Energy-balanced And Efficient Routing Algorithms In Wireless Sensor Networks

Wireless sensor network (WSN), as a new networking technology, consists of alarge amount of sensor nodes with limited resource, nodes that are randomly deployedin the target region through collaborating with each other.to accomplish various taskssuch as data acquisition and transmission, event detection and so forth. It is very criticalfor WSN by efficiently utilizing the limited resource to attain special functions, whichare required in different applications. The performances of routing protocols, likeenergy saving, energy balancing and load balancing, have an important effect on thelifetime of a network, especially when in monitoring applications steady data streamsare transmitted continuously. Therefore, my dissertation focuses on designingenergy-balanced and efficient routing algorithms to improve the energy utilization andeventually prolong the lifetime of networks applied in above relevant scenarios. Twosuch routing algorithms are developed through my three-year research work:①Aiming at the deficiency of uneven distribution of cluster heads (CHs), energyand load in sensor networks by using classical clustering routing protocols, like LEACH,I proposed an energy efficient uniform tree-clustering routing protocol (EUCR) forWSN. This algorithm firstly emended the threshold T(n) in LEACH by taking node’sreceived signal strength indication, residual energy and number of neighbors intoaccount and computing the optimal number of CHs in a network, and therefore avoidcircumstances in which minuscule cluster was formed and low-energy node became acluster head. After the preliminary election of CHs by using the improved threshold, thedistribution of CHs was further amended according to clustering radius and appropriatedistance between them, which assured the uniform distribution of CHs and a balancedscale between clusters. Secondly, an optimizing routing tree was constructed, basing onthe hops and residual energy, to make CHs transmit data in multi-hop, helping reducethe energy consumption. The strategy of engendering an assistant CH and of CHalternation within clusters evenly allocated load and energy consumption among clustermembers and saved expenditure of reconstruction in every round. Simulation resultsshow that EUCR has better energy efficiency and load balancing comparing to LEACH,and prolongs the network lifetime.②When EUCR was used in applications with continuous and steady data streams,the energy consumption and load of different CHs can’t be well balanced by merely adopting the simple routing tree of CHs in EUCR because of a large amount ofcommunication tasks, which leaded to a serial of problems like uneven energydistribution, hot spots, energy hole, packet loss caused by overloading and energyconsumption of retransmission. Aiming at these issues, a distributed multipleattribute-based energy-balance routing algorithm (DMAE) for WSN was proposed.Introducing a load prediction mechanism and total transmission energy consumptionTEC, this algorithm constructed an information table of forward neighbors for eachnode, including forward neighbor ID, residual energy, comprehensive load and TEC.Every node then by using relative entropy calculated the weights of these attributes andestablished the model of multiple attribute decision making. According to the model,each node forwarded its data by selecting the optimal forward neighbor to avoid pathloopback, disperse data streams in network, balance energy consumption of all nodesand ease the problem of “hot spots”. Simulation results show that DMAE has betterenergy-balancing effect, improves the energy efficiency of network and prolongs thenetwork lifetime.