On evidence preservation requirements for forensic-ready systems

Forensic readiness denotes the capability of a system to support digital forensic investigations of potential, known incidents by preserving in advance data that could serve as evidence explaining how an incident occurred. Given the increasing rate at which (potentially criminal) incidents occur, designing software systems that are forensic-ready can facilitate and reduce the costs of digital forensic investigations. However, to date, little or no attention has been given to how forensic-ready software systems can be designed systematically. In this paper we propose to explicitly represent evidence preservation requirements prescribing preservation of the minimal amount of data that would be relevant to a future digital investigation. We formalise evidence preservation requirements and propose an approach for synthesising specifications for systems to meet these requirements. We present our prototype implementation— based on a satisfiability solver and a logic-based learner—which we use to evaluate our approach, applying it to two digital forensic corpora. Our evaluation suggests that our approach preserves relevant data that could support hypotheses of potential incidents. Moreover, it enables significant reduction in the volume of data that would need to be examined during an investigation.