2nd National Conference on Innovative Paradigms in Engineering and Technology (NCIPET 2013) |
Foundation of Computer Science USA |
NCIPET - Number 11 |
March 2012 |
Authors: Jawade Nilima R, Bawane N G, Dongre Manoj M |
8425db9c-2a71-4e2b-9dac-12188e270937 |
Jawade Nilima R, Bawane N G, Dongre Manoj M . QOS Improvement by Load Balancing using Nodal and Neighboring Load in Multipath Routing. 2nd National Conference on Innovative Paradigms in Engineering and Technology (NCIPET 2013). NCIPET, 11 (March 2012), 1-6.
The Mobile Ad hoc NETwork (MANET) is a collection of wireless mobile nodes. These mobile nodes do not have any fixed network infrastructure or centralized network administration. IP routing protocols determine a route for the data packet so that it should travel across the network and reach the intended destination. The table driven routing protocol pre compute the routes initially, which become stale or invalid due to high mobility. while the on-demand routing protocols concentrate their routing load on few centrally located nodes. One similarity among them is that both use shortest path as their route selection criterion. The heavily loaded nodes in the network lead to congestion and bottlenecks because of dynamic network topology, energy limitation and bandwidth. Frequent link breakages occur due to high mobility. One solution to this problem is to have a routing strategy that could balance and distribute the traffic load more evenly through each mobile node. Existing on-demand ad hoc routing protocols such as AODV and AOMDV does not provide load balancing. Here we propose a load balancing scheme which utilizes the nodal and neighboring information to select a route for the packet to travel across the network to balance the load evenly and avoid congestion and bottlenecks. The load metrics utilized are the number of active paths the node support along with its queue length. As the load is balanced efficiently in the proposed methodology, it improves the packet delivery ratio, throughput and network life time. The overhead of the advertised messages like RREQ is reduced by forwarding them selectively.