BinaryInferno: A Semantic-Driven Approach to Field Inference for Binary Message Formats.
Jared Chandler, Adam Wick, and Kathleen Fisher.
In Proceedings of the Symposium on Network and Distributed System Security (To Appear: NDSS’23), 2023. PDF

Abstract: We present BinaryInferno, a fully automatic tool for reverse engineering binary message formats. Given a set of messages with the same format, the tool uses an ensemble of detectors to infer a collection of partial descriptions and then automatically integrates the partial descriptions into a semantically-meaningful description that can be used to parse future packets with the same format. As its ensemble, BinaryInferno uses a modular and extensible set of targeted detectors, including detectors for identifying atomic data types such as IEEE floats, timestamps, and integer length fields; for finding boundaries between adjacent fields using Shannon entropy; and for discovering variable-length sequences by searching for common serialization idioms. We evaluate BinaryInferno's performance on sets of packets drawn from 10 binary protocols. Our semantic-driven approach significantly decreases false positive rates and increases precision when compared to the previous state of the art. For top-level protocols we identify field boundaries with an average precision of 0.69, an average recall of 0.73, and an average false positive rate of 0.04, significantly outperforming five other state-of-the-art protocol reverse engineering tools on the same data sets: AWRE (0.18, 0.03, 0.04), FIELDHUNTER (0.68, 0.37, 0.01), NEMESYS (0.31, 0.44, 0.11), NETPLIER (0.29, 0.75, 0.22), and NETZOB (0.57, 0.42, 0.03). We believe our improvements in precision and false positive rates represent what our target user most wants: semantically meaningful descriptions with fewer false positives.

Deceptive Self-Attack for Cyber-Defense.
Jared Chandler, and Adam Wick.
In Proceedings of the 56th Hawaii International Conference on System Sciences, 2023. PDF

Abstract: The asymmetry between cyber-defense and cyber-offense is well-known; defenders must perfectly protect their systems, while attackers need only find one flaw. Defensive cyber-deception has been proposed as a way to mitigate this problem, by using various techniques designed to require attackers to defend themselves from misdirection, false data, and counter-attack. In this paper, we propose a new cyber-deception technique: deceptive self-attack (DSA). DSA modifies network and systems to give the appearance that an unknown third party is also at work attacking the same systems. It is our contention that the presence of this (deceptive) adversary pressures real adversaries in novel ways useful to cyber-defense; and discuss these effects. As a study in DSA, we present and evaluate SoundTheAlarm, a SMT-solver based system for generating deceptive self-attack network traffic. SoundTheAlarm uses public attack signatures from the Suricata intrusion detection system to automatically generate network traffic consistent with a particular cyber-attack signature.

Invasion of the botnet snatchers: A case study in applied malware cyberdeception.
Jared Chandler, Kathleen Fisher, Erin Chapman, Eric Davis, and Adam Wick.
In Proceedings of the 53rd Hawaii International Conference on System Sciences, 2020. PDF

Abstract: In this paper, we provide the initial steps towards a botnet deception mechanism, which we call 2face. 2face provides deception capabilities in both directions – upward, to the command and control (CnC) server, and downward, towards the botnet nodes – to provide administrators with the tools they need to discover and eradicate an infestation within their network without alerting the botnet owner that they have been discovered. The key to 2face is a set of mechanisms for rapidly reverse engineering the protocols used within a botnet. The resulting protocol descriptions can then be used with the 2face network deception tool to generate high-quality deceptive messaging, against the attacker. As context for our work, we show how 2face can be used to help reverse engineer and then generate deceptive traffic for the Mirai protocol. We also discuss how this work could be extended to address future threats.