The Ieee802.15.4 Wireless Sensor Network Performance Analysis And Improvement Research

The rapid development of sensor wireless communication technology promotes the appearance and evolution of Wireless Sensor Network(WSN). It is generally accepted that the definition of WSN includes large scale, self-organization, multi-hop and no infrastructure supporting. WSN node is always structure-identical, low cost and small size. The nodes are randomly distributed in the monitoring area and collect data. It is required that WSN must have a long life time.The IEEE 802.15.4 standard is released by the IEEE 802.15 group 4, which is suitable for Low-Rate Wireless Personal Area Network(LR-WPAN). The standard defines LR-WPAN's PHY layer and MAC layer. The standard with the features of low rate, low power, short communication range and simple structure is very fit for WSN. The performance of IEEE 802.15.4 standard in a large-scale WSN application is analyzed. To minimize the energy consumption of the entire network and to allow adequate network coverage, IEEE 802.15.4 peer-to-peer topology is selected, and configured to a beacon-enabled cluster-tree structure. The analysis consists of models for CSMA-CA mechanism and MAC operations specified by IEEE 802.15.4. The performances of device and coordinator are analyzed in terms of power consumption and throughput, and the effect of SO and BO for the network performance is analyzed. The analysis results are verified with simulations using OPNET.Then we apply 2.4GHz IEEE802.15.4 compliant RF transceiver CC2420 design a wireless sensor network system, including the design and implementation of node and route protocol. And we analyze the energy consumption of our node. In order to balance the great difference of consumption between device and coordinator in the network, we propose a new clustering algorithm, called Energy and Maximal Independent Set Based distributed algorithm for minimum connected set(EMISB). The algorithm utilizes node's energy status and 2-hop neighbor information to select cluster head, other nodes will join the cluster which head is the nearest one. Simulation results show that, compared to the other clustering algorithm, the new algorithm can prolong the lifetime of the network evidently while balancing energy consumption and increasing energy efficiency.