Font Size: a A A

Research And Implementation Of XML Compression Algorithm And Security

Posted on:2010-11-10Degree:MasterType:Thesis
Country:ChinaCandidate:Z S TongFull Text:PDF
GTID:2178360272991529Subject:Computer application technology
Abstract/Summary:PDF Full Text Request
With the trend of market globalization, the competition of market keeps increasingly fierce. Thus, the cross-sectoral and inter-regional manufacturing business model has evolved. ERP(Enterprise Resource Planning) is a cross-platform, internetized business model and management tool, which serves companies. ERP system uses the ideas of information technology and advanced management. It comes to be a modern business model for the widely use of global enterprises. In ERP, heteromerous data will lead to difficulties of data loading between different applications. Therefore, A2A and B2B data exchange in ERP often uses XML as data carrier.This article gets started from the large amount of XML data and high security demand in ERP system. It focuses on the problem of network bandwidth caused by the structural redundancy of XML. With the research of XML compression algorithms and security technology in ERP environment, the work of this paper is as following:One weakpoint of existing queriable XML compression algorithms, such as XGrind, is that the encoding relies on DTD files, which may not be available in ERP system. Another weakpoint is that XGrind has to scan XML file twice, which may impact the compression efficient. To conquer these problems, this paper proposed a queriable XML compression algorithm (named ELZ) based on Lempel-Ziv. No DTD file input is needed, so that it is suitable for XML files without DTD in ERP. ELZ scans XML file once, which will reduce the processing time. The algorithm uses the same-structure conversion strategy, and keeps the structure and semantics of compressed XML document same as the original one. The compression of data section depends on data type.To improve the compression ratio of ELZ, another compression algorithm EHF was proposed. It uses the ideas of Huffman encoding. Frequency of XML elements/attributes is analysed before compression, and a Huffman tree is created. Thus, EHF generates better code than ELZ, and increases compression ratio by 5%~10%.For XML data confidentiality, this paper analysed the public key cryptography algorithm, CRT-RSA. It proved that: Wiener's continued fractions attack does not work on CRT-RSA. For data reliability, a way to verify XML is designed based on Merkle.Finally, an XML data exchange system was implemented to use the compression algorithms and security technologies stated above. It is tested and applied in SAP internal management system. It does not use DTD which is the bottleneck of XML compression and decreases 30% compression time with comparison to XGrind.
Keywords/Search Tags:ERP, XML Compression, XPath, CRT-RSA, Continued Fraction Attack
PDF Full Text Request
Related items