Dynamic Approach for Real time and Non-Real time Packet Scheduling in WSNs.

Abstract

Wireless sensor networks (WSNs) have a large kind of military and civil applications. Packet Scheduling is one in all the foremost essential functions of the nodes within the Wireless Sensor Networks. The sensor nodes are high-powered by batteries with restricted energy. Aggressive or enterprising environments wherever the sensor nodes are deployed or the sheared variety of the sensors prevents replacement or recharge of the batteries. So in WSNs, planning of various kinds of packets is of important energy consumptions and end-to-end task transmission delays. Most of the existing packet-scheduling mechanisms of WSN use First Come First Serve (FCFS), non preemptive priority and preemptive priority scheduling algorithms. These algorithms deserve a high process overhead and long end-to-end task transmission delay because of improper allocation of data packets to queues in multilevel queue scheduling algorithms. Moreover, these algorithms are not dynamic to the dynamic requirements of WSN applications since their planning policies are preset. During this paper, we tend to propose a Dynamic multilevel Priority (DMP) packet scheduling scheme. Within the proposed scheme, each node, except those at the last level of the virtual hierarchy within the zone- primarily based topology of WSN, has 3 levels of priority queues. real time packets are placed into the very high priority queue and may preempt data packets in alternative queues. Non-real-time packets are placed into 2 alternative queues supported a precise threshold of their calculable time interval. Leaf nodes have 2 queues for real time and non-real-time data packets since they are doing not receive task from alternative nodes and so, reduces end-to-end delay