A dynamic scheduling model for MANETs using order statistics

Though providing Quality of Service (QoS) for MANET is a challenge, it becomes a necessity because of its applications in critical scenarios. Users belonging to various profiles use MANETs for various purposes. Sometimes, even a low profiled user may need to send an urgent message in time critical applications. Hence providing prioritization based on user profile and urgency is necessary. In this paper we propose a Dynamic Differentiated, Delay threshold based priority model and MAC protocol for MANETs (3D-MAC). Our contributions are, calculation of dynamic urgency index to cater to urgent traffic, a novel scheduling model using order statistics to avoid starvation of low priority nodes, adjustment of contention window size during collisions to avoid packet drops and method to improve throughput of low priority, while the network has only low priority nodes. The model is simulated in ns2. We compare our results with the IEEE802.11e and justify that, our model gives 16% more average throughput during collision. Further, when Low priority nodes are only present in the network, their throughput is increased by 23%.