Research On Optimizing Energy Consumption In Wireless Sensor Network

With the rapid development of wireless technology, sensor network, as a boominginformation collection technology in recent years, has been increasingly given a great amountof attention in many fields.Comparing with the traditional wireless network, WSN has some particular characteristic:energy constraints, redundant low-rate data, limited transmission range and computingcapability. Energy efficiency is the main performance metric of WSN. The main factors ofenergy consumption are data sensing, data processing and radio communication. Due tolimitations of physical layer, protocol design and implement on network layer and data linklayer is the most effective way to reduce energy consumption.The main work and innovation of this paper are summarized as follows.In view of the defects of the back-off mechanism in SMAC protocol, DMAC(Dynamic-MAC) that introduce the fast collision resolution is proposed. SMAC is analgorithm based on the Carrier Sense Multiple Access (CSMA) work pattern. Along with thechange of data flow, DMAC can change the size of contention window exponentially. Nodesreduce the contention window size to enlarge the nodes’ sleep time when the data flowdecreased. In contrast, while the data flow increased, nodes enlarge contention window size toreduce the probability of data collision. It can save the energy consumption effectively.DMAC combine with both CSMA and time division multiple access (TDMA) workpattern. In the CSMA pattern, when the size of contention window greater than a threshold andthe there are still some collision in the network, DMAC turn into TDMA work pattern. It cantotally avoid the data collision and re-emitted. It also has several advantages, such as zerodelay of access channel and high utilization rate of bandwidth. The results of simulationexperiment on the platform OMNeT++indicate that our strategy, compared to the SMAC andTHMAC, can prolong network lifetime and reduce the end-to-end delay.Clustering can be used in the routing algorithm to enhance the scalability in WirelessSensor Network. A drawback, almost clustering algorithms have, is that the nodes cannotcommunicate with others who are out of the cluster after its head node dead. This defect mayresult in that the information collected by nodes cannot be transmitted to sink and surplus energy in the network is unbalanced. According to this defect, this paper proposed analgorithm called EDC (energy-efficient dual-pathed clustering). Nodes can transmit withothers even after their head nodes dead.All of nodes in the network have a backup path in the EDC. It can transmit data throughsuch paths after its head node dead. The principles of allotting the backup nodes are as followsby priority: other head node that has the greatest surplus energy in the radio range; the node inother cluster that has the greatest surplus energy in the radio range; the node in the samecluster that has the greatest surplus energy in the radio range. Extensive simulations onOMNET++have been performed to validate the performance of the EDC strategy.