Research On Wireless Sensor Network Service In Cloud Computing

Based on wireless sensor network (WSN) application services and the advantage of cloud platform resource, a service architecture of WSN on cloud environment is proposed in the paper. By the architecture, three experiments are performed for managing and refactoring data, optimizing and minting network coverage, recognition and detection holes.In details, the main content of this paper includes:1. The management architecture of WSN on cloud environment. This study reviews characteristics of cloud platform and WSN. A three-tier service architecture on Hadoop is proposed, including resource layer, platform layer and application layer. In this architecture, our research is focus on the application layer which provides different application interface on Hadoop.2. The data storage and reconstruction of WSN on cloud environment. Three data transmission models (real-time, timer and quantitative) are designed for different applications. For better data storage, three policies (single, logical and block file) are proposed in cloud environment. For the space-time characteristics of WSN, four methods of data reconstruction are proposed including node, time, position and mixed reconstruction. The result of experiments show that reconstruction of cloud computing is more efficient than the stand-alone computing.3. The coverage control and monitoring of WSN on cloud environment. To optimize problem of coverage in deployment and operation of WSN, the parallel genetic algorithm is proposed based on cloud environment by adjusting node state and energy level. For meeting requirements of operation and maintenance in WSN, a parallel data statistical method is proposed based on network monitoring of WSN. Experimental results show that high precision and efficiency are achieved.4. The monitor of holes in WSN based on cloud environment. Discovering network holes is important for WSN. Based on the node degrees and global network information, rules are designed for discovering holes and equidistant line algorithm is proposed on cloud environment. Using historical information and abnormal states, a recognition algorithm is proposed for discovering holes. The results show that the position of the hole is characterized quickly and accurately.