A node scheduling based on partition for WSN

A node scheduling based on partition was proposed for wireless sensor network (WSN). At first, the monitoring area was partitioned into equal virtual grid. The algorithm defined two working modes: a monitoring mode and a tracking mode. The effective lifetime that the WSN can work effectively (not the lifetime before all nodes die) were stressed. So the coverage and connectivity of the network was discussed and the schedulable node set was provided in which the nodes can work in turn while the coverage and connectivity of the network was kept. Then a sleep scheduling strategy was designed to monitoring interest situation and a waking up scheduling strategy was presented to track unusual event. For monitoring mode, only one node was assigned to work in each grid in accordance with a scheduling cycle to reduce the energy consumption. The monitoring scheduling strategy could not only prolong the lifetime of WSN but also balance the work time among the nodes. In addition, the failure rate was introduced to improve the system reliability. For tracking mode, more nodes were woken up to get more precise data and avoid losing the target since there are coverage holes in some cases. A wakeup sensor threshold and a sleep time threshold were introduced to control the energy consumption as little as possible. The proposed algorithm and PEAS were simulated by NS2 with the network scale from 200 to 1000 nodes. The simulation results show that the proposed algorithm performed much better than PEAS both in energy consumption and the network effective lifetime. The number of wakeup nodes along the event trail and their working time show that the proposed method could track the event accurately while conserving energy. As a result, the proposed algorithm can extend the effective lifetime of the network almost twice as PEAS.