Reversible Dna Information Hiding

In this paper, we address two ways of difference expansion based reversible DNA watermarking using noncoding DNA sequence. Unlike image data, the reversible DNA watermarking should consider the string structure of a DNA sequence, the organism functionality, the perfect recovery, and the high embedding capacity. We convert the string sequence of four characters in noncoding region to the decimal coded values and embed the watermark bit into coded values by using pairs of neighbor coded values and consecutive DE-MBE (C-DE-MBE) using the previous embedded coded values as the current estimated one. Using sequences of bacterium and archaea types, experimental results verified that our approaches have more high embedding capacity than conventional methods and produce no false start codon and recover perfectly the host sequence without the reference sequence. Especially C-DE-MBE can embed more high two times than DE-MBE.