Lifetime Maximization in Wireless Sensor Networks Using a Mobile Sink with Nonzero Traveling Time

It is a known phenomenon that in a wireless sensor network, sensors communicating directly with a sink deplete their energy faster than the others. As a result, the so-called neighboring sensors can die, disconnecting some of the sinks from the rest of the network, even though most of the sensors ar...

Full description

Saved in:
Bibliographic Details
Published in:Computer journal 2011-11, Vol.54 (12), p.1987-1999
Main Authors: Keskin, M. E., Altinel, I. K., Aras, N., Ersoy, C.
Format: Article
Language:English
Subjects:
Computer simulation
Disengaging
Linear programming
Mathematical models
Mathematical programming
Maximization
Mobility
Networks
Remedies
Sensors
Studies
Wireless communication
Wireless networks
Citations: Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:It is a known phenomenon that in a wireless sensor network, sensors communicating directly with a sink deplete their energy faster than the others. As a result, the so-called neighboring sensors can die, disconnecting some of the sinks from the rest of the network, even though most of the sensors are still fully functional. One possible remedy is to balance the relaying load of the sensors using mobile sinks and controlling their mobility, which has attracted the interest of researchers. In this work, we extend the relevant literature by introducing two new mathematical programming models. They intend to maximize the network lifetime through the controlled mobility of a sink with nonzero travel times with and without limiting the number of hops by which the data originating from the sensors reach the sink. Both models allow more than one tour of the sink during the network lifetime and determine the optimal sink route and sojourn times. Since the models are computationally difficult to solve, we propose efficient heuristics methods to compute near-optimal solutions. On the basis of the computational results performed on randomly generated problem instances, we can say that their performance is remarkable. [PUBLICATION ABSTRACT]
ISSN:0010-4620
1460-2067
DOI:10.1093/comjnl/bxr048