A Chaos-Based Arithmetic Logic Unit and Implications for Obfuscation

It is no secret that modern computer systems are vulnerable to threats such as side-channel attack or reverse engineering whereby sensitive data or code could be unintentionally leaked to an adversary. It is the premise of this work that the mitigation of such security threats can be achieved by leveraging the inherent complexity of emerging chaos-based computing (computer systems built from chaotic oscillators). More specifically, this paper considers a chaos-based arithmetic logic unit which consists of many unique implementations for each possible operation. Generalizing to a chaos-based computer, a large number of implementations per operation can enable the obfuscation of critical code or data. In such a system, any two functionally equivalent operations are unique in terms of control parameters, power profiles, and so on. Furthermore, many possible implementations for each operational code can be leveraged to compile a program that is uniquely defined in terms of what the user knows -- such knowledge which itself could be protected via encryption. The frequencies of the various operations are shown to approach that of a probabilistic system as the circuit is allowed to evolve in time. Further, the difficulty of a successful attack is assumed to be directly related to the number of unique op-code sets possible which is shown to grow exponentially with allowed evolution time for the proposed chaos-based arithmetic logic unit.