Study On Routing Of Water Environment Monitoring Network For The Three-Gorges Reservoir Area

The Three Gorges Project of Yangtze River is one of the most remarkable projects around the world. It brings enormous economic and social benefits on the aspect of flood control, electricity generation and shipping et al. Thus, Three Gorges reservoir water environmental safety has been highly concerned and became a hot spot in research area. Since existing monitoring method and equipment can’t satisfy the demand, it is urgent to adopt new ideas and devices to enhance the monitoring ability about reservoir water environmental.Wireless sensor networks consist of large numbers of cheap and low-power sensor nodes, which can form a multi-hop self-organizing network system, have the abilities of sensing data from the monitoring area, data processing and transmitting. When deployed into the Three Gorges reservoir, sensor nodes can construct a water environment monitoring network that efficiently and continuously monitor the water quality in real time. This kind of monitoring method has important theoretical and practical significance for improving our country’s ability of water environment monitoring.Numerous sensor nodes are deployed into the water environment monitoring area, and each one is powered by battery which is inconvenient to be recharged. Thus it is crucial to manage the energy of a node to save its energy and prolong the lifetime of the monitoring network system. Usually, communication model of a sensor node consumes the majority of energy, and then its energy efficiency can be improved by optimizing the routing protocol for the network. This thesis takes Three Gorges reservoir water environment monitoring as application background and combines the zonal distribution characteristics, devotes to study cluster-based energy saving routing protocols for wireless sensor networks. The main contributions are listed as follows:(1) A game theoretical cluster-based routing protocol is proposed. To construct the cluster topology, each sensor node only needs to exchange messages with its neighbors. Thus this protocol is completely distributed and can adapt to zonal distribution networks.(2) A punishment mechanism is adopted in the clustering game when modeling the cluster head selection problem, and residual energy of a node is a determinant factor to the punishment intensity. After the game, each sensor node gets an equilibrium probability to be the cluster head, and decides whether to be a cluster head according to this probability so that a balance can be achieved between saving energy and providing network services.(3) For a sensor node, if it is successfully selected as a cluster head based on its equilibrium probability, it acts as a tentative cluster head and will further compete for the final cluster head. The main objectives of this method are avoiding neighbor nodes become the cluster head simultaneously and getting a wide area covered by all cluster heads.(4) A cost function is designed according to the degree of cluster heads and the distance to the base station. Each normal node chooses its own cluster head based on this function so that energy consumption among cluster heads can be balanced.(5) Simulation experiments are implemented to evaluate the performance of the proposed protocol, and compared with classical clustering protocol LEACH and game theoretical clustering protocol CROSS, LGCA.