The batch steganography problem is how best to split a steganographic payload between multiple covers. This paper makes some progress towards an information-theoretic analysis of batch steganography by describing a novel mathematical abstraction we call perturbation hiding. As well as providing a new challenge for information hiding research, it brings into focus the information asymmetry in steganalysis of multiple objects: Kerckhoffs' Principle must be interpreted carefully. Our main result is the solution of the perturbation hiding problem for a certain class of distributions, and the implication for batch steganographic embedding. However, numerical computations show that the result does not hold for all distributions, and we provide some additional asymptotic results to help explore the problem more widely. © 2008 Springer Berlin Heidelberg.
CITATION STYLE
Ker, A. D. (2008). Perturbation hiding and the batch steganography problem. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (Vol. 5284 LNCS, pp. 45–59). Springer Verlag. https://doi.org/10.1007/978-3-540-88961-8_4
Mendeley helps you to discover research relevant for your work.