Querying XML data based on improved prefix encoding

XML has become the de facto standard of data presentation and exchange on the Web and Internet. Efficiently store and query XML documents have been the primary urgent task. At present, most of XML index and query are based on encoding the XML document tree. XML encoding is that node or edge of XML document tree is encoded with the only encoding according to some method. Through the encoding, structural relation can be judged and the XML document tree needn´t be traversed. So XML structure query is often transformed into structure join calculation. In order to support query especially structure query, all kinds of XML encoding schemes are proposed. And most of XML query are based on prefix encoding schemes. Prefix encoding is based on path and completely keeps the absolute path from root to target node, then it is convenient to judge structural relation. This paper analyzes and studies some existing prefix encoding schemes, which have long codes when node depth is excessive deep and the document tree has more children or more siblings. Then excessively long codes need extra memory space and matching of structural join. So as to enhance encoding performance and query efficiency, and to lessen encoding length, therefore, by analyzing the current prefix encoding schemes, IPE encoding (Improved Prefix Encoding) is proposes in this paper. IPE encoding is an improved XML prefix encoding. With this encoding, we can reduce the length of prefix encoding and conveniently judge structural relation, when XML document tree has more nodes. Based on IPE encoding and inverted index tables, path query algorithm is presented; accordingly query speed is improved effectively.