Mobile Database Key Technologies

In pace with rapid development and increasingly maturation in wireless network technology, mobile computing and mobile data process plays a more and more important role. The technology of mobile database makes it possible for people to access data at any place and any time. The technology of mobile database opens a new field to the database application and is of important significance in the development of the parallel and distributed technology. However, we are still facing some new challenges because of the mobile nodes, low bandwidth and disconnection of wireless network. In current mobile database system, functions are not very effective, too many hypotheses and limitations exist in most of mechanism and algorithm and some key functions cannot be supported.This paper mainly studys the technologies of route protocols in Ad Hoc network, data caching and query optimizing in mobile database system. This paper presents a new group route algorithm based on Ad Hoc network, which partitions the nodes to some groups, broadcasts the request of route to different groups, consequently finds out an near optimized path rapidly.The algorithm lightens the load on network, and lays a foundation for mobile database communication. This paper brings forward a dynamic group cache mechanism, which fully considers data access frequency, validity and consistency of data, and which reduces the network load and improve the response speed by reducing the distance of data transmission. This paper puts forward two basic querying algorithms: one is single node query algorithm based on cache, and the other is double node join algorithms based on cost model. Single node query algorithm improves the querying efficiency by the full use of cache in group. Double node join algorithm makes out three query projects. Node dynamically chooses query project according to the query costand network condition. In the end, based on the above algorithms, the paper realizes a mobile distributed database prototype system - MDDB and gives experiment results and performance analyses on single node query algorithm based on group cach and double node join algorithm.