Research On XML Data Broadcast In Wireless Environments

With the rapid development of wireless network technologies, users with mobile devices can access a large amount of information at anytime from anywhere. Broadcast, as an efficient way for public information delivery to a large number of listeners, offers great scalability, good power consumption, and efficient bandwidth utilization. Most of the current broadcast researches focus on broadcasting data items with unique key values. The requests are key-based queries and the indexing methods are also key-based. To retrieve information from the broadcast, users must know the item key in advance.On the other hand, XML has rapidly gained popularity as the de facto standard to represent such information in data transmission and exchange. As a result, broadcasting XML data in wireless networks is a natural choice for disseminating XML data to a large number of mobile clients. However, traditional air indexing techniques and broadcast schedule techniques can not be applied to XML data broadcast directly since XML data are semi-structured.In this paper, we study XML data broadcast problems in wireless environments. We propose a new set of techniques to solve the problems. Our major contributions can be described as follows: 1. we have designed two basic structures of air index in XML data broadcast; 2. we have proposed air index pruning techniques to eliminate redundant data in the basic indexes; 3. we have proposed air index combining techniques to further eliminate redundant data between the basic indexes; 4. we have proposed air index building techniques and optimizations to accelerate the building process of air index; 5. we have proposed efficient XML data broadcast scheduling techniques: XML document pruning technique and XML document splitting technique. Our preliminary experiments show that the basic air index structures can support Mobile Units' XPath queries. Using the air index pruning technique and combining technique, we can build concise air index for XML data broadcast. Using air index building techniques and optimizations, we can build air index more efficiently. And using our broadcast scheduling techniques, we can significantly optimize the organizations of XML data on the wireless channel.