Pseudo-random number generator (PRNG) combined with text based watermarking for cryptography application / Lew Chee Hon

• Main Author: Lew , Chee Hon
• Format: Thesis
• Published: 2017
• Subjects: QA75 Electronic computers. Computer science; QA76 Computer software
• Online Access: http://studentsrepo.um.edu.my/11962/2/Lew_Chee_Hon.pdf; http://studentsrepo.um.edu.my/11962/1/Lew_Chee_Hon.pdf; http://studentsrepo.um.edu.my/11962/

Text digital watermarking,which is an important research direction for information hiding, can be used for copyright protection of text documents. Digital text protection received increasing attention in the last decade due to massive digital data documents via internet, hence digital watermarking was considered as a potential solution for solving cryptography problems since many E-commerce applications protect their data transactions through the internet. These transactions include sensitive document transfers, digital signature authentications, digital watermarking for data security protection, and digital data storage and linkages. This thesis proposed a digital text based watermarking algorithm combined with Pseudo-Random Number Generator(PRNG) for cryptography application. Related work was surveyed in digital watermarking, cryptography and design methodology before implementing the text based watermarking method (embedded and extract/detection of watermarks).The implementation results on text based digital watermarking combined with cryptographic techniques are presented. This thesis intends to provide a reference finding for newcomers in security designs and to promote more activities in these security issues. This experiment demonstrates that better accuracy of extracted watermark due to text based watermarking is obtained when combined with PRNG. The accuracy of the extracted watermark under text with PRNG exceeded 97%. The sentence selected for embedding watermarking bit are 85% higher in speed due to faster embedding watermarking bit string.The relation of information-hiding capacity is 1:5.5 which means that for every 5.5 terms of text, one bit of watermark can be hidden data. A comparison was made between a PRNG generator with the proposed RSA key generator and of text based watermarking in the proposed work. Therefore, security is improved as it is difficult to detect watermark (imperceptibility) when combined with the developed PRNG.