Implementation of Energy-Efficient Routing Protocols for Mobile Ad hoc Networks (MANET
No Thumbnail Available
Date
2004-08
Authors
Journal Title
Journal ISSN
Volume Title
Publisher
Addis Ababa University
Abstract
The nodes in a mobile ad hoc network also form its routing infrastructure. Previous research
shows that the idle power consumption in the nodes is significant, as the network interfaces on
them is always on in order to maintain the routing fabric. As mobile nodes are dependent on
battery power, there is a need for protocols that minimize energy consumption. In this thesis, the
Dynamic Source Routing protocol used for on-demand routing in an ad hoc network is modified
to reduce the power consumption in nodes by adaptively putting their interfaces to sleep.
In an ad hoc network, it is impossible to predict accurately when it is all right for a node to put its
network interface to sleep, using only its own information. In the approach presented, the time
slot during which the interface is on is alternated with a time slot during which the interface is
put to sleep. The duration of the on period depends only on indigenously available information
about the number of packets the interface receives during this time slot. In the absence of any
network activity in the on slots, the sleep period is linearly increased up to a maximum. The
report explains all the factors that can affect the performance of the modified routing protocol
and its influence on the energy consumption in the network. The penalty of increase in delay and
packet loss is unavoidable and the levels of the factors are identified to minimize the penalty.
The modified protocol is implemented in the Ns-2 network simulator. A linear equation is used
to model the energy consumption for each node in the network. Simulations are conducted to test
the modified protocol and the factors varied to study their impact. The results are compared with
those obtained from the simulations using the unmodified DSR protocol. The results show
average energy savings per-packet of up to 25% with an average of 2-3ms per-packet increase in
the delay. The packet loss is comparable to the unmodified DSR protocol.