Making a network that functions without any infrastructure, central administration
and routers is achieved through ad hoc mode of communication. Mobile Ad hoc
Networks (MANET) comprise a set of mobile nodes that independently move in any
direction, organize by itself, leave or connect with the network when required. Multihop
radio relaying, decentralized nature, low node density, less memory, less power,
light weight mobile devices and dynamic network configuration are the features of
MANETs which make it different from other wired and wireless networks. Thus routing
in MANETs happens to be one of the most challenging tasks. Most of the existing
routing protocols for MANETs are developed with the assumptions that there exists at
least one path between the source node and the destination node at all times throughout the packet forwarding process. Even if any situation arises in the network that the
nodes are misbehaving, switched off or moved out of the path, alternative paths can
be selected in order to precede the communication without any hindrance. Routing
protocols are developed with route repair and route maintenance mechanisms. But all
such protocols are designed only with the assumption that all the nodes in the network
are cooperative and at least one node is present at all the time in the transmission
range of the other. But these assumptions do not work well for some of the real-life
environments and such type of network is said to be intermittently connected MANET
(ICMAN). Routing of packets in intermittently connected network (Li and Rus, 2000;
Jain et al., 2004; Jones et al., 2005; Zhang, 2007; and Erramilli, 2008) is based on storecarry-
and-forward model. Suppose, if a node that is not connected to other nodes has
a message copy to forward, it stores and carries the message in its buffer until there
arises an appropriate communication opportunity. Many examples are given for such
networks in real life, including ecology monitoring, search and rescue operations,
vehicular ad hoc network, military networks, etc.
The paper is organized as follows. Section 2 explains the literature review.
Section 3 describes the node selection strategy along with the model calculations.
Section 4 explains the results obtained and the parameters used for the simulation.
Finally, the paper ends with conclusion.
|