2 \usepackage{beamerthemefemto}
3 \usepackage[T1]{fontenc}
4 \usepackage{amsfonts,amsmath,amssymb,stmaryrd}
5 \usepackage[frenchb]{babel}
7 \usepackage[utf8]{inputenc}
8 %\usepackage[section]{algorithm}
9 %\usepackage{algorithmic}
12 \usepackage{spverbatim}
22 \usepackage{algorithmic}
23 %\usepackage[ruled,english,boxed,linesnumbered]{algorithm2e}
24 %\usepackage[english]{algorithme}
25 \usepackage{subfigure}
32 \def\setgrouptext#1{\gdef\grouptext{#1}}
33 \newenvironment{groupeditems}{\begin{displaymath}\left.\vbox\bgroup\setgrouptext}{%
34 \egroup\right\rbrace\hbox{\grouptext}\end{displaymath}}
36 \newcommand*\rot{\rotatebox{90}}
37 \newcommand*\OK{\ding{51}}
39 \newcommand{\mcrot}[4]{\multicolumn{#1}{#2}{\rlap{\rotatebox{#3}{#4}~}}}
41 \newcommand*{\twoelementtable}[3][l]%
43 \renewcommand{\arraystretch}{0.8}%
44 \begin{tabular}[t]{@{}#1@{}}%
51 \DeclareGraphicsExtensions{.jpg, .png , .pdf, .bmp, .pdftex}
53 \setbeamertemplate{section in toc}[sections numbered]
54 \setbeamertemplate{subsection in toc}[subsections numbered]
59 \frametitle{Presentation Outline}
60 \tableofcontents[currentsection]
65 \title{\textbf{Distributed Coverage Optimization Techniques for Improving Lifetime of Wireless Sensor Networks} \\\vspace{0.1cm}\hspace{2cm}\textbf{\textcolor{cyan}{\small PhD Dissertation Defense}}}
66 \author{\textbf{\textcolor{green}{Ali Kadhum IDREES}} \\\vspace{0.5cm} \small Under Supervision: \\\textcolor{cyan}{\small Raphaël COUTURIER, Karine DESCHINKEL \& Michel SALOMON} \\\vspace{0.2cm} \textcolor{blue}{ University of Franche-Comté - FEMTO-ST - DISC Dept. - AND Team} \\\vspace{0.2cm}~~~~~~~~~~~~~~~~\textbf{\textcolor{green}{1 October 2015 }}}
68 %\institute[FEMTO-ST, DISC]{\textit{FEMTO-ST - DISC Departement - AND Team}}
75 % ____ _____ ____ _ _ _____
76 % | _ \| ____| __ )| | | |_ _|
77 % | | | | _| | _ \| | | | | |
78 % | |_| | |___| |_) | |_| | | |
79 % |____/|_____|____/ \___/ |_|
87 \setbeamertemplate{background}{\titrefemto}
95 \setbeamertemplate{background}{\pagefemto}
101 \begin{frame} {Problem Definition, Solution, and Objectives}
104 \includegraphics[width=0.475\textwidth]{Figures/6}
106 % \includegraphics[width=0.475\textwidth]{Figures/8}
108 \includegraphics[width=0.475\textwidth]{Figures/10}
110 % \includegraphics[width=0.475\textwidth]{Figures/13}
113 \begin{block}{\textcolor{white}{ MAIN QUESTION?}}
114 How to reduce the redundancy while coverage preservation for prolong the network lifetime continuously and effectively when monitoring a certain area of interest?
122 \begin{frame}{Problem Definition, Solution, and Objectives}
124 \begin{block}{\textcolor{white}{OUR SOLUTION}}
125 The area of interest is divided into subregions using a divide-and conquer method and then combine two efficient techniques :
128 \item Leader Election for each subregion.
129 % \item Activity Scheduling based optimization is planned for each subregion.
134 \includegraphics[width=0.475\textwidth]{Figures/div}
136 \includegraphics[width=0.475\textwidth]{Figures/div2}
144 \begin{frame}{Problem Definition, Solution, and Objectives}
146 \begin{block}{\textcolor{white}{OUR SOLUTION}}
148 %\item Leader Election for each subregion.
149 \item Activity Scheduling based optimization is planned for each subregion.
154 \includegraphics[width=0.775\textwidth]{Figures/act}
163 \begin{frame}{Problem Definition, Solution, and Objectives}
165 \begin{block}{\textcolor{white}{Dissertation Objectives}}
166 Develop energy-efficient distributed optimization protocols that should be able to:
168 \item Schedule node activities by optimize both coverage and lifetime.
169 \item Combine two efficient techniques: leader election and sensor activity scheduling.
170 \item Perform a distributed optimization process.
183 \frametitle{Presentation Outline}
185 \tableofcontents[section,subsection]
193 \section{\small {State of the Art}}
199 \begin{frame}{Wireless Sensor Networks (WSNs)}
203 \column{.58\textwidth}
206 \includegraphics[height = 3cm]{Figures/WSNT.jpg}
214 \item Electronic Low-cost tiny device.
215 \item Sense, process and transmit data.
216 \item Limited energy, memory and processing capabilities.
220 \column{.52\textwidth}
223 \includegraphics[height = 4.5cm]{Figures/WSN.jpg}
227 \includegraphics[height = 2cm]{Figures/sn.jpg}
231 % \begin{femtoBlock} {}% {SOME APPLICATIONS OF WSNs \\}
233 % \includegraphics[height =1 cm]{1.png}
234 % \includegraphics[height =1cm]{2.png}\\
235 % \includegraphics[height =1cm]{5.jpg}
236 % \includegraphics[height = 1cm]{traffic.jpg}
237 % \includegraphics[height = 1cm]{3.png}
252 \begin{frame}{Types of Wireless Sensor Networks}
257 %\column{.52\textwidth}
259 % \item Terrestrial WSNs.
260 % \item Underground WSNs.
261 % \item Underwater WSNs.
262 % \item Multimedia WSNs.
267 % \column{.58\textwidth}
269 \includegraphics[height = 7cm]{Figures/typesWSN.pdf}
279 \begin{frame}{Applications}
283 \includegraphics[height = 7cm]{Figures/WSNAP.pdf}
291 \begin{frame}{Energy-Efficient Mechanisms of a working WSN}
296 \includegraphics[height = 5cm]{Figures/WSN-M.pdf}
300 %\begin{frame}{Energy-Efficient Mechanisms of a working WSN}
304 % \includegraphics[height = 7cm]{Figures/WSN-S.pdf}
311 \begin{frame}{Network Lifetime}
313 \begin{block}{\textcolor{white} {Some network lifetime defintions:}}
314 \begin{enumerate}[i)]
315 \item \small Time spent until death of the first sensor ( or cluster head ).
316 \item Time spent until death of all wireless sensor nodes in WSN.
317 \item Time spent by WSN in covering each target by at least one sensor.
318 \item Time during which the area of interest is covered by at least k nodes.
319 \item Elapsed time until losing the connectivity or the coverage.
323 \begin{block}{\textcolor{white} {Network lifetime In this dissertation:}}
324 Time elapsed until the coverage ratio becomes less than a predetermined threshold $\alpha$.
333 \begin{frame}{Coverage in Wireless Sensor Networks}
335 \begin{block} <1-> {\textcolor{white} {Coverage Definition:}}
336 \textcolor{blue} {Coverage} reflects how well a sensor field is monitored efficiently using as less energy as possible.
341 \begin{block} <2-> {\textcolor{white} {Coverage Types:}}
343 \item \small \textcolor{blue} {Area coverage:} every point inside an area has to be monitored.
344 \item \textcolor{blue} {Target coverage:} is to cover only a finite number of discrete points called targets.
346 \item \textcolor{blue} {Barrier coverage:} is to detect targets as they cross a barrier such as in intrusion detection and border surveillance applications.
352 \begin{block} <3-> {\textcolor{white} {Coverage type in this dissertation:}}
353 The work presented in this dissertation deals with area coverage.
361 \begin{frame}{Existing Works}
363 \begin{block} {\textcolor{white} {Coverage Approaches:}}
364 Most existing coverage approaches in literature classified into
365 \begin{enumerate}[A)]
366 \item Full centralized coverage algorithms.
368 \item Optimal or near optimal solution.
369 \item low computation power for the sensors (except for base station).
370 \item High communication overhead.
371 \item Not scalable for large WSNs.
373 \item Full distributed coverage algorithms.
375 \item Lower quality solution.
376 \item High communication overhead especially for dense WSNs.
377 \item Reliable and scalable for large WSNs.
384 \begin{block} {\textcolor{white} {Coverage protocols in this dissertation:}}
385 The protocols presented in this dissertation combine between the two above approaches.
395 \section{\small {Distributed Lifetime Coverage Optimization Protocol (DiLCO)}}
401 \begin{frame}{\small DiLCO Protocol $\blacktriangleright$ Assumptions and Network Model:}
405 \begin{femtoBlock} {} %{Assumptions and Network Model:}
409 \column{.50\textwidth}
413 \begin{enumerate} [$\divideontimes$]
414 \item Static Wireless Sensors.
415 \item Uniform deployment.
416 \item High density deployment.
417 \item Homogeneous in terms of:
419 \item Sensing, Communication, and Processing capabilities
421 \item Heterogeneous Energy.
422 \item Its $R_c\geq 2R_s$.
423 \item Multi-hop communication.
424 \item Know Its location by:
426 \item Embedded GPS or
427 \item Location Discovery Algorithm.
433 \column{.50\textwidth}
434 \begin{enumerate} [$\divideontimes$]
435 \item Using two kinds of packet:
438 \item ActiveSleep packet.
440 \item Five status for each node:
442 \item LISTENING, ACTIVE, SLEEP, COMPUTATION, and COMMUNICATION.
446 \begin{femtoBlock} { \small Primary point coverage model}
449 \includegraphics[height = 4.0cm]{Figures/fig21.pdf}
463 \begin{frame}{\small DiLCO Protocol $\blacktriangleright$ Main Idea}
465 \begin{femtoBlock} {}%{Main Idea:\\}
467 \includegraphics[height = 2.5cm]{Figures/OneSensingRound.jpg}
471 \item \textcolor{blue}{ \textbf{INFORMATION EXCHANGE:}}\\
472 Sensors exchanges through multi-hop communication, their:
474 \item Position coordinates,
475 \item current remaining energy,
476 \item sensor node ID, and
477 \item number of its one-hop live neighbors.
490 \begin{frame}{\small DiLCO Protocol $\blacktriangleright$ Main Idea}
492 \begin{femtoBlock} {}%{Main Idea:\\}
494 \begin{enumerate} [2.]
496 \item \textcolor{blue}{ \textbf{ LEADER ELECTION:}}\\
497 The selection criteria are, in order of importance:
499 \item larger number of neighbors,
500 \item larger remaining energy, and then in case of equality,
505 \begin{enumerate} [3.]
506 \item \textcolor{blue}{ \textbf{ DECISION:}} \\
507 Leader solves an integer program(see next slide) to:
509 \item Select which sensors will be activated in the sensing phase.
510 \item Send Active-Sleep packet to each sensor in the subregion.
513 \begin{enumerate} [4.]
514 \item \textcolor{blue}{ \textbf{ SENSING:}} \\
515 Based on Active-Sleep Packet Information:
517 \item Active sensors will execute their sensing task.
518 \item Sleep sensors will wait a time equal to the period of sensing to wakeup.
531 \begin{frame}{\small DiLCO Protocol $\blacktriangleright$ Coverage Problem Formulation}
532 \begin{femtoBlock} { }
533 \noindent Our coverage optimization problem can then be formulated as follows:
534 \begin{equation*} \label{eq:ip2r}
537 \min \sum_{p \in P} (w_{\theta} \Theta_{p} + w_{U} U_{p})&\\
538 \textrm{subject to :}&\\
539 \sum_{j \in J} \alpha_{jp} X_{j} - \Theta_{p}+ U_{p} =1, &\forall p \in P\\
541 %\sum_{t \in T} X_{j,t} \leq \frac{RE_j}{e_t} &\forall j \in J \\
543 \Theta_{p}\in \mathbb{N}, &\forall p \in P\\
544 U_{p} \in \{0,1\}, &\forall p \in P \\
545 X_{j} \in \{0,1\}, &\forall j \in J
551 \item $X_{j}$ : indicates whether or not the sensor $j$ is actively sensing (1
552 if yes and 0 if not);
553 \item $\Theta_{p}$ : {\it overcoverage}, the number of sensors minus one that
554 are covering the primary point $p$;
555 \item $U_{p}$ : {\it undercoverage}, indicates whether or not the primary point
556 $p$ is being covered (1 if not covered and 0 if covered).
567 \begin{frame}{\small DiLCO Protocol $\blacktriangleright$ DiLCO Protocol Algorithm}
568 %\begin{femtoBlock} {}
570 %\includegraphics[height = 7.2cm]{Figures/algo.jpeg}
571 \includegraphics[height = 7.2cm]{Figures/Algo1.png}
582 \begin{frame}{\small DiLCO Protocol $\blacktriangleright$ Simulation Framework}
586 \caption{Relevant parameters for network initializing.}
590 Parameter & Value \\ [0.5ex]
592 Sensing Field & $(50 \times 25)~m^2 $ \\
593 Nodes Number & 50, 100, 150, 200 and 250~nodes \\
594 Initial Energy & 500-700~joules \\
595 Sensing Period & 60 Minutes \\
596 $E_{th}$ & 36 Joules\\
601 Modeling Language & A Mathematical Programming Language (AMPL) \\
602 Optimization Solver & GNU linear Programming Kit (GLPK) \\
603 Network Simulator & Discrete Event Simulator OMNeT++
614 \begin{frame}{\small DiLCO Protocol $\blacktriangleright$ Energy Model \& Performance Metrics }
616 \begin{femtoBlock} {Energy Consumption Model}
621 %\caption{Power consumption values}
623 \begin{tabular}{|l||cccc|}
625 {\bf Sensor status} & MCU & Radio & Sensing & {\it Power (mW)} \\
627 LISTENING & On & On & On & 20.05 \\
628 ACTIVE & On & Off & On & 9.72 \\
629 SLEEP & Off & Off & Off & 0.02 \\
630 COMPUTATION & On & On & On & 26.83 \\
632 \multicolumn{4}{|l}{Energy needed to send or receive a 2-bit content message} & 0.515 \\
639 \begin{femtoBlock} {Performance Metrics}
641 \begin{enumerate}[$\mapsto$]
642 \item {{\bf Network Lifetime}}
643 \item {{\bf Coverage Ratio (CR)}}
644 \item {{\bf Energy Consumption}}
645 \item{{\bf Number of Active Sensors Ratio (ASR)}}
646 \item {{\bf Execution Time}}
647 %\item {{\bf Stopped Simulation Runs}}
658 \begin{frame}{ \small DiLCO Protocol $\blacktriangleright$ Performance Comparison}
663 \includegraphics[scale=0.5] {Figures/R3/CR.eps}
664 \caption{Coverage ratio for 150 deployed nodes}
665 \label{Figures/ch4/R3/CR}
677 \begin{frame}{ \small DiLCO Protocol $\blacktriangleright$ Performance Comparison}
682 \includegraphics[scale=0.5]{Figures/R3/ASR.eps}
683 \caption{Active sensors ratio for 150 deployed nodes }
684 \label{Figures/ch4/R3/ASR}
692 %\begin{frame}{ \small DiLCO Protocol $\blacktriangleright$ Performance Comparison}
696 %\includegraphics[scale=0.5]{Figures/R3/SR.eps}
697 %\caption{Percentage of stopped simulation runs for 150 deployed nodes }
698 %\label{Figures/ch4/R3/SR}
706 \begin{frame}{ \small DiLCO Protocol $\blacktriangleright$ Performance Comparison}
710 \column{.50\textwidth}
711 \includegraphics[scale=0.35]{Figures/R3/EC95.eps}
712 \footnotesize \\~~~~~~~~~~~~~~~~~~~~~~~~~~~~~(a)\\
713 \column{.50\textwidth}
714 \includegraphics[scale=0.35]{Figures/R3/EC50.eps}
715 \footnotesize \\~~~~~~~~~~~~~~~~~~~~~~~~~~~~~(b) \\
717 \caption{Energy consumption for (a) $Lifetime_{95}$ and (b) $Lifetime_{50}$}
718 \label{Figures/ch4/R3/EC}
728 \begin{frame}{ \small DiLCO Protocol $\blacktriangleright$ Performance Comparison}
732 \column{.50\textwidth}
733 \includegraphics[scale=0.35]{Figures/R3/LT95.eps}
734 \footnotesize \\~~~~~~~~~~~~~~~~~~~~~~~~~~~~~(a)\\
735 \column{.50\textwidth}
736 \includegraphics[scale=0.35]{Figures/R3/LT50.eps}
737 \footnotesize \\~~~~~~~~~~~~~~~~~~~~~~~~~~~~~(b) \\
739 \caption{Network lifetime for (a) $Lifetime_{95}$ and (b) $Lifetime_{50}$}
740 \label{Figures/ch4/R3/LT}
753 \section{\small{Multiround Distributed Lifetime Coverage Optimization Protocol (MuDiLCO)}}
759 \begin{frame}{\small MuDiLCO Protocol $\blacktriangleright$ Main Idia}
762 \includegraphics[width=110mm]{Figures/GeneralModel.jpg}
763 \caption{MuDiLCO protocol.}
772 \begin{frame}{\small MuDiLCO Protocol $\blacktriangleright$ Multiround Coverage Problem Formulation}
775 Our coverage optimization problem can then be formulated as follows
778 \min \sum_{t=1}^{T} \sum_{p=1}^{P} \left(W_{\theta}* \Theta_{t,p} + W_{U} * U_{t,p} \right) \label{eq15}
784 \sum_{j=1}^{|J|} \alpha_{j,p} * X_{t,j} = \Theta_{t,p} - U_{t,p} + 1 \label{eq16} \hspace{6 mm} \forall p \in P, t = 1,\dots,T
788 \sum_{t=1}^{T} X_{t,j} \leq \lfloor {RE_{j}/E_{th}} \rfloor \hspace{6 mm} \forall j \in J, t = 1,\dots,T
793 X_{t,j} \in \lbrace0,1\rbrace, \hspace{10 mm} \forall j \in J, t = 1,\dots,T \label{eq17}
797 U_{t,p} \in \lbrace0,1\rbrace, \hspace{10 mm}\forall p \in P, t = 1,\dots,T \label{eq18}
801 \Theta_{t,p} \geq 0 \hspace{10 mm}\forall p \in P, t = 1,\dots,T \label{eq178}
814 \begin{frame}{\small MuDiLCO Protocol $\blacktriangleright$ MuDiLCO Protocol Algorithm}
816 \begin{femtoBlock} {}
818 %\includegraphics[height = 7.2cm]{Figures/algo2.jpeg}
819 \includegraphics[height = 7.2cm]{Figures/Algo2.png}
827 \begin{frame}{\small MuDiLCO Protocol $\blacktriangleright$ Results Analysis and Comparison}
831 \includegraphics[scale=0.5] {Figures/R1/CR.pdf}
832 \caption{Average coverage ratio for 150 deployed nodes}
841 \begin{frame}{\small MuDiLCO Protocol $\blacktriangleright$ Results Analysis and Comparison}
845 \includegraphics[scale=0.5]{Figures/R1/ASR.pdf}
846 \caption{Active sensors ratio for 150 deployed nodes}
855 %\begin{frame}{\small MuDiLCO Protocol $\blacktriangleright$ Results Analysis and Comparison}
859 %\includegraphics[scale=0.5]{Figures/R1/SR.pdf}
860 %\caption{Cumulative percentage of stopped simulation runs for 150 deployed nodes }
868 \begin{frame}{\small MuDiLCO Protocol $\blacktriangleright$ Results Analysis and Comparison}
872 \includegraphics[scale=0.5]{Figures/R1/T.pdf}
873 \caption{Execution Time (in seconds)}
882 \begin{frame}{\small MuDiLCO Protocol $\blacktriangleright$ Results Analysis and Comparison}
886 \column{.50\textwidth}
887 \includegraphics[scale=0.35]{Figures/R1/EC95.eps}
888 \footnotesize \\~~~~~~~~~~~~~~~~~~~~~~~~~~~~~(a)\\
889 \column{.50\textwidth}
890 \includegraphics[scale=0.35]{Figures/R1/EC50.eps}
891 \footnotesize \\~~~~~~~~~~~~~~~~~~~~~~~~~~~~~(b) \\
893 \caption{Energy consumption for (a) $Lifetime_{95}$ and (b) $Lifetime_{50}$}
894 \label{Figures/ch4t/R3/EC}
902 \begin{frame}{\small MuDiLCO Protocol $\blacktriangleright$ Results Analysis and Comparison}
906 \column{.50\textwidth}
907 \includegraphics[scale=0.35]{Figures/R1/LT95.eps}
908 \footnotesize \\~~~~~~~~~~~~~~~~~~~~~~~~~~~~~(a)\\
909 \column{.50\textwidth}
910 \includegraphics[scale=0.35]{Figures/R1/LT50.eps}
911 \footnotesize \\~~~~~~~~~~~~~~~~~~~~~~~~~~~~~(b) \\
913 \caption{Network lifetime for (a) $Lifetime_{95}$ and (b) $Lifetime_{50}$}
914 \label{Figures/ch4/Rh3/EC}
922 \section{\small {Perimeter-based Coverage Optimization (PeCO) to Improve Lifetime in WSNs
929 \begin{frame}{\small PeCO Protocol $\blacktriangleright$ Assumptions and Models}
934 \column{.50\textwidth}
935 \includegraphics[scale=0.40]{Figures/ch6/pcm.jpg}
936 \footnotesize \\~~~~~~~~~~~~~~~~~~~~~~~~~~(a)\\
937 \column{.50\textwidth}
938 $$\alpha = \arccos \left(\dfrac{Dist(u,v)}{2R_s}
940 \includegraphics[scale=0.40]{Figures/ch6/twosensors.jpg}
941 \footnotesize \\~~~~~~~~~~~~~~~~~~~~~~~~~~(b) \\
943 \caption{(a) Perimeter coverage of sensor node 0 and (b) finding the arc of
944 $u$'s perimeter covered by $v$.}
953 \begin{frame}{\small PeCO Protocol $\blacktriangleright$ Assumptions and Models}
958 \column{.50\textwidth}
959 \includegraphics[scale=0.33]{Figures/ch6/expcm2.jpg}
960 \footnotesize \\~~~~~~~~~~~~~~~~~~~~~~~~~~(a)\\
961 \column{.50\textwidth}
962 \includegraphics[scale=0.38]{Figures/tbl.jpeg}
963 \footnotesize \\~~~~~~~~~~~~~~~~~~~~~~~~~~(b) \\
965 \caption{(a) Maximum coverage levels for perimeter of sensor node $0$. and (b) Coverage intervals and contributing sensors for sensor node 0.}
977 \begin{frame}{\small PeCO Protocol $\blacktriangleright$ PeCO Protocol Algorithm}
979 %\includegraphics[height = 7.2cm]{Figures/algo6.jpeg}
983 \includegraphics[height = 7.2cm]{Figures/Algo3.png}
991 \begin{frame}{\small PeCO Protocol $\blacktriangleright$ Perimeter-based Coverage Problem Formulation}
996 \includegraphics[scale=0.5]{Figures/ch6/formula6.png}
1003 %%%%%%%%%%%%%%%%%%%%
1005 %%%%%%%%%%%%%%%%%%%%
1006 \begin{frame}{\small PeCO Protocol $\blacktriangleright$ Performance Evaluation and Analysis}
1010 \includegraphics[scale=0.5] {Figures/ch6/R/CR.eps}
1011 \caption{Coverage ratio for 200 deployed nodes.}
1017 %%%%%%%%%%%%%%%%%%%%
1019 %%%%%%%%%%%%%%%%%%%%
1020 \begin{frame}{\small PeCO Protocol $\blacktriangleright$ Performance Evaluation and Analysis}
1024 \includegraphics[scale=0.5]{Figures/ch6/R/ASR.eps}
1025 \caption{Active sensors ratio for 200 deployed nodes.}
1031 %%%%%%%%%%%%%%%%%%%%
1033 %%%%%%%%%%%%%%%%%%%%
1034 \begin{frame}{\small PeCO Protocol $\blacktriangleright$ Performance Evaluation and Analysis}
1038 \column{.50\textwidth}
1039 \includegraphics[scale=0.35]{Figures/ch6/R/EC95.eps}
1040 \footnotesize \\~~~~~~~~~~~~~~~~~~~~~~~~~~~~~(a)\\
1041 \column{.50\textwidth}
1042 \includegraphics[scale=0.35]{Figures/ch6/R/EC50.eps}
1043 \footnotesize \\~~~~~~~~~~~~~~~~~~~~~~~~~~~~~(b) \\
1045 \caption{Energy consumption per period for (a)~$Lifetime_{95}$ and (b)~$Lifetime_{50}$.}
1052 %%%%%%%%%%%%%%%%%%%%
1054 %%%%%%%%%%%%%%%%%%%%
1055 \begin{frame}{\small PeCO Protocol $\blacktriangleright$ Performance Evaluation and Analysis}
1059 \column{.50\textwidth}
1060 \includegraphics[scale=0.35]{Figures/ch6/R/LT95.eps}
1061 \footnotesize \\~~~~~~~~~~~~~~~~~~~~~~~~~~~~~(a)\\
1062 \column{.50\textwidth}
1063 \includegraphics[scale=0.35]{Figures/ch6/R/LT50.eps}
1064 \footnotesize \\~~~~~~~~~~~~~~~~~~~~~~~~~~~~~(b) \\
1066 \caption{Network Lifetime for (a)~$Lifetime_{95}$ and (b)~$Lifetime_{50}$.}
1072 %%%%%%%%%%%%%%%%%%%%
1074 %%%%%%%%%%%%%%%%%%%%
1075 %\begin{frame}{\small PeCO Protocol $\blacktriangleright$ Performance Evaluation and Analysis}
1077 %\begin{figure} [h!]
1078 %\centering \includegraphics[scale=0.5]{Figures/ch6/R/LTa.eps}
1079 %\caption{Network lifetime for different coverage ratios.}
1086 %%%%%%%%%%%%%%%%%%%%
1088 %%%%%%%%%%%%%%%%%%%%
1089 \section{\small {Conclusion and Perspectives}}
1092 %%%%%%%%%%%%%%%%%%%%
1094 %%%%%%%%%%%%%%%%%%%%
1095 \begin{frame}{Conclusion}
1096 \begin{enumerate} [$\blacktriangleright$]
1098 \item Two-step approaches are proposed to optimize both coverage and lifetime performances, where:
1100 \item Sensing field is divided into smaller subregions using divide-and-conquer method.
1101 \item One of the proposed optimization protocols is applied in each subregion in a distributed parallel way.
1103 \item The proposed protocols (DiLCO, MuDiLCO, PeCO) combine two efficient mechanisms:
1105 \item Network leader election, and
1106 \item Sensor activity scheduling based optimization.
1108 \item Our protocols are periodic where each period consists of 4
1111 \item Information exchange,
1112 \item Network leader election,
1113 \item Decision based optimization, and
1124 %%%%%%%%%%%%%%%%%%%%
1126 %%%%%%%%%%%%%%%%%%%%
1127 \begin{frame}{Conclusion}
1128 \begin{enumerate} [$\blacktriangleright$]
1130 \item DiLCO and PeCO provide a schedule for one round per period.
1131 \item MuDiLCO provides a schedule for multiple rounds per period.
1132 \item Comparison results show that DiLCO, MuDiLCO, and PeCO protocols:
1134 \item maintain the coverage for a larger number of rounds.
1135 \item use less active nodes to save energy efficiently during sensing.
1136 \item are more powerful against network disconnections.
1137 \item perform the optimization with suitable execution times.
1138 \item consume less energy.
1139 \item prolong the network lifetime.
1146 %%%%%%%%%%%%%%%%%%%%
1148 %%%%%%%%%%%%%%%%%%%%
1149 \begin{frame}{Perspectives}
1150 \begin{enumerate} [$\blacktriangleright$]
1151 \item The optimal number of subregions will be investigated.
1152 \item Design a heterogeneous integrated optimization protocol to integrate coverage, routing, and data aggregation protocols.
1153 \item Extend PeCO protocol so that the schedules are planned for multiple
1155 \item We plan to consider particle swarm optimization or evolutionary algorithms to obtain quickly near optimal solutions.
1156 \item Improve our mathematical models to take into account heterogeneous sensors from both energy and node characteristics point of views.
1157 \item The cluster head will be selected in a distributed way and based on local information.
1164 %%%%%%%%%%%%%%%%%%%%
1166 %%%%%%%%%%%%%%%%%%%%
1167 %\begin{frame}{Mes perspectives}
1171 %%%%%%%%%%%%%%%%%%%%
1173 %%%%%%%%%%%%%%%%%%%%
1177 \textcolor{BleuFemto}{Thank You for Your Attention!}\\\vspace{2cm}
1178 \textcolor{BleuFemto}{Questions?}\\