Abstract
Wireless cloud computing is presented as a new computing paradigm which integrates cloud computing into wireless environment. It takes full advantages of cloud computing which has potential to transform a large part of the IT industry, making hardware or software even more attractive as a service and shaping the way in which hardware is purchased and designed. The objectives of wireless cloud computing are to satisfy more users demand, efficiently utilize a pool of resources including wireless network, storage and computation resources and optimize energy on wireless devices. We examine new wireless cloud computing technologies, and indicate the main issues and challenges for their development in the future, among which resource management problem is presented. Combining the current agent architectures and resource optimization strategies, we present an agent-based resource management (ABRM) to deal with multiple data and computation intensive applications of users demand. It offers a promising solution of selecting the best service provider, saving energy on a wireless device and efficiently utilized wireless network resource under user’s request constraint. In this thesis, we mainly accomplish three research issues. Firstly, we solve the problem of selecting wireless cloud service provider in wireless cloud environment. We present combined analytic hierarchy process (AHP) and grey relation analysis (GRA) method for guaranteed wireless users being always best connected to wireless cloud. This method achieves the best available wireless cloud service provider will be selected to meet users’ demand and maintain the high service performance. Secondly, we address the energy optimization problem on wireless devices. We present adaptive offloading strategy to extend storage space and computation capability as well as saving battery power for performing a number of computation intensive applications on a wireless device. The grouped job strategy is employed to partition meta-jobs. The fuzzy control model is presented for saving energy on wireless device. The network-aware sufferage heuristic algorithm is employed for saving wireless device idle power via virtual machines which belong to service provider. Thirdly, we solve the problem of delivering multimedia content in wireless cloud environment. We present convex optimization algorithm for efficiently utilizing wireless network resources. The Lagrange duality model is used to model the wireless network optimization problem. Also, a weighted server provisioning algorithm is employed for balancing the content server throughput, which aims to satisfy more users’ requirement with minimum bandwidth. Then, continuous multimedia application service will be maintained.