Research On Tree Routing Topology And Energy Schemes In Wireless Sensor Networks

Wireless Sensor Networks (WSN) is a hotspot of information technology after Internet. It will have a profound influence on many areas in 21st century. WSN is a self-organized multi-hop network which is formed of a large number of cheap micro sensor nodes deployed around the areas by wireless channels. WSN is a novel and advanced network system, it takes on widespread application domains, such as military affairs, industry, environment monitor and medical treatment. The power and hardware resource of WSN nodes are very scarce, the operation and communication functions are very limited. Generally, it is hard to change battery for nodes. A key project in WSN is using the limited energy in a reasonable way in order to maximize the network lifetime. A efficient routing protocol which is power saving is a important technique to achieve the target.This dissertation is supported by the fund of Chongqing Science and Technology Committee (Subject No.2005BB2198). Routing protocol, energy efficient technique and implementation of WSN are the main parts of this dissertation. Based on the view of these, this dissertation makes the following contributions:①Analyzes and compares some WSN routing protocols. Based on the application demands of WSN, presents and summarizes the routing metric for prevalent routing protocols, analyzes characteristic for each protocol, points out the importance and necessity of energy metric for WSN routing protocol.②Based on the WSN network stacks, analyzes some power-saving techniques. Pointed out the reasons of"energy hole"and"hotspot node"in WSN. Deduce the relationship about number geographical positions of the base node and the transmitting load of nodes in different areas. On the assumption of nodes being uniformly distributed in networks, it is demonstrated that more base nodes, uniformity distributing of base nodes, and more base nodes located in the center of monitor area can balance the transmitting energy consumption of network. Advises some schemes to reduce the transmitting load caused by many-to-one data traffic of tree routing topology.③Based on the framework of distributed tree building routing algorithm, and combined with power-aware, data aggregation and relay nodes, this dissertation designed a energy efficient routing protocol. The protocol adopted the technique which combined by Minimum Transmission (MT) metric and the residual power of nodes. The energy efficient protocol integrated the excellences of table-driven and source-initiated on-demand routing. The routing message sending of the protocol used the event-driven and time-driven to reduce cost for the routing maintenance. Protocol makes use of data aggregation to reduce the cost of communication, the efficiency of message header can be improved. Flooding routing, distributed tree building routing , and energy efficient routing are simulated by 40 nodes. Simulation results show that the energy efficient routing protocol performs much better than flooding routing and distributed tree building routing. In the energy efficient protocol, the average transmit energy consumption of node is only 28.38% of distributed tree building routing algorithm, the max transmit energy consumption is only 15.56% of distributed tree building routing algorithm, the reception rate of nodes compared with that of distributed tree building routing algorithm is less than 2%, reliability is not decreased by data aggregation. Energy efficient protocol can reduce power consumption of nodes and improve unbalance power consumption of nodes effectively. It is also proved that the new algorithm has good reliability.④The energy efficient routing protocol is implemented in embedded operation system TinyOS by nesC language. This dissertation expatiates the principle, implementation and approach for the energy efficient routing algorithm in TinyOS. Based on the Micaz platform and the MIB600 gateway, the new protocol which includes data aggregation is implemented by the application of collecting temperature. User acquires data by ethernet through socket interface. The new algorithm under the compiler of TinyOS operation system are 420 bytes ROM and 50 byte RAM more than the distributed tree building routing algorithm.