Research and Design of Smart Home System Based on Cloud Services
|School||Hangzhou University of Electronic Science and Technology|
|Course||Electronics and Communication Engineering|
|Keywords||smart home cloud computing Eucalyptus Hadoop|
With the development of smart home technology and the popularity of the products, theperformance of intelligence and comfort is far from enough for people’s currently requirements, themain problems are: the complicated installation, more bottlenecks of data, the low resourceutilization rate and the poor real-time performance. By studying the architecture of smart homesystem based on cloud-service, a smart home scheme of remote control is given in this paper, whichcan provide the most service with the least amount of resources and the lowest cost takingadvantage of the high availability, multi-service, and low cost of cloud-computing.Firstly in this paper, technologies related to smart home system and cloud-computing arestudied, the problems appeared in the development of smart home technology are investigated, andthe advantage of cloud-computing is analyzed. On this basis, the docking of smart home system andcloud-computing illustrates how to solve the problems related to smart home system, making use ofthe advantages of cloud-computing. Secondly, the paper introduces the key technologies of cloudcomputing such as virtualization, studies the three different service modes provided by the cloudcomputing platform, which are IaaS, PaaS and SaaS, and designs a kind of cloud services systemfor smart home. Again, it analyzes the main platforms using cloud-computing technology, choosesthe advantaged open source cloud computing platform, namely Eucalyptus, as the infrastructureplatform of the cloud service system, and implements a basic cloud service system after studyingEucalyptus and it’s components such as CLC, Walrus, CC, SC, and NC. Then, according to thearchitecture and service mode of cloud service, a cloud service system suitable for smart homesystem is selected, and various modes of service are realized through system construction, mirrorframe and expansion of the platform mirror frame. Finally, the paper puts forward a cloud servicesystem for smart home which combined with Eucalyptus and Hadoop, studies different platforms,thus improving it. Using the Hadoop distributed storage computing platform to expand the datastorage for virtual machines in cloud service system, and developing the Eucalyptus and Hadoop, itoffers the developer application interfaces for cloud applications, and providing users with cloudservices.After various implementations of the platform, ultimately available cloud service system fit forsmart home system is presented, it gives users the IaaS, PaaS, and SaaS services through deployingthe smart home system on each virtual machine of cloud service system. It not only solves problemssuch as the multifarious installation operation in house, the requirements of large data storage, andhigh-speed processing capacity requirements; also improves the system resource utilization, theaccuracy and real-time performance of remote control; addition, reduces the cost of smart home greatly, because the cloud service system is built with cheap hardware platform, and the scalabilityof cloud services system brings convenience for later research.