An Efficient Address Assignment Mechanism for Mobile Ad-Hoc Networks

Abstract

In a mobile ad-hoc network (MANET), in order to route a packet between any nodes, nodes should have their unique IP address in the network. In our previous work, we introduced and compared three IP assignment mechanisms, namely RADA, LiA, and LiACR. In RADA, a randomly-selected IP address in a specified address space is assigned to a joining node, which results in poor utilization of the address space with a great deal of address conflict. LiA allows a joining node to be assigned to the current maximum address + 1, that is, linearly from the address space. Although LiA utilizes the address space better, it takes a long time to complete address assignments due to address conflict in case that several joining nodes require the same IP address. LiACR allows simultaneously joining nodes to have their IP addresses according to the node ID-based order. However, it relies on reliable exchange of control messages in the wireless network. Since broadcasting is inherently unreliable, we therefore propose an enhanced version of the LiACR protocol, called E-LiACR, which copes with the unreliable broadcasting through the help of neighbor nodes. Through ns-2 simulations, we show that E-LiACR performs better than LiACR in terms of IP address allocation time, number of address conflicts, and control message overhead.

This work was supported by the Korea Research Foundation Grant funded by the Korean Government (MOEHRD) (KRF-2005-003-D00295).