In this paper, we use ns2 to simulate our new routing methods. We compare the delivery ratio in four different routing methods and analysis the impacts that affect the delivery ratio. We choose our new routing, named OOPFE-Routing and has been accepted in ICCNT2012 [8], with other three routing methods to analysis the issues affecting routing performance. We will specially focus on the packet-drop-problems that don't consider the network security and don't depend on buffer queue in delay tolerant networks. From our experimental results, the number of delivery ratio of OOPFE is even better than Epidemic-Routing which is high-bound in theory. The interesting reason is the overhead of OOPFE lower than Epidemic. We will research about the reason of overhead. There are three major reasons; the first reason is the numbers of repeatedly received packets. The second reason is the numbers of packets dropped by our setting, when the size of buffer-queue is less than 100%. The last reason is the numbers of packets dropped by other five types of reason in the routing time. Furthermore, we compare the detail reasons in the collision reason from four different routings and especially made a special analysis for multiple copy packets to understand the characteristics of the different routing protocols.
Relation:
Journal of Convergence Information Technology 8(4), pp.1-11
