1 \documentclass[14]{article}
7 %\usepackage[T1]{fontenc}
8 %\usepackage[latin1]{inputenc}
10 \renewcommand{\labelenumii}{\labelenumi\arabic{enumii}}
11 %\titleformat*{\section}{\Large\bfseries}
13 %\title{Response to the reviewers of \bf "Perimeter-based Coverage Optimization to Improve Lifetime in Wireless Sensor Networks"}
14 %\author{Ali Kadhum Idrees, Karine Deschinkela, Michel Salomon and Raphael Couturier}
22 \vspace{-0.5cm}\hspace{-2cm}FEMTO-ST Institute, UMR 6714 CNRS
24 \hspace{-2cm}University Bourgogne Franche-Comt\'e
26 \hspace{-2cm}IUT Belfort-Montb\'eliard, BP 527, 90016 Belfort Cedex, France.
31 On the decision for the article
33 ``Multiround Distributed Lifetime Coverage Optimization \\ Protocol in Wireless
36 by Ali Kadhum Idrees, Karine Deschinkel, Michel Salomon, and Raph\"ael Couturier
42 \noindent Dear Editor,
44 After a careful reading of your last decision letter, we are quite disappointed
45 by the rejection of our article named: ``Multiround Distributed Lifetime
46 Coverage Optimization Protocol in Wireless Sensor Networks'', submitted for
47 publication in the AD HOC NETWORKS journal. Indeed, we have started the
48 submission process in September 2014 and obtained the comments of only ONE
49 reviewer during July 2015. Its suggestion were very helpful and we incorporated
50 them in the revised article submitted in September 2015.
52 In particular, we have stopped the resolution of the Branch-and-Bound method
53 after a time threshold empirically defined and we retain the best feasible
54 solution found by the solver, as it was suggested by the reviewer. We made our
55 best to carefully address the issues raised by the referee and revise our paper
56 accordingly. So we would like to clarify some of the points raised by the
57 reviewer, since some them seem not to be not relevant or fair.
59 \noindent {\bf 1.} The authors have partially taken into account the comments
60 of my previous review. Additional content has been added, but these additions
61 are sometime confusing.\\
63 \textcolor{blue}{\textbf{\textsc{Answer:} The reviewer does not clearly indicate
64 which addition is confusing.}}\\
66 \noindent {\bf 2.} The answer that the authors have provided to my comments
67 should have been inserted into the paper, this is not always the case (i.e., my
68 comment about the duration of the rounds).\\
70 \textcolor{blue}{\textbf{\textsc{Answer:} We clearly indicated in Section~3.2
71 that the rounds are of equal duration and we explained that this parameter
72 should be set according to the types of application (see our answer in part
73 ``minor comments'' in our previous answer).}}\\
75 \noindent {\bf 3.} In Section~3.1, all nodes are assumed to be ``homogeneous
76 from the point of view of energy provision''. Why is such an hypothesis
77 necessary? This assumption is likely to be satisfied only when the WSN is
78 deployed for the first time, with new sensors. But after using the network for
79 the first time, the aforementioned hypothesis is not likely to be satisfied
82 \textcolor{blue}{\textbf{\textsc{Answer:} Maybe the honorable reviewer misread
83 the sentence in Section~3.1: ``We assume that all nodes are {\bf
84 homogeneous} in terms of communication and processing capabilities, and
85 {\bf heterogeneous} from the point of view of energy provision.'' }}\\
87 \noindent {\bf 4.} The last paragraph of Section 3.1 is very confusing. It
88 seems that the author attempt to propose area coverage instead of target
89 coverage, but instead of defining ``primary points'' inside the area to be
90 covered, they define points inside the sensing range of the sensors, that are
91 obviously covered when the corresponding sensor is active. There exists works in
92 the literature for area coverage, the authors should read them.\\
94 \textcolor{blue}{\textbf{\textsc{Answer:} This remark is rather offensive. Of
95 course we read literature for area coverage. We use the metric ``Coverage
96 Ratio'' (defined in Section~4.3) to measure how much the area is covered.
97 But the optimization process to decide which sensor has to be active or not
98 in each round is based on the coverage of only a specified set of points
99 called primary points. So the area coverage problem is transformed into the
100 target coverage problem.}}\\
102 \noindent {\bf 5.} In Section 3.2, the reason why a subregion is defined in such
103 a way that the distance between any two sensors in the same subregion is less
104 than 3 hops is not justified. The reader is not told if the proposed protocol
105 works better when the number of subregions is low or high. No algorithm for
106 defining the subregions is given. It pertains to a clustering problem, for which
107 a large number of algorithms exists, but without a sound definition of what is a
108 'good' partitioning for the proposed protocol, the reader cannot select a
109 clustering algorithm. \\
111 \textcolor{blue}{\textbf{\textsc{Answer:} The choice of the number of
112 subregions is discussed in the last paragraph of Section~4.2. More
113 particularly, we explain that this parameter should be chosen by taking into
114 account the trade-off between the benefit of the optimization problem
115 induced by the number of sensors in a subregion and the time needed to solve
116 it. As said at the beginning of Section~3.2, the area of interest is
117 ``divided into regular homogeneous subregions using a simple
118 divide-and-conquer algorithm.''}}\\
120 \noindent {\bf 6.} The ILP of Section 3.5 aims at addressing a bi-objective
121 problem. Since full coverage is required, the ILP should first ensure total
122 coverage, and then minimize overcoverage. The current ILP is not a proper
123 formulation for reaching this objective, as undercoverage can be compensated by
124 overcoverage. The primary and secondary objective are then mixed up into a
125 single objective function with no interpretable meaning. Of course, an objective
126 function value of 100 is better than an objective value of 101, but one cannot
127 tell if a gap of 1 makes a major difference or not. The objectives should be
128 distinguished, and the problem should be addressed as a multiobjective
129 optimization problem.\\
131 \textcolor{blue}{\textbf{\textsc{Answer:} As mentioned in the paper, the ILP of
132 Section~3.5 is based on the model proposed by F. Pedraza, A. L. Medaglia,
133 and A. Garcia (``Efficient coverage algorithms for wireless sensor
134 networks'') with some modifications. The originality of the model is to
135 solve both objectives in a parallel fashion: maximizing the coverage and
136 minimizing the overcoverage. Nevertheless the weights $w_\theta$ and $w_U$
137 must be properly chosen so as to guarantee that the number of points which
138 are covered during each round is maximum. By choosing $w_{U}$ much larger
139 than $w_{\theta}$, the coverage of a maximum of primary points is
140 ensured. Then for the same number of covered primary points, the solution
141 with a minimal number of active sensors is preferred. It has been formally
142 proven in the paper mentioned above that this guarantee is satisfied for a
143 constant weighting $w_{U}$ greater than $\left|P\right|$ (when $w_{\theta}$
147 \noindent {\bf 6.} The content of Section 4.3, where different metrics are
148 proposed to assess the solution quality, is a sign of an ill formulated problem:
149 how to comment on the performance of an algorithm on a criterion (say network
150 lifetime) if the ILP does not take this objective into account? The performances
151 with these additional objectives are likely to be related to the ILP solver
152 used: many optimal solutions to the ILP of Section 3 may have a very different
153 impact in terms of these additional metrics. Hence, measuring them is
156 \textcolor{blue}{\textbf{\textsc{Answer:} We disagree this remark. It is quite
157 possible to optimize a criterion to have an impact on another one. In the
158 problem formulation proposed here, the number of active sensors is minimized
159 in each round for a maximal level of coverage. Limiting the activation time
160 of each sensor has a direct impact on its lifetime and consequently on the
161 network lifetime, as shown in our experimental results. For example, such an
162 idea is used in the models developed for brachytherapy treatment planning to
163 improve the quality of a dose distribution ("Comparison of inverse planning
164 simulated annealing and geometrical optimization for prostate high-dose-rate
165 brachytherapy", I-Chow J. Hsu1, E. Lessard, V. Weinberg, J. Pouliot, {\it
166 Brachytherapy} Volume 3, Issue 3, 2004, Pages 147-152): the objective in
167 the problem formulation is to minimize a weighted sum of the differences
168 between prescribed doses and obtained doses in reference points, whereas
169 many criterion (like dose-volume histograms, conformal index COIN) are used
170 for quantitative evaluation of dose plans.}}\\
171 %si vous avez d'autres exemples plus parlants?
173 We hope that these observations will allow you to revise your decision
174 concerning our manuscript. If possible, we would like to have comments of an
