| With the rapid development of communication technology, the future wirelessnetwork which is a heterogeneous system composed by different RATs,has overlappedcoverage, complementary technical features and a variety of business needs, and etc. Assignificant differences in service capabilities of different wireless networks, how toimplement the coordination of heterogeneous networks and the complementation of avariety of resources, which of the advanced resource management techniques and policystrategy are needed to integrate various radio networks have become one of the hotreaserch issues of future communication system, and also been the objective of thisdissertation.From the system, schemes and algorithms, this dissertation proposes a new radioresource management mechanism based on the autonomous policy strategy. The key ofthe scheme is that it presents an approach for generating the optimal policy based oncase reasoning, which can be used for radio resource management and optimization inan autonomous and efficient mode in heterogeneous wireless networks. It searches forthe most appropriate reuse policy in the case library by utilizing case retrieval andmatching algorithms, and improves the new policy’s efficiency and quality by utilizingpolicy reuse algorithm based on the probability of similarity. The approach alsointroduces the greedy algorithm for generating coping policies when this is no caseavailable. In addition, it real-time updates the policy case library based on the learning.Simulation results show that the proposed approach owns efficient online learningability. |