Research On Data Transmission Optimization Of Wireless Sensor Networks

Wireless Sensor Networks is a network consists of sensor nodes which have the ability of sensing, computing and communication.The wireless sensor network can fix up fast and the nodes are inexpensive. The correlation between nodes is low, and they can make the network automatically. It has a strong ability to adapt to the environment. So it has a very wide range of applications in military, civil, communication and so on. However, the energy of wireless sensor nodes are often limited, thery use the battery as the energy source generally. Moreover, transmission energy consumption is relatively large in the nodes. How to reduce the energy consumption and prolong the network life cycle is the hotspot of the research of wireless sensor networks.Firstly, the chain process of data aggregation algorithm based on the chain structure using greedy algorithm,it can greatly reduce the transmission distance and reduce the energy consumption. However, there are some problems in the link selection, and the delay of the data is large. MSDARP algorithm proposed in this paper is a multi-chain structure rather than single-chain but the idea of single-chain is still used. MSDARP aims to solve the problem of energy consumption and large delay due to unreasonable link allocation. It can be proved that the algorithm can reduce the energy consumption of the network.Secondly, the paper proposes a data aggregation algorithm (Data Density-Based LEACH, DDB-LEACH), which is based on the traditional LEACH algorithm.The cluster head selection methods in LEACH algorithm and some of its improved algorithms are very rigid, cluster heads can not play a good role.In the case of data traffic distribution is not well-distributed, the nodes in the higher part of the data flow will have high intensity of work and die early. Thus will reduce the overall network life. The DDB-LEACH algorithm will give more attention to the nodes in the high traffic area and grasp the overall network. Thus selecting the cluster head and divide the cluster according to the current network traffic state.