1}. Clearly then, the minimum
number of (k − 1)-holes in any r ck -coloring on G is the (k − 1)-hole index of G. We
refer the collection of all r ck -colorings on G with ρk (G) number of (k − 1)-holes
as Λρk (G). Moreover, the minimum span of a radio k-coloring on G with at most
(k − 2) consecutive holes is defined as max-(k − 2)-hole span of G and denoted
by μk (G). Note that the 1-hole index and 2-hole index of the graph in Fig. 1.1 is 2.
Again, the 1-hole index and 2-hole index of the same graph are zero as shown in
Figs. 1.2 and 1.3.
Now, an r c2 -coloring, i.e., a λ2,1 -coloring does not have two consecutive holes.
Clearly then, ρ2 (G) is the minimum number of holes in any λ2,1 -coloring on G which
is often referred as hole index of G, denoted by ρ(G). Thus, the 1-hole index and
hole index of G are same. Also, Λρ2 (G) is referred as Λρ (G). Moreover, the max-0-
hole span of G is referred as no-hole span of G, and μ2 (G) is alternatively written
L(2, 1): 0 3 L(3, 2, 1): 0 6
2 8 4 14
4 6 8 11
1 2
Fig. 1.1 Example of graph with positive 1-hole index and 2-hole index
2 4
0 1
No hole rc 2−coloring
5
3
2
0
2
5
0 Set of holes={4}
3
1
1
L(2,1)−coloring
Fig. 1.2 Examples of r c2 -colorings with and without holes
6
4
2 9
0 9
1 3
2 4 0
6
3 7
Set of holes={5,7,8} Set of holes={1,5,8}
L(3,2,1)−coloring/ radio coloring
Fig. 1.3 Examples of r c3 -colorings/radio colorings with holes

G1 G2 G3
Fig. 1.4 Examples of non-full colorable graphs
Fig. 1.5 One more example

of non-full colorable graphs
as simply μ(G). Note that r ck (G) ≤ μk (G) and in particular, λ2,1 (G) ≤ μ(G). If
λ2,1 (G) = μ(G), then the graph G admits a λ2,1 -coloring with no hole. Such λ2,1 -
coloring of G is said to be no hole λ2,1 -coloring or consecutive λ2,1 -coloring of G
and the graph G is said to be a full colorable graph.
It has been shown in [13] that the following graphs are full colorable:
(i) any connected graph G with Δ(G) = 2 except P3 , C3 , C4 and C6 ;
(ii) any tree T with Δ(T ) ≥ 3 except stars;
(iii) any connected non-tree graph G of order at least six with Δ(G) = 3 and
λ2,1 (G) = 4 except the graphs in the Fig. 1.4;
(iv) any connected non-tree graph G of order either 7 or 8 with Δ(G) = 4 and
λ2,1 (G) = 5 except the graph in the Fig. 1.5.
Several interesting combinatorial properties of holes in a λ2,1 -coloring as well as
existence of full colorable graphs have been studied in [1, 13, 16, 30, 31]. Recently,
holes in a r ck -coloring has been studied in [40], and several interesting properties
have been explored. In fact, the study of holes in an r ck -coloring on a graph is an
emerging topic in literature. However, this problem is comparatively more explored
for r c2 -coloring, i.e., a λ2,1 -coloring of a graph than for general k.
1.2 k-Islands and an Equivalence Relation on Λρk (G)
We now define k-islands and using this notion, we will define an equivalence relation
on Λρk (G). These notions will be very helpful in our subsequent discussion. Based
on the fact that an r ck -coloring of a graph G may have at most (k − 1) consecutive
holes [39], an Ω-set and a k-island have been defined in [40]. Let L ∈ Λρk (G). An
Ω-set of L is a non-empty set A of non-negative integers, assigned by L as colors,
such that s ∈ A only if 0 ≤ |s − s | < k, for some s ∈ A. A maximal Ω-set of
L ∈ Λρk (G) is a k-island or simply an island of L, if there is no confusion regarding
k. The minimum and maximum element of an island I are said to be the left and
right coast of I , respectively, denoted by lc(I ) and r c(I ) accordingly. The left and
right coasts of I are together called the coastal colors or coastal labels of I . If the
set of coastal colors of some island is a singleton set, then the island is said to be an
atoll.
An Equivalence Relation on Λρk (G):
First, we discuss a recoloring of an r ck -coloring in Λρk (G), referred as an
α-recoloring (see [40]), which eventually will induce an equivalence relation on
Λρk (G). Let L ∈ Λρk (G) and I0 , I1 , . . . , Iρk be the islands of L such that lc(I j+1 ) =
r c(I j ) + k, 0 ≤ j ≤ ρk − 1. For any i, 0 ≤ i ≤ ρk , and m, 0 ≤ m ≤ (ρk − i), let us
define a new radio k-coloring L̂ on G as follows:

L(u), if L(u) ∈/ i+m
s=i Is ,
L̂(u) = i+m
r c(Ii+m ) − t, if L(u) ∈ s=i Is and L(u) = lc(Ii ) + t, t ≥ 0.
It is easy to verify that L̂ ∈ Λρk (G) and the islands I0 , I1 , . . . , Iρk of L̂ are such that
I j = I j , for j ∈
/ {i, i +1, . . . , i +m} and r c(Ii+m−t ) = r c(Ii+m )−(lc(Ii+t )−lc(Ii )),
lc(Ii+m−t ) = r c(Ii+m ) − (r c(Ii+t ) − lc(Ii )), for 0 ≤ t ≤ m. This recoloring L̂ is an
α-recoloring of L.
Let η be a relation on Λρk (G) defined by L 1 ηL 2 if and only if L 2 is obtained from
L 1 using of finite α-recolorings. It is easy to verify that η is reflexive, symmetric,
and transitive. Thus, η is an equivalence relation on Λρk (G). Therefore, any r ck -
coloring in Λρk (G) is η-related to some L ∈ Λρk (G) with islands I0 , I1 , . . . , Iρk
such that lc(I j+1 ) = r c(I j ) + k, 0 ≤ j ≤ ρk − 1 and |I0 | ≤ |I1 | ≤ · · · ≤ |Iρk |.
Thus, we may assume, without loss of generality, that for any L ∈ Λρk (G), if the
islands of L are I0 , I1 , . . . , Iρk with lc(I j+1 ) = r c(I j ) + k, 0 ≤ j ≤ ρk − 1, then
|I0 | ≤ |I1 | ≤ · · · ≤ |Iρk |. We refer this finite sequence (|I0 |, |I1 |, . . . , |Iρk |) as the
k-island sequence of L or simply the island sequence of L, if there is no confusion
regarding k. For the graph G in Fig. 1.1, the λ2,1 -coloring in Λρ2 (G), i.e., Λρ (G), has
2-islands I0 = {0, 1, 2, 3, 4}, I1 = {6}, I2 = {8}, and the r c3 -coloring in Λρ3 (G) has
3-islands I0 = {0, 2, 4, 6, 8}, I1 = {11}, I2 = {14}. Hence, (1, 1, 5) is a 2-island
sequence as well as a 3-island sequence of G.
1.3 An Upper Bound of ρk (G)nd Related Issues
In this section, an upper bound of ρk (G) has been given. Moreover, various interesting
features regarding order, regularity (i.e., each vertex having the same degree), radio
k-chromatic number, domination number as well as perfect domination number of
G (defined later) can be found when ρk(G) attains its upper bound. We will start with
the following results which will give a structural insight of a graph G in presence of
a (k − 1)-hole of an r ck -coloring in Λρk (G). But first, we need this definition. Let G
and H are two graphs. Then G + H is the disjoint union of G and H having vertex
set V (G) ∪ V (H ) and edge set E(G) ∪ E(H ). Thus, mG is the disjoint union of m
pairwise disjoint copies of G.
Lemma 1.1 [40] Let G be a graph with ρk (G) ≥ 1 and L ∈ Λρk (G). If {i + 1, i +
2, . . . , i + k − 1} is a (k − 1)-hole in L, then li = li+k and the subgraph of G induced
by L i ∪ L i+k is li K 2 .
Proof Case I: Let li+k ≥ 2. If possible, let x ∈ L i+k be such that x v, where v is
an arbitrary vertex in L i . We define L̂ by

L(u), if u = x,
L̂(u) =
i + k − 1, if u = x.
Clearly, L̂ is an r ck -coloring with fewer (k − 1)-holes, leading to a contradiction.

Case II: Let li+k = 1 and L i+k = {x}. If possible, let x v, where v is an
arbitrary vertex in L i . We define L̂ by

L(u), if L(u) ≤ i,
L̂(u) =
L(u) − 1, if L(u) ≥ i + k.
Then L̂ is a radio k-coloring with span r ck (G) − 1, a contradiction.

Hence, every vertex of L i+k is adjacent to a unique vertex in L i . Similarly, any
vertex L i is adjacent to a unique vertex in L i+k . From this, the proof follows.
The following result is going to play a key role in our subsequent discussion.
Corollary 1.1 [40] Let G be a graph with ρk (G) ≥ 1 and I, J are two distinct
islands of L ∈ Λρk (G) with x and y as two coastal colors of I and J , respectively.
Then l x = l y and the subgraph of G induced by L x ∪ L y is l x K 2 .
Proof Applying finite number of suitable α-recolorings, we assume, without loss

of generality, that x = r c(I ), y = lc(J ) and y = x + k. The proof follows from
Lemma 1.1.
We are now in a position to obtain an upper bound of ρk (G).
Theorem 1.1 [40] Let G be a graph with ρk (G) ≥ 1. Then ρk (G) ≤ Δ(G), where
Δ(G) is the maximum degree of any vertex in G.
Proof Let I0 , I1 , . . . , Iρk be the islands of any L ∈ Λρk (G). Let x be any coastal
color of I0 and u ∈ L x . Then by Corollary 1.1, u ∼ v y , where v y ∈ L y in G,
for each coastal color y of I j , 1 ≤ j ≤ ρk (G). Hence, d(u) ≥ ρk (G). Therefore,
ρk (G) ≤ Δ(G).
When ρk (G) attains its upper bound, G shows some interesting structural proper-
ties. Before we discuss the situation, let us take a look at few definitions. In a graph
G, a set S ⊆ V (G) is a dominating set if for every vertex u ∈ V (G) \ S, there is

a v ∈ S such that u ∼ v (see [42]). The minimum order of a dominating set is the
dominating number of G, denoted by γ(G). A set S ⊂ V (G) is said to be a perfect
dominating set if every vertex in V (G) \ S is adjacent to exactly one vertex in S
(see [20]). The minimum cardinality of a perfect dominating set of G is the perfect
domination number of G, denoted by γ p (G). Clearly γ(G) ≤ γ p (G).
Now, we state the following lower bound for domination number. It will help us
to prove Theorem 1.3.
Theorem 1.2 [20, 41] For any graph G of order n, γ(G) ≥ n

Δ(G)+1
.
Now, we explore the structure of G, including its regularity, order, domination

number, etc., when ρk (G) = Δ(G).
Theorem 1.3 [40] Let G be a graph of order n with ρk (G) ≥ 1 and ρk (G) =
Δ(G) = Δ. Then
(i) G is a Δ-regular graph;

(ii) r ck (G) = kΔ;
(iii) n ≡ 0 (mod Δ + 1);
(iv) γ(G) = γ p (G) = Δ+1
n
;
(v) If n = Δ + 1, then μk (G) = r ck (G) + 1.
Proof Let I0 , I1 , . . . , Iρk be the islands of any L ∈ Λρk (G) with lc(I j+1 ) = r c(I j ) +
k, 0 ≤ j ≤ ρk − 1 such that |I0 | ≤ |I1 | ≤ · · · ≤ |Iρk |. Then the number of islands of
L is ρk + 1.
We claim that every island of L is an atoll. If not, then there exists an island I
of L such that lc(I ) = r c(I ). Then by Corollary 1.1, for any vertex v with L(v)
as a coastal color of any other island J of L, d(v) ≥ ρk + 1 = Δ + 1, which is a
contradiction. Hence, every island is a singleton set, i.e., an atoll. Thus, for every
vertex v of G, L(v) is a coastal color of some island of L and therefore every vertex
has same degree, by Corollary 1.1. Hence, G is Δ-regular.
Since each island of L is atoll, I j = {k j}, 0 ≤ j ≤ Δ. Hence, r ck (G) = kΔ.
Moreover, by Corollary 1.1, lki = lk j = l (say), 0 ≤ i < j ≤ Δ. Therefore,
n = l(Δ + 1), i.e., n ≡ 0 (mod Δ + 1).
Again, by Corollary 1.1, the subgraph induced by L ki ∪ L k j is l K 2 , for any i, j
with 0 ≤ i < j ≤ Δ. Also, Ii = {ki}, 0 ≤ i ≤ Δ. Therefore, L ki is a dominating set
as well as a perfect dominating set of G, for every i, 0 ≤ i ≤ Δ. Hence, γ(G) ≤ Δ+1 n
.
Using Theorem 1.2, we get γ(G) = Δ+1 . Hence, γ p (G) = Δ+1 .
n n
Let n = Δ + 1. Then l ≥ 2. Since lc(I j ) = r c(I j ), for every 0 ≤ j ≤ Δ, using

Corollary 1.1, we get a path P : v0 , v1 , . . . , vΔ in G such that L(v j ) ∈ I j = {k j},
0 ≤ j ≤ Δ. Now, the new coloring L̂ on G defined by

L(u), if u = v j , 0 ≤ j ≤ Δ,
L̂(u) =
L(u) + 1, if u = v j , 0 ≤ j ≤ Δ.
is a radio k-coloring of G with no (k − 1)-hole. Hence, μk (G) ≤ span( L̂) =

r ck (G)+1. But since ρk (G) > 0, μk (G) > r ck (G). Therefore, μk (G) = r ck (G)+1.
1.4 A New Family of Graphs: GΔ,t
Things become very interesting when ρ2 (G) = Δ(G), i.e., ρ(G) = Δ(G). By
Theorem 1.3, the order of G is t (Δ + 1), for some integer t ≥ 1. Let GΔ,t be the
family of connected graphs G with ρ(G) = Δ(G) = Δ and order t (Δ + 1), for some
integer t ≥ 1. Then using Theorem 1.3, li = t, for i = 0, 2, 4, . . . , λ2,1 (G) = 2Δ.
Note that GΔ,1 = {K Δ+1 }.
1.4.1 A Characterization of GΔ,2
We now focus on t = 2. For any Δ ≥ 2, let G be the bipartite graph obtained by

deleting a perfect matching from K Δ+1,Δ+1 , where a perfect matching is a set M of
edges such that no two edges in M share a vertex, and each vertex of the graph is
incident to exactly one edge in M. Then G is connected Δ-regular graph with order
2(Δ+1). It can be easily verified that G ∈ GΔ,2 . Now, the question is whether every
connected Δ-regular graph with order 2(Δ + 1) belongs to GΔ,2 .
Before we answer this question, we will state the following result which will
be helpful in the subsequent discussion. Note that a cycle through all the vertices
of a graph is a Hamiltonian cycle and any graph having a Hamiltonian cycle is a
Hamiltonian graph. Also, any path through all the vertices of a graph is a Hamiltonian
path in the graph. Clearly, a Hamiltonian graph contains a Hamiltonian path.
Theorem 1.4 [42] Let G be a graph on n ≥ 3 vertices having minimum degree

δ(G) ≥ n2 . Then G is a Hamiltonian graph.
Now we are in a position to prove the following result.
Lemma 1.2 [16] If G is a connected Δ-regular graph of order 2(Δ + 1), then
G ∈ GΔ,2 or λ2,1 (G) = 2Δ + 1.
Sketch of proof. Clearly, G c is a (Δ + 1) regular graph of order 2(Δ + 1). As δ(G c ) =

1
2
|V (G c )|, by Theorem 1.4, G c has a Hamiltonian path, say, P : v0 , v1 , . . . , v2Δ+1 .
Hence by assigning color i to vi , 0 ≤ i ≤ 2Δ + 1, we get λ2,1 (G) ≤ 2Δ + 1.
We shall now show that if λ2,1 (G) ≤ 2Δ, then G ∈ GΔ,2 . For this, using the
distance condition of L(2, 1)-coloring and regularity as well as order of G, we would
show that if α is a multiple color of any λ2,1 -coloring L, then lα = 2 and α − 1,
λ2,1 (G)
α + 1 are two holes in L which in turn will help us to show |V (G)| = Σi=0 mi ≤
λ2,1 (G)+2
2 2 , implying λ2,1 (G) ≥ 2Δ. Hence, λ2,1 (G) = 2Δ and using this we can
show ρ(G) = Δ.
Fig. 1.6 Illustrating an vs,1 vr,2

example of e-exchange
vr,1 vs,2
Remark 1.1 The above result says about the radio 2-chromatic number of G if G is
a connected Δ-regular graph of order 2(Δ + 1) and G ∈/ GΔ,2 .
We now discuss a graph construction (introduced in [16]) which will be helpful

in characterizing GΔ,2 .
The S-Exchange of Sum of Two Graphs:
Let G 1 , G 2 be two graphs both isomorphic to G with V (G) = {v1 , v2 , . . . , vn }
and φi (i = 1, 2) be graph isomorphisms from G to G i , where φi (v j ) = v j,i . Also
let e = vr vs be an edge of G. Let T (e) = {vr,1 vs,2 , vr,2 vs,1 }. Then the e-exchange of
graph G 1 +G 2 , denoted by X e (G 1 +G 2 ), is the graph with the vertex set V (G 1 +G 2 )
and edge set (E(G 1 + G 2 ) − {φ1 (e), φ2 (e)}) ∪ T (e).
Let S ⊆ E(G). Then the S-exchange of G 1 + G 2 , denoted by X S (G 1 + G 2 ),
is the graph with vertex set V (G 1 + G 2 ) and the edge set (E(G 1 + G 2 ) −
∪e∈S {φ1 (e), φ2 (e)}) ∪ (∪e∈S T (e)), where for each e, T (e) is defined earlier.
For example, if G is isomorphic to the complete graph K 3 and S = E(G), then
X S (G 1 + G 2 ) is isomorphic to the cycle C6 . Also if G is isomorphic to the complete
graph K 4 and e = vr vs is any edge of G, then X e (G 1 + G 2 ) is isomorphic to the
graph shown in Fig. 1.6.
We now state the following theorem which gives a characterization of GΔ,2 in
terms of S-exchange.
Theorem 1.5 [16] Let G be a connected Δ-regular graph of order 2(Δ + 1). Then
G ∈ GΔ,2 if and only if there is a set S ⊆ E(K Δ+1 ) such that G is isomorphic to
X S (K Δ+1 + K Δ+1 ).
1.4.2 GΔ,t and the t-Neighborhood Property
We shall now address the more general issue of connected Δ-regular graphs having
order t (Δ + 1), for any t ≥ 2. Král’ et al. [24] identified a special property of
connected Δ-regular graph with order t (Δ + 1). A connected Δ-regular graph with
order t (Δ + 1) is said to have the t-neighborhood property if for any two disjoint
sets U and W of vertices of G, the following holds: If no vertex of U is adjacent to
any vertex of W and any pair of vertices in U or W are at least three distance apart,
then |U | + |W | ≤ t.
We now have the following result.
Fig. 1.7 Illustrating H4

u/ u u
1 1 2
v/ v v
1 1 2
Theorem 1.6 [24] Let G be a connected Δ-regular graph of order t (Δ + 1) with

λ2,1 (G) ≤ 2Δ such that G satisfies the t-neighborhood property. Then λ2,1 (G) = 2Δ
and ρ(G) = Δ and hence G ∈ GΔ,t .
Sketch of proof. Clearly, there is an L(2, 1)-coloring of G with span at most 2Δ.
Using the t-neighborhood property of G, m i + m i+1 ≤ t. For any odd integer i 0 ,
i 0 −1
1 ≤ i 0 ≤ 2Δ, t (Δ + 1) = Σi=0
2Δ
m i = (Σi=0
2
m 2i + m 2i+1 ) + (Σ Δ i0 +1 m 2i−1 + m 2i ) −
i= 2
m i0 ≤ t (Δ + 1) − m i0 . Therefore, m i0 = 0, for every odd integer i 0 , 1 ≤ i 0 ≤ 2Δ.
So m 2i = t, 0 ≤ i ≤ Δ. From this, the proof follows.
Thus, Theorem 1.6 says that if G is a connected Δ-regular graph of order t (Δ+1)
with λ2,1 (G) ≤ 2Δ, then satisfying the t-neighborhood property is a sufficient
condition for obtaining λ2,1 (G) = 2Δ and ρ(G) = Δ. We now show that the
condition is not necessary. In fact, we will prove that there is a connected Δ-regular
graph of order 3(Δ + 1) which does not satisfy the 3-neighborhood property and for
which λ2,1 (G) = 2Δ and ρ(G) = Δ.
Before we prove this, we shall state the following lemmas whose proofs follow
easily.
Lemma 1.3 [31] Let G be a graph isomorphic to K m+1 − e, where the edge e = uv.
Then λ2,1 (G) = 2m −1 and for any λ2,1 -coloring L of G, {L(u), L(v)} = {2i, 2i +1}
for some i ∈ {0, 1, . . . , m − 1}.
Lemma 1.4 [18] Let H be a subgraph of G. Then λ2,1 (H ) ≤ λ2,1 (G).
Now, we construct a graph Hm (m ≥ 3) consisting of two components. First one is
K m+1 − {e1 , e2 }, where e1 = u 1 v1 and e2 = u 1 v1 are two non-incident edges (called
the left component of Hm ); the other component of K m+1 −e, where e = u 2 v2 (called
the right component of Hm ). Now, we add two edges u 1 u 2 and v1 v2 (See Fig. 1.7).
We prove the following result.
Lemma 1.5 [31] λ2,1 (Hm ) = 2m, for m ≥ 3.
Proof Label u 1 , v1 , u 2 , v2 , u 1 , v1 by 0, 2, 2, 0, 2m − 2, 2m, respectively, and the
remaining (m − 3) vertices in the left component by 4, 6, . . . , 2m − 4 and the
remaining (m−1) vertices in the right component by 4, 6, . . . , 2m. This is an L(2, 1)-
coloring of Hm of span 2m and so λ2,1 (Hm ) ≤ 2m.
If possible, let λ2,1 (Hm ) < 2m. Since Hm has K m+1 − u 2 v2 as its subgraph
(the right component), so by Lemma 1.4, λ2,1 (Hm ) = 2m − 1. Let L be a λ2,1 -
coloring of Hm . Without loss of generality, we can assume that L(u 2 ) = 2i and
Fig. 1.8 Illustrating G 4

u u 1/ u u
3 1 2
v v/ v v
3 1 1 2
L(v2 ) = 2i + 1, by Lemma 1.3. Then the vertices in the right component of Hm are
colored with 0, 2, 4, . . . , 2i, 2i +1, 2i +3, . . . , 2m −1. Therefore, {L(u 1 ), L(v1 )} ⊆
{1, 3, 5, . . . , 2i − 1, 2i + 2, 2i + 4, . . . , 2m − 2} as L satisfies the distance two
condition. Hence, |L(u 1 ) − L(v1 )| ≥ 2. Therefore, {L(u 1 ), L(v1 )} = {2 j, 2 j + 1},
for some j ∈ {0, 1, . . . , m − 1} and so the vertices in the left component are colored
by 0, 2, 4, . . . , 2 j, 2 j + 1, 2 j + 3, . . . , 2m − 1. But as i = j and L(u 2 ) = 2i,
L(v2 ) = 2i + 1, so this leads to a contradiction. Hence, λ2,1 (Hm ) = 2m.
We now construct a graph G m which consists of three components: one is K m+1 −
{e1 , e2 }, where e1 = u 1 v1 and e2 = u 1 v1 are two non-incident edges (called middle
component); the other two components are K m+1 − ei , where ei = u i vi , i = 2, 3
(called left and right components, respectively). Now, add edges u 1 u 2 , v1 v2 , u 1 u 3
and v1 v3 (See Fig. 1.8).
Lemma 1.6 [31] λ2,1 (G m ) = 2m, for m ≥ 3.
Proof Since Hm is a subgraph of G m , λ2,1 (G m ) ≥ 2m, by Lemmas 1.4 and 1.5.
Again, label the vertices u 1 , v1 , u 1 , v1 , u 2 , v2 , u 3 , v3 with 0, 2, 2m − 2, 2m, 2, 0, 2m,
2m − 2, respectively; the remaining (m − 1) vertices of the left (right, respectively)
component by distinct colors from the set {0, 2, . . . , 2m −4} ({4, 6, . . . , 2m}, respec-
tively). Lastly, label the remaining (m − 3) vertices in the middle component by the
distinct colors from {4, 6, . . . , 2m − 4}. This is an L(2, 1)-coloring of G m and hence
λ2,1 (G m ) = 2m.
Now we are in a position to discuss the following result.
Theorem 1.7 [31] For each integer m ≥ 3, there exists a connected m-regular graph
G of order 3(m + 1) without 3-neighborhood property such that λ2,1 (G) = 2m and
ρ(G) = m.
Sketch of proof. Clearly G m is a connected m-regular graph of order 3(Δ + 1).
By Lemma 1.6, λ2,1 (G m ) = 2m. Consider the set of vertices U = {u 2 , u 3 } and
W = {v2 , v3 } in G m . Clearly, each pair of vertices of U or W is three distance apart,
and no vertex of U is adjacent to any vertex of W . Moreover, |U | + |W | = 4 > 3.
Hence, G m does not satisfy the 3-neighborhood property.
Now, it can be shown (see [31]) that for any λ2,1 -coloring L of G m , if u, v are any
two vertices in the same component, i.e., in the middle or left or right component,
of G m , then L(u), L(v) cannot be consecutive integers. Hence, 1, 3, 5, . . . , 2m − 1
are holes in L. So, ρ(G m ) = m.
a0 a
1
b 0,0 b 1,0 b 2,0
b0,2 b0,1 b b b b 2,1

1,2 1,1 2,2
a2
Fig. 1.9 Illustrating Ω3
1.4.3 A Conjecture on G r,t
In this subsection, we study a conjecture regarding Gr,t and its disproof. But first we
need to understand the backdrop in which the conjecture shall find its relevance. For
this, we will study the following two families of connected r -regular graphs.
We begin with a class Ωr of connected r -regular graphs, for any r ≥ 1, introduced
in [16]. Let X = r K r and Y = r K 1 . Let V (Ωr ) = V (X ) ∪ V (Y ), where
−1
(i) V (X ) = ∪ri=0 Bi , Bi = {bi, j | 0 ≤ j ≤ r − 1} and (ii) V (Y ) =
{a0 , a1 , . . . , ar −1 }.
−1
Let E(Ωr ) = R ∪ S where (i) R = ∪ri=0 Ri , Ri = {bi, j bi,k : 0 ≤ j < k ≤ r − 1}
r −1
and (ii) S = ∪i=0 Si , Si = {ai bm,i : 0 ≤ m ≤ r − 1}. Clearly, Ωr is a connected
r -regular graph with order r (r + 1) (See Fig. 1.9).
It can be easily verified that Ω1 and Ω2 are isomorphic to K 2 and C6 , respectively.
We now observe the following facts regarding distances in Ωr .
For 0 ≤ i < j ≤ r − 1, d(ai , a j ) = 3.
For 0 ≤ i, j, k, l ≤ r − 1,
⎧
⎨ 1, if i = k and j = l,
d(bi, j , bk,l ) = 2, if i = k and j = l,
⎩
3, otherwise;
Also for 0 ≤ i, j, k ≤ r − 1,

1, if i = k,
d(ai , b j,k ) =
2, otherwise;
Using these facts, we can say that Ωr satisfies the r -neighborhood property.
Now let Bk = {bi, j | ( j − i) ≡ k (mod r )}, 0 ≤ k ≤ r − 1. Clearly, |Bk | = r
and each pair of vertices of Bk and Y are at distance three. Hence, we can assign an
L(2, 1)-coloring on Ωr as follows:
Fig. 1.10 Illustrating Ω4,2 a0 a1
b b b
0,0 b 1,0 1,1
0,1
b b0,2 b b
0,3 1,3 1,2

2k, if v ∈ Bk ,
L(v) =
2r, if v = ai , where i = 0, 1, . . . , r − 1.
Hence, λ2,1 (Ωr ) ≤ 2r . Therefore by Theorem 1.6, Ωr ∈ Gr,r , for every r ≥ 1.

Thus, for every r ≥ 1, Gr,r is non-empty.
Given any r ≥ 2, to show that Gr,t is non-empty for every t < r , Georges
et al. constructed another graph Ωr,t in [16]. Let X = t K r and Y = t K 1 . Then
V (Ωr,t ) = V (X ) ∪ V (Y ), where
(i) V (X ) = ∪i=0
t−1
Bi , Bi = {bi, j | 0 ≤ j ≤ r −1} and (ii) V (Y ) = {a0 , a1 , . . . , at−1 }.
Also E(Ωr,t ) = R ∪ S ∪ T , where (i) R = ∪i=0 t−1
Ri , Ri = {bi, j bi,k : 0 ≤ j < k ≤
r − 1}, (ii) S = ∪i=0 Si , Si = {ai bm,i : 0 ≤ m ≤ t − 1} and (iii) T = ∪i=0
t−1 t−1
Ti ,
Ti = {ai bi, j : t ≤ j ≤ r − 1} (See Fig. 1.10). Clearly Ωr,t is a connected r -regular
graph with order t (r + 1). It can be easily verified that Ω2,1 is isomorphic to K 3 and
in general Ωr,1 is isomorphic to K r +1 . Argument similar to the analysis of Ωr can
show that Ωr,t ∈ Gr,t .
Georges et al. were unable to prove that Gr,t is non-empty for t > r and at this point
they conjectured in [16] that Gr,t = ∅, for every t > r . This conjecture was proven
to be wrong by Král’ et al. in [24]. For this, they constructed an (α + β − 1)-regular
connected graph α,β of order αβ(α + β) with ρ(α,β ) = Δ(α,β ) = α + β − 1.
Now, V (α,β ) consists of two sets Vg = {[a, b, ā] : 1 ≤ a ≤ α, 1 ≤ b ≤ β and 1 ≤
ā ≤ α} and Vr = {[a, b, b̄] : 1 ≤ a ≤ α, 1 ≤ b ≤ β and 1 ≤ b̄ ≤ β}. The
vertices of Vg and Vr are called green and red vertices, respectively. Two distinct
green vertices [a, b, ā] and [a , b , ā ] are adjacent if and only if b = b and ā = ā .
Similarly, two distinct red vertices [a, b, b̄] and [a , b , b̄ ] are adjacent if and only
if a = a and b̄ = b̄ . A green vertex [a, b, ā] is adjacent to a red vertex [a , b , b̄ ]
if and only if a = a and b = b (Fig. 1.11).
We now prove the following theorem.
Theorem 1.8 [24] For every α, β ≥ 1, λ2,1 (α,β ) ≤ 2(α + β − 1).
Proof For i = 0, 1, . . . , α − 1, define Ai = {u = [a, b, ā] ∈ Vg : i ≡ a +

ā (mod α)}. Also, for j = 0, 1, . . . , β − 1, define B j = {v = [a, b, b̄] ∈ Vr : j ≡
b + b̄ (mod β)}. Then Ai and B j are independent sets for all i and j. Also, Vg =
α−1 β−1
∪i=0 Ai and Vr = ∪ j=0 B j . Moreover, we claim that any two vertices in an Ai or
B j are at least three distances apart. If not, then without loss of generality, we may
assume that two green vertices u = [a, b, ā] and v = [a , b , ā ] in Ai are two
distance apart. Let the common neighbor of u and v be a green vertex. Then b = b
Another random document with
no related content on Scribd:
and numbers of people we should like to meet here. You
remember he had come back invalided from the Soudan, and
though he was quite young, about thirty, he had gained great
distinction. He was Major Caruth, I think, and his name must
have been Alexander, for his mother called him 'Alec.'
Everyone liked them both, but we used to think him just a
little reserved."
"I thought him extremely polite—quite a model of courtesy, in

fact."
"Well, yes, he was, and especially to the elder ladies; but he

never showed any marked attention to any of the younger
ones. He was the most devoted son possible, and it was quite
beautiful to see the manner in which he looked up in his
mother's face when she came to his side with that inquiring
glance on hers."
"He had nearly died, and he was all she had," replied Mrs.
Evans. "Heir to a fine property, I believe. I scarcely think that
Mrs. Caruth could be the one Joyce mentioned. Was it likely
there would be any intimacy between the daughter of a poor
country clergyman and people of position like those Caruths?"
"I do not know. You see, clergymen go everywhere."
"But not always their wives and daughters," said Mrs. Evans.
"Did you notice the name of the place Joyce's friend lived at? I
have the address of those we met at Mentone; it was Ferns—
something—crag, probably."
"Was it Fernsclough?" said Augusta, eagerly.
"I really believe it was."
"Then the lady is the same. Her place is Fernsclough, Salop."

"Well, what of that? Her giving Joyce a character to go out as
children's maid puts away the suggestion of intimacy at once.
She might do that, and never speak to or communicate again
with one who was disgracing herself by taking a sort of
servant's place."
"Joyce said that Mrs. Caruth wanted her to go to Fernsclough

for an indefinite time."
"Perhaps that was an invention, in order to raise herself in our

eyes, my dear child. I have seen more of life and character
than you have, Augusta."
"I can hardly think that," replied the girl; "I could not imagine
Joyce saying an untrue word. She is not that sort of girl. And,
mamma, she is my cousin and a lady, though she is not rich. I
cannot help feeling sorry for her. If these friends of hers
should turn out to be the Caruths we met, and at some future
time we should see them again, what will they think of us for
letting Joyce go?"
"Think, you foolish girl! What can they think? Just that as she
was too proud to go to Fernsclough, she was too headstrong
to be guided by us, and went her own wilful way. You need not
trouble your head about that."
But Augusta was not quite happy, in spite of her mother's

assurances; and Adelaide was still less so.
CHAPTER IV.
THOUGH Joyce Mirlees' twenty-first birthday brought some
clouds and storms, it was not wholly without peace and
brightness. More than a dozen letters reached her from
various quarters. Her uncle did not forget Joyce, but wrote
warmly and lovingly, and promised to be at The Chase before
she left it.
Other letters were from old friends at Welton, who did not fail
to send birthday greetings and simple gifts to their former
pastor's daughter. One packet, containing some beautiful
fancy articles, came to her from her Sunday scholars, who had
worked them for the dear teacher whose absence they
regretted more and more, they said. Yet it was plain that one
and all pictured Joyce amongst loving kinsfolk, and amid
luxuries of every kind, for they seemed half afraid that their
simple tokens of love would look very poor and mean amongst
her birthday gifts in her new and splendid home.
If those who had bestowed such patient labour on the dainty

articles could have seen how Joyce looked at them through
gathering tears, but with a glad face, and heard her soft
whisper, "Not alone in the world. Not forgotten, though
absent, thank God!" they would have been more than repaid.
The very answering of these gave Joyce happy employment

during the afternoon. Besides, she had not been without
personal greetings. The very servants at The Chase had
learned to love their master's orphan niece, who spoke gently,
and thought of and for them, as they went about their daily
duties. They ventured to offer good wishes, and one little
country girl begged her to accept a pin-cushion which she had
risen earlier to make for Miss Joyce.
There were loving words, too, from Sarah Keene, who

alternately rejoiced and wept over her nursling, bewailing her
coming departure with one breath, and expressing her firm
conviction in the next, that it would be overruled for good,
and that her darling would be above all of them yet.
There was one more letter not named hitherto, which, though
full of kindness, brought some disappointment. The writer,
Mrs. Caruth, said all that could be expected from an old and
true friend. But there was no other message, though she
mentioned casually that her son, being quite well, had
rejoined his regiment instead of availing himself of the longer
leave at his disposal.
It was still early evening, and Joyce was in her own room,
when she heard a light tap at the door, and the words, "May I
come in, Cousin Joyce?"
The voice was Adelaide's, but the tone of it was so different

from her ordinary one that Joyce could hardly believe her
ears. She, however, opened the door and convinced herself
that her visitor was indeed Adelaide, the elder and much more
beautiful of her two handsome cousins. She also somewhat
resembled Mr. Evans in disposition; but, like him, had rarely
courage to express her sentiments when they differed from
those of her mother and sister.
"May I come in?" she repeated, as she hesitated on the

threshold of Joyce's room.
"Certainly. I am glad, very glad, to have you."
"That is kind, Cousin Joyce; kinder than I deserve. I am come

to make a confession, Joyce; I have been very unkind to you.
Will you forgive me?"
"I do not understand. You have done nothing," said Joyce,

amazed at the visit, words, and look of her cousin, who had
taken her hand, and was holding it between both her own.
"Perhaps I have not done much, after all," she said; "but one
has often as much cause to grieve for the not doing what is
right and kind as for active unkindness. Cousin Joyce, I have
had a revelation to-day. I have had a peep at my own heart
and life, and I am dissatisfied with both, especially in
connection with yourself. When you spoke to my mother this
morning and told her what you were going to do, how you had
made up your mind to leave the only relatives you have in the
world, because under their roof you had a shelter, not a home,
I felt so sorry for you, so ashamed for ourselves. It was your
birthday morning. You are twenty-one to-day. I was the same
four months ago, and then my mother did not know how to
lavish enough of costly things upon me. I had cards—works of
art that had cost pounds; flowers in profusion, letters,
messages, callers, jewellery, finery of all kinds, and a grand
evening party given in my honour. And you, Cousin Joyce, had
nothing but the coldest greeting, and an offer of our
secondhand and third-best clothes. Please let me finish—" for
Joyce would have stopped the confession half-way. "I do not
know how it was brought about, but I seemed to see
everything you had endured under this roof from the day of
your coming. No welcome, no sympathy, no home, no
friends."
"Yes, my uncle has always been kind, and I have had Sarah
Keene. Besides, I was but a stranger who had to win the
affection of strangers, though they might be relatives; and I
really believe you care for me after all!" cried Joyce, looking
up into Adelaide's face, and smiling through the tears which
her cousin's words had brought to her eyes. "Forgive me,
Adelaide. I want forgiveness, too, for I have judged you rather
hardly, I am afraid."
"No, you have not; I have never been kind, but I want to be
now." And two pairs of arms went out, and two girls' lips met
for the first time in mutual affection and forgiveness. Then
they sat down side by side, each encircling the other with one
embracing arm.
"We shall be friends as well as cousins for the future. Until

now, we have been neither," said Adelaide. "I wish you were
not going away, Joyce. If you will stay, I will try to make The
Chase more of a home to you than it has been. But how can
you, after what mamma said this morning? I think that
proposal about the dresses and your helping to alter ours was
too dreadful."
And the girl blushed with shame at the recollection.
"I should not have minded about working early and late if you
had wanted help and we had worked together," said Joyce. "If
any one here had been ill, I should have thought nothing too
much to do for them, night or day. Supposing that my uncle
had been poor, and had given me a home with his children, I
would have slaved for him and them most cheerfully, and
taken care that his kindness should have cost him nothing in
the end. But you are all rich, and every wish can be gratified;
and the thought of being sent to sew under the orders of
Russell was—"
"Hush, dear Joyce! I cannot bear it," interposed Adelaide, as

she laid her white hand on her cousin's lips. "That alone would
have driven you from us, and after what mamma said, you
cannot stay. Now you must show you have forgiven me by
taking this little birthday gift," and drawing a ring-case from
her pocket, Adelaide tried to place a beautiful ring on Joyce's
finger.
"Do not ask me, dear; I cannot take it," said Joyce.
"I bought it myself, and I have so large an allowance that it

cost me nothing; I wish it had. The having too much money
takes from us the joy of self-sacrifice."
"I cannot take it," repeated Joyce. "How would that diamond
look on the hand of a maid to little children? Besides, I have
rings that belonged to my mother, if I wished to wear any."
"You have not forgiven me," sighed Adelaide.
"Yes, and I will take a gift, too, and prize it. Spend ten
shillings on a little brooch in cut steel, and I will wear it, and
never part with it while I live. And give me your likeness; I
should like to have it, though I shall always picture your face
as it looks to-night."
"You shall have these trifles, Joyce, and I will keep this, no
matter how long, until you are willing to wear it." And
restoring the ring to its case she put it into her pocket. "Now
what else can I do for you?" she asked.
"My uncle breakfasts earlier than you and the rest do. I have
been used to pour out his coffee and join him at table. I think
he will miss me at first. Will you sometimes breakfast with
him?"
"How selfish I have been not to notice this, or care for his
loneliness! Rely on me, I will breakfast with him always,
unless by some special chance I have been up very late the
night before."
"I shall neither be missed nor wanted," said Joyce. "Indeed, I

begin to fear I shall soon be forgotten."
But she smiled as she said it, for she was glad to think that
the father and daughter would be brought together by her
own departure.
Then these two girls became more confidential, and Joyce

gave her cousin every particular respecting the work she had
undertaken, the manner in which she had obtained the
situation, and of the fact that Mrs. Caruth was sending her
own maid to accompany her on her journey to Springfield
Park.
"It seems quite amusing to think that one who is travelling

with such an object should be so attended, does it not?" asked
Joyce.
Adelaide looked thoughtful, then replied, "Mrs. Caruth must

think a great deal about you. Does she understand what you
are going to do?"
"I am not sure, but I do know she is my friend. She was

almost like a mother to me until I was about seventeen, and
when I had none of my own. Then—"
"Then what?"
"Her son came home for a time, and she had him, and I
became more of a companion to my father."
"I believe I have seen both Mrs. Caruth and her son. Does she
call him Alec?"
"Always. He is about thirty-two now. You see I was only nine

when he was twenty, and as the child of his old tutor, he made
a pet and playfellow of me. It seems strange that we should
both be grown-up people after a few years."
"He is very fond of his mother, and she of him," said Adelaide.
"Indeed, he seems a good, noble-minded man altogether.
Augusta thought there was no one like him during the eight
weeks we spent at Mentone."
A statement which did not appear to give unqualified

satisfaction to Joyce, for she paused a moment, then, in a
constrained voice, though with an attempt at archness, she
asked—
"Did Major Caruth think there was no one like Augusta?"
"He neither troubled himself about her nor any other girl. I
mean so far as paying special attention went. He was
everything that was kind and courteous, but the elder ladies
and the children absorbed the larger share of his time—
somewhat, I think, to the disgust of the grown-up girls. If
hazarded a guess, it would be that he had no heart left to
give, and that he was far too noble and true a man to pay
unmeaning attentions, which could lead to nothing but regrets
and pain for another. I suppose he has no sister, or he would
be a model 'brother of girls.'"
"No, but he is a brother of girls for all that. He would be to all

such, if circumstances called for his help, what the son of a
pure-minded, virtuous, Christian mother should be. I know
him so well."
Joyce's face was lighted up by a bright, glad look, born of

precious memories, but it faded as she said, "I am not likely
to meet Major Caruth again. I was Miss Mirlees, and a power
at Welton, as the parson's daughter in a country parish always
is, you know. Three days hence I shall be 'only a servant.'"
"Joyce, you must give up this plan of yours; I cannot bear to

think of it. My father cares for you; I want you at The Chase.
Augusta will come over to my side, for she is not nearly so
hard as she seems. We have both been carefully educated in
selfishness, and even a first step in the right direction costs a
great effort. But I can stir her to it, and we will make a
combined attack on my mother, who must give in. Say you
will stay."
"Not now, dear. But if ever the time should come when I can
be sure you all wish for me, or if I am needed by any, I will
return."
Adelaide was obliged to be satisfied with this. The girls parted
with expressions of affection and pledges of future friendship;
and Joyce laid her head on her pillow with a lighter heart than
she had done for months past.
Mr. Evans was expected home the evening before his niece
was to leave The Chase, but in place of him came a telegram
—
"Accident on line. Train delayed, but none injured. Expect me

at noon to-morrow."
Joyce was to leave the station at nine, so her uncle would not
arrive till after she was gone. Mrs. Evans declined to see her,
but sent word that when Joyce came to her senses, and was
prepared to submit and acknowledge she had done wrong,
she might write and say so.
Augusta, doubtless urged thereto by her sister, rose early

enough to say farewell to her cousin. Sarah Keene watched
her out of sight as well as she could through falling tears, and
prayed for a blessing on her head, and Adelaide, bravely
mounted beside Joyce in the shabby conveyance which took
her and her luggage to the station, whispered cheery words to
the very last moment, when, in company with Dobson, Mrs.
Caruth's staid waiting-woman, she started on her journey.
Moved still further by the new and better feelings just born in
her heart, Adelaide declined to drive with Mrs. Evans and
Augusta, and went instead to meet her father on his return at
noon.
It was a great surprise to Mr. Evans when he saw Adelaide's

beautiful face glowing with eager expectation, in search of
some traveller whose arrival she anticipated. He did not for a
moment associate her presence with his own home-coming,
until her eyes met his as the train stopped, and stepping
forward, she exclaimed—
"Papa, I am so glad you are here safe and sound!" And lifting
her face to his she kissed him lovingly again and again, then
slipping her arm through his, went with him to the carriage
which awaited them.
"That first kiss was poor Cousin Joyce's," she said. "She left it
for you, and I promised to deliver it."
"Joyce's! She is surely not gone? I thought you would all have
joined to keep her until my return. My only sister's only child
to leave The Chase in such haste!"
"She could not stay. I tried hard to persuade her, for, papa, I
am sorry I have not been kinder to Joyce. We are friends now,
dear friends, and I hope we shall always be so. I cannot blame
Joyce for going. How could she stay? But you do not know all
yet. I trust things will turn out better than they seem to
promise. I think I ought to tell you all about Joyce's birthday
and what was said, only you must promise to say nothing to
mamma. I cannot help thinking she is a little sorry now, and
she is more likely to feel regret about Joyce's going if no one
speaks of it."
Then Adelaide told her father all that had passed, and Mr.
Evans listened, not altogether sadly, for his daughter made
the most of all that had been bright for Joyce on her birthday
—the loving letters and souvenirs from Welton, Mrs. Caruth's
consideration for her cousin's safe convoy, the opening of
hearts between themselves, and the new-born friendship,
which was to bind them more closely than the ties of
relationship had done.
"And," continued Adelaide, "Joyce will never disgrace the

name she bears. I only wish I were more like her."
There was much to cheer Mr. Evans in what he heard from his
daughter, and acting upon her suggestion, he made no
allusion to Joyce's departure. His silence was both a relief and
a reproach to his wife, who expected a scene, and was
conscious that, in spite of her desire to free herself from a
sense of responsibility, she could not even excuse herself for
her treatment of Joyce.
On the following morning, when Mr. Evans went down,

expecting to take his breakfast in solitude, and feeling how
much he should miss Joyce's gentle ministry, he found
Adelaide already seated at the table. She rose as he entered
and lifted her face for a kiss.
"Now another, papa," she said. "That is for Joyce. You must
give me one every morning for her, as I am her deputy."
It was such a new thing for Mr. Evans to be greeted thus by

his own children, that he could hardly realize that he was
awake, but he showered many kisses on the fair, bright face
that waited for them.
"I did not expect to see you, my dear," he said.
"No, dear papa, but I must try to be a better daughter. I told

you yesterday that I was beginning to learn new lessons. If I
become what I wish to be, remember, Joyce was my first
teacher. When I asked what I could do for her, she told me
what I might do in a little way for you. But for her, I should
not be here; however, I will not leave you to a lonely meal
again."
And Adelaide kept her promise.
CHAPTER V.
JOYCE reached the station nearest to Springfield Park at three
o'clock, having had a change of trains, and a stoppage of an
hour and a half on the road. Mrs. Caruth's maid, returning
direct to Fernsclough, would rejoin her mistress before six.
Her train would not, however, start for twenty minutes, so she
was able to tell her mistress that she had seen Miss Mirlees in
charge of a grey-haired coachman, who, with two little girls,
awaited her arrival.
"Are you the young person for Springfield Park?" asked the
man.
Joyce replied in the affirmative.
"I am the coachman. The groom would have brought only a

trap, but the little ladies were wild to see their new maid, and
Mrs. Ross would only trust the children with me."
The man intended Joyce to understand that to drive any but

members of the family and their friends would be beneath the
dignity of so old a servant, and that the presence of the little
girls explained his own.
"No doubt Mrs. Ross feels that the children are safest with
you," said Joyce.
"Just so. She has had time to know what I am, for I drove her
when she was no bigger than the least of them, and I was in
her father's service. Now you step in next the eldest one—
Miss Mary. She should have been a boy by rights, but nobody
would like to change her for one now. Your things will be
brought by that lad, who has a trap close at hand. They are all
together, I suppose?"
Joyce pointed to her belongings on the platform, said farewell

to her escort, and sent messages of thanks and love to Mrs.
Caruth. Then she followed the coachman to a little carriage, in
which were seated two lovely children in the present charge of
the station-master's daughter.
"Come in," cried the elder child. "We wanted to see you, so
mamma let Price bring us. I am Mary, 'papa's Molly,' they call
me, and that is Alice. She turns her face away because she is
shy, but she will be friends soon. Mamma said we must be
very good and not make you sorry, because you have no
father and mother."
Tears sprang into Joyce's eyes, which the child noted instantly,
and her own face grew sorrowful.
"Why do you cry?" she said. "Let me kiss the tears away, as
mamma does mine, if I am only sorry, not naughty."
The winsome creature pulled Joyce's head down to her own

and smiled, until her new attendant was fain to smile in
response.
"There, that is right. Now look how pretty the park is, and see
the deer under the trees. They feed out of our hands, and
they will know you very soon, because you will be with us."
Joyce saw that her new surroundings would be even more

beautiful than her uncle's home, and she drank in with delight
the loveliness which met her eyes on every side, whilst Mary
prattled unceasingly till they reached the house. There she
was met by a pleasant, motherly person, who introduced
herself as Mrs. Powell, the housekeeper, and led her upstairs
to a good-sized cheerful room, very comfortably furnished,
and opening into a still larger one, in which were two little
beds. Both rooms again opened into the day nursery, a
delightful apartment, in which everything suggested the
personal superintendence of a thoughtful, loving mother.
The little girls had been taken charge of by Mrs. Ross's maid,
Paterson, and the housekeeper told Joyce that when she was
ready she was to come to her own room for refreshment.
"Here are your boxes in good time," said Mrs. Powell; and
thus Joyce was able to make the needed change in her dress.
She was about to go down, when, recollecting her new
position, she turned back for one of the aprons, ironed so
carefully by Sarah Keene's hands, and over which, as badges
of coming servitude for her darling, she had shed many a tear.
"Never mind," thought Joyce; "they are honourable badges, so

long as they accompany faithful performance of duty, work
done as in God's sight, and depending for its success on His
blessing."
So, with a bright face, the reflection of a brave heart, she

went down, after having occupied a few moments in thanking
God for a safe journey and a kind reception.
"I always have an early cup of tea," said Mrs. Powell, "and I
thought it would be the best for you, along with something
more substantial, after a journey. Your future meals will be
taken upstairs with the children. Mrs. Ross will see you in the
morning; but she and the master are away—only for the day;
they will be back to-night. My mistress trusted you to me, and
I promised to make you comfortable," said Mrs. Powell, with a
look of great kindness in her motherly face.
"It was very good of her to leave me in such hands," said

Joyce, with an answering smile. Then Mrs. Powell dropped her
voice to a whisper—
"Let me say a word about yourself, my dear. My mistress

trusts me, and she said—only to me, mind—that the friend
who wrote in answer to her inquiries had told her a little of
your history. How that you were a lady, used to be served
instead of serving others, and that if you chose to accept a
home with her, there was one open to you; but that you
preferred service to a life of dependence."
"What did Mrs. Ross say? I hope she did not think I wished to
deceive her in any way," said Joyce.
"No, indeed. She honours you for preferring work to

dependence, and says that if she finds you what she has been
led to expect, you, in turn, shall find a real home and true
friends at Springfield Park. There, my dear, I hope you will
sleep the sounder for knowing this; and if it will comfort you
to hear it, my heart warms to you, and you have one friend
already."
To Joyce this was like having her old friend Sarah Keene by
her side, and she thanked the kindly housekeeper most
heartily and gratefully for her encouraging words.
But the tea was being neglected, and Mrs. Powell turned
Joyce's attention in that direction; so, impelled by a healthy
girlish appetite, she made a hearty meal, much to her new
friend's satisfaction.
One hour after she spent with the children, of whom, however,
she was not to take formal charge until the morning. Then the
housekeeper, being at leisure, showed her through the house
and a portion of the gardens, and finally left her to indulge in
happier thoughts than she could have imagined would be
possible to her under her new circumstances.
Joyce rose early and dressed the children, the little one having
overcome her shyness, and being now willing to make friends.
She was sitting, telling them a baby-story, when Mrs. Ross
entered the nursery after breakfast, and greeted her with the
utmost kindness.
At the sight of their mother, the children rushed to her side,

and, clasped in her arms, forgot for the time their anxiety to
know the end of Joyce's fairy story.
How the girl sped at Springfield Park may be gathered from a
letter, written after three months' experience, to Sarah Keene.
Many shorter letters had been exchanged between Mr. Evans,
Adelaide, the old nurse, and Joyce; but she purposely
refrained from saying much about her position, until a
sufficient time had elapsed to allow her to form a fair
judgment as to the wisdom of the step she had taken.
MRS. ROSS SENT A NOTE A LITTLE LATER.
"
Springfield Park, Sept. 6th."
"After three months, dear old nurse and friend,
I can say that I am glad I came here. Every one
is good to me; the children are so sweet that it is
delightful to work for them; and I do work,
Sarah."
"I try to earn every penny, and I have proof

that Mrs. Ross is satisfied. Yesterday she told me
how glad she was that the children had learned to
love me, and that she was much pleased with my
mode of managing them. Then she gave me my
quarter's wages, and I found considerable
sweetness in receiving my first earnings. I was to
have seventeen pounds a year and all found; but
Mrs. Ross placed a five-pound note in my hand,
and would not receive any change."
"You must know I cannot occupy my time in

only dressing and attending to the children and
their clothes; the former are so docile, the latter
so handsome and abundant that they receive
little damage, and when at all shabby they are
given away; so I began to teach, and turned
everything I knew to account in order to benefit
my darling charges."
"Mrs. Ross found out what we were doing, and

said, 'You are teaching my children to love
information by leading them gently, and making it
attractive. How have you acquired such an
excellent method?'"
"'I taught in our Welton Sunday Schools,' I

said. 'My little scholars were the children of the
very poor; but I took more pains with them
because their learning time is short and their
opportunities are few. If my method has any
merit, it is owing to my dear father's example,
which I tried to copy.' My eyes filled. I could not
keep back my tears when I thought of him, and
of all I owed to his loving training."
"Mrs. Ross laid a gentle hand on my shoulder,

and said, 'Do not cry, Joyce. I feel deeply for you.
It must be hard to look back and think how things
were whilst he lived. I have heard so much of
your father's excellences, and how you were both
loved by rich and poor.'"
"'I am not unhappy,' I replied. 'Service here is

not servitude, and I am much better satisfied to
earn my bread than to owe it to the charity of
another.'"
"'You are right; but I should be wrong to accept

the faithful labours of a governess in return for a
nursemaid's salary. Henceforth you will receive
forty pounds a year, and, Miss Mirlees, I shall look
for you, with the children, in the drawing-room
daily, when we have no formal company and are
alone, or have only a few friends.'"
"I began to wonder if my old Welton frocks

would be good enough, but that evening a parcel
came to me, containing a dress-length of good
mourning silk, with all requisites for making it up.
Mrs. Ross sent a note a little later, to say that it
was a mark of the satisfaction felt by her husband
and herself at the improvement in their children."
"I can now wear my dear mother's watch and

ornaments without their seeming unsuitable, and
I shall once again find myself amongst people of

Textbook Mathematical and Statistical Applications in Life Sciences and Engineering 1St Edition Avishek Adhikari Ebook All Chapter PDF

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Textbook Mathematical and Statistical Applications in Life Sciences and Engineering 1St Edition Avishek Adhikari Ebook All Chapter PDF

Uploaded by

Copyright:

Available Formats

Mathematical and Statistical

Applications in Life Sciences and

Mathematical and Statistical Applications in Food

Mathematical Methods in Engineering and Applied

An Introduction to Statistical Analysis in Research:

Mathematical and Statistical Methods for Actuarial

Recent Advances in Mathematics for Engineering

Physical Sciences and Engineering Advances in Life

Handbook of Fractional Calculus with Applications Vol 7

Handbook of Fractional Calculus with Applications Vol 8

Yogendra Prasad Chaubey

Mathematical and Statistical

Mahima Ranjan Adhikari

ISBN 978-981-10-5369-6 ISBN 978-981-10-5370-2 (eBook)

© Springer Nature Singapore Pte Ltd. 2017

Printed on acid-free paper

This Springer imprint is published by Springer Nature

This is the commemoration volume celebrating the 10th foundation anniversary

• Focus on mathematical and statistical applications in a single volume,

Kolkata, India Avishek Adhikari

8 Nonparametric Estimation of Mean Residual Life Function Using

About the Editors

Yogendra Prasad Chaubey, Ph.D. is Professor of Statistics in Department of Mathematics and

Sujit K. Ghosh Department of Statistics, NC State University, Raleigh, NC, USA

Constantin Udrişte Faculty of Applied Sciences, Department of

Ushnish Sarkar and Avishek Adhikari

Abstract Frequency assignment problem (FAP) of radio networks is a very active

Channel assignment problem or frequency assignment problem is one of the most

U. Sarkar · A. Adhikari (B)

© Springer Nature Singapore Pte Ltd. 2017 3

| f (x) − f (y)| ≥ l x y , (1.1)

|L(u) − L(v)| ≥ k + 1 − d(u, v) (1.2)

While assigning colors (i.e., frequencies) to the vertices (i.e., transceivers) of a

1.1.2 Important Definitions, Notations, and a Brief Survey

regarding G and k. The vertices of L i are represented by vij , 1 ≤ j ≤ li , and if li = 1,

we replace vij by vi . In a r ck -coloring L on G, a color i is referred as a multiple color

1.1.3 Introduction to Holes

An interesting difference between general graph coloring and radio k-coloring is

L(2, 1): 0 3 L(3, 2, 1): 0 6

Fig. 1.2 Examples of r c2 -colorings with and without holes

Set of holes={5,7,8} Set of holes={1,5,8}

L(3,2,1)−coloring/ radio coloring

Fig. 1.3 Examples of r c3 -colorings/radio colorings with holes

Fig. 1.4 Examples of non-full colorable graphs

Fig. 1.5 One more example

1.2 k-Islands and an Equivalence Relation on Λρk (G)

1.3 An Upper Bound of ρk (G)nd Related Issues

Clearly, L̂ is an r ck -coloring with fewer (k − 1)-holes, leading to a contradiction.

Then L̂ is a radio k-coloring with span r ck (G) − 1, a contradiction.

Proof Applying finite number of suitable α-recolorings, we assume, without loss

We are now in a position to obtain an upper bound of ρk (G).

G, a set S ⊆ V (G) is a dominating set if for every vertex u ∈ V (G) \ S, there is

Theorem 1.2 [20, 41] For any graph G of order n, γ(G) ≥ n

Now, we explore the structure of G, including its regularity, order, domination

(i) G is a Δ-regular graph;

Let n = Δ + 1. Then l ≥ 2. Since lc(I j ) = r c(I j ), for every 0 ≤ j ≤ Δ, using

is a radio k-coloring of G with no (k − 1)-hole. Hence, μk (G) ≤ span( L̂) =

1.4 A New Family of Graphs: GΔ,t

1.4.1 A Characterization of GΔ,2

We now focus on t = 2. For any Δ ≥ 2, let G be the bipartite graph obtained by

Theorem 1.4 [42] Let G be a graph on n ≥ 3 vertices having minimum degree

Now we are in a position to prove the following result.

Sketch of proof. Clearly, G c is a (Δ + 1) regular graph of order 2(Δ + 1). As δ(G c ) =