This action might not be possible to undo. Are you sure you want to continue?

(IEEE TRANSACTIONS ON MOBILE COMPUTING, VOL. 8, NO. 4, APRIL 2009) PUBLISHED BY:

7/29/2011

**Abdelmorhit El Rhazi, Samuel Pierre, Senior
**

(Member, IEEE) UNDER THE GUIDENCE OF

**Dr. KINGSLEY J. SINGH
**

DEAN,

(SCHOOL OF MECHANICAL DPT.)

PRESENTED BY:

**VISHNU DEV LAWANIYA
**

M.TECH. (CAD) REG. No 1531010025

1

SRM UNIVERSITY, TAMILNADU

ABSTRACT

The main challenge in wireless sensor network deployment pertains to optimizing energy consumption when collecting data from sensor nodes. This paper proposes a new centralized clustering method for a data collection mechanism in wireless sensor networks, which is based on network energy maps and Quality-of-Service (QoS) requirements. The clustering problem is modeled as a hypergraph partitioning and its resolution is based on a tabu search heuristic. Our approach defines moves using largest size cliques in a feasibility cluster graph.

7/29/2011 Compared to other methods (CPLEX-based method. Index Terms Wireless sensor network. simulated annealing-based method). 3 . the results show that our tabu search-based approach returns high-quality solutions in terms of cluster cost and execution time. tabu search. this approach is suitable for handling network extensibility in a satisfactory manner. energy map. distributed method. data collect. As a result. clustering methods.

humidity).. Two mechanisms are used to reduce energy consumption: Message aggregation Filtering of redundant data .capture and measure specific elements from the physical world (e. temperature. pressure.INTRODUCTION SENSORS :.g.

7/29/2011 These mechanisms generally use clustering methods in order to coordinate aggregation and filtering. Clustering methods belong to either one of two categories: Distributed Centralized 5 .

However. the problem is modeled as a graph partitioning problem with particular constraints that render this problem NP-hard. Then.7/29/2011 The centralized approach assumes that the existence of a particular node is cognizant of the information pertaining to the other network nodes. the major drawbacks of this category are linked to additional costs engendered by communicating the network node information and the time required to solve an optimization problem. The central node determines clusters by solving this partitioning problem. 6 .

Such a clustering mechanism is used to collect data in sensor networks. 7 . Centralized clustering is modeled as hyper graph partitioning. The first original aspect of this investigation consists of adding these constraints to the clustering mechanism that helps the data collection algorithm in order to reduce energy consumption and provide applications with the information required without burdening them with unnecessary data.7/29/2011 This paper proposes a new centralized clustering mechanism equipped with energy maps and constrained by Quality-of-Service (QoS) requirements. The novel method proposes the use of a TABU SEARCH heuristic to solve this problem.

7/29/2011 Data aggregation and data filtering are two methods that reduce the quantity of data received by applications. 8 . The aggregation data mechanism allows for the gathering of several measures into one record whose size is less than the extent of the initial records. we propose a novel data collection approach for sensor networks that use energy maps and QoS requirements to reduce power consumption while increasing network coverage.

ROBLEM FORMULATION 7/29/2011 The mechanism comprises two phases: During the first phase. which receives the application query and obtains results from other nodes before returning them to the applications. 9 . the applications specify their QoS requirements regarding the data required by the applications. They send their requests to a particular node S. The collector node builds the clusters. called the collector node. optimally using the QoS requirements and the energy map information.

7/29/2011 During the second phase. computing aggregate functions. 10 . The cluster head is in charge of various activities: coordinating the data collection within its cluster. the cluster heads must provide the collector node with combined measurements for each period. II. and sending results to a node collector. III. Split the network nodes into a set of clusters Gi that satisfies the application requirements. Prolong the network lifetime. filtering redundant measurements. Reduce energy consumption. The goal of the clustering algorithm is to I.

7/29/2011 Clusters are built according to the following criteria: Maximize network coverage using the energy map. Gather nodes likely to hold redundant measurements. The considered network contains a set V of m stationary nodes whose localizations are known. Gather nodes located within the same zone delimited by the application. The communication model can be described as multihop. 11 .

A query duration. mut. T. measurements are considered redundant. The network required for the query run a total time whose value is equal to T. it could decrease the step value or even ignore 12 the transfer of such value. fq. Within a single zone. A measurement uncertainty threshold. A zone size step. If the difference between two simultaneous measurements from two different nodes in the same zone (fourth requirement) is inferior to mut. The step value determines the zone length. The network provides results to the application every time the duration fq expires. . then one of them is considered redundant.An application can specify the following7/29/2011 QoS requirements: Data collection frequency. If an application requires more precision.

An edge ( i . It is capable of determining feasible clusters.e. it is clear that a clique in Gr embodies a feasible cluster.7/29/2011 A TABU SEARCH A ROACH In order to facilitate the usage of tabu search for CBP. Nodes that satisfy Constraint. j ) is defined in graph Gr between nodes i and j if they satisfy Constraints (8) and (9).. Consequently. i. 13 . ensure zone coverage. are called active nodes. A clique consists of a set of nodes that are adjacent to one another. a new graph called Gr is defined. The vertices of Gr represent the network nodes.

determine the content and size of tabu lists. 2. 3. design an algorithm that returns an initial solution. define the aspiration criteria. 4.7/29/2011 Five steps should be conducted in order to adapt tabu search heuristics to solve a particular problem: 1. define moves mðÞ that determine the neighborhood N(s) of a solution s. 14 . 5. design intensification and diversification mechanisms.

3. The maximal number of iterations.7/29/2011 The algorithm ends when one of the following three conditions occurs: 1. All possible moves are prohibited by the tabu lists. where the best solution is not enhanced successively. 2. 15 . has been reached. The maximal number of iterations allowed has been reached.

The algorithm determines the largest size clique that contains the selected active node i with its adjacent nodes in graph Gr. For each iteration. which have yet to be covered by F0. 16 . the first active node i. not yet covered by the initial solution F0. This clique is considered a new cluster and node I becomes the cluster head. is selected.7/29/2011 Initial Solution It starts sorting active nodes according to their degree in graph Gr decreasingly.

Consequently. this leads to the fact that an initial solution could not be feasible.7/29/2011 The algorithm does not ensure that all nonactive nodes are assigned to a cluster. However.. if node i is not covered by any cluster when the algorithm ends. nodes made up of at least one cluster does not consist of a clique in the graph Gr.e. i. it is assigned to a cluster whose head is adjacent to node i. 17 . A penalty equation to evaluate a solution is proposed in the following sections.

A Move Involving a Regular Node. i. Furthermore. This is due to the fact that an active node could be a cluster head and thus build a new cluster. A Move Involving a Cluster Head. 7/29/2011 18 .e. and the second move involves an active node. a nonactive node..The Neighborhood N(s) Definition Two types of moves are distinguished: the first move involves an ordinary node. A Move Involving an Active Node. the third move that involves a cluster head and allows removing an existing cluster from a solution is also considered.

19 .7/29/2011 COM UTATIONAL EX ERIENCE AND RESULTS In order to evaluate the performance of these novel algorithms. another approach was devised to solve the partitioning problem based on CPLEX [18] in order to compare the quality of the solutions found by tabu search. they were implemented with the use of C++ and the Boost Graph Library (BGL) [10] and tested with sensor networks of different sizes and topologies. several experiments were conducted to evaluate the impact of the tabu search method parameters. On the one hand. On the other hand.

a 1-Gbyte memory. and an Intel processor.7/29/2011 The algorithms run on a 1.80-GHz Pentium 4.4-2).4. 20 . equipped with a Linux server (Red Hat 3.

Analysis of the Impact of Tabu Search arameter 7/29/2011 Impact of the tabu list size on the solution cost. 21 . The best results are obtained using tabu lists whose size is similar to the number of nodes. between 75 and 120..e. i.

22 .7/29/2011 Impact of the maximum number of iterations on the solution cost.

7/29/2011 Impact of the maximum number of iterations on the execution time. 23 .

7/29/2011 Comparing Tabu Search Approach with C LEX-Based Method Comparing the solution costs (tabu search and CPLEX): Square topology. 24 .

25 .7/29/2011 Comparing execution time (tabu search and CPLEX): Square topology.

Comparing Tabu Search-Based to Simulated AnnealingBased Approaches 7/29/2011 Comparing the solution cost (tabu search and simulated annealing). 26 .

this centralized approach suffers from a major drawback linked to the additional costs generated by communicating the network node information and the time required to solve an optimization problem. Furthermore. a third approach based on simulated annealing heuristic. Finally. 27 . results show that a tabu search-based resolution method provides quality solutions in terms of cluster cost and execution time. compared to a distributed approach. The performance is compared to that obtained by a second resolution CPLEX-based method. Nevertheless. and an existing algorithm (TAG).CONCLUSION 7/29/2011 The performance of this novel approach was evaluated with different network sizes and topologies. it behaves well with network extensibility.

Sign up to vote on this title

UsefulNot useful- 017
- xin_Dcovpid
- 06497797
- Lee 2002
- IJRET_110306027
- Hit Manual
- Bertilsson Et Al 2011 Multidimensional Consideration of Anthropometric Diversity
- din molec
- Schaefer 1991 Closed-Form Solution Qualitative Equations
- Genetic Algorithm-based Clustering Technique
- Stochastic Modeling in Operations Research
- Lag Selection in Subset VAR Models
- Cyrus Montecarlo
- Kalman Filter for Control of Sattelites
- Automatic Simulation of Cyclic Voltammograms by the Adaptive Huber Method for Weakly Singular Second Kind Volterra Integral Equations 2010 Electrochimica Acta
- Measures Of Dispersion
- A Recursive Random Search Algorithm 1
- Operation Research
- 8
- R.N
- Blind and Total Identiﬁcation of ARMA M
- Hierarchical Clustering With Multi view point Based Similarity Measure
- Wp97 Hypersizer Optimization Approach
- EE-63
- Final Thesis Pk12
- Control Charts for Lognormal Data
- Cure Internet Verification Based on Image Processing
- Bch
- 50120130406008
- Tabu Search Final

Are you sure?

This action might not be possible to undo. Are you sure you want to continue?

We've moved you to where you read on your other device.

Get the full title to continue

Get the full title to continue reading from where you left off, or restart the preview.

scribd