A smart-gentry based software system for secret program execution

Currently generic executable programs can only be encrypted during transmission and storage. To execute the program itself and the data it operates on must be decrypted. If the execution system is not trusted or compromised, both the program code and data are endangered. Recent advances in homomorphic cryptography show how additions and multiplications can be executed in encrypted space, i.e. without decrypting the information, the arithmetic operations themselves are not encrypted. To date, a universal implementation of a homomorphic system, capable of executing arbitrary programs and allowing for practical experiences is still missing. In this paper we present the first method to compute a non-linear arbitrary secret program on an untrusted resource using fully homomorphic encrypted circuits. We use our own implementation of the Smart-Gentry crypto-system as a foundation and define a processor architecture which is capable of executing encrypted programs on encrypted data. Unlike other approaches, such as static one-pass boolean circuit simulations, our system supports read and write memory access, dynamic parameters and non-linear programs, that render branch-decisions at runtime and cannot be represented in a circuit with hard-wired in-circuit parameters and data. Our implementation comprises the runtime environment for an encrypted program and an assembler to generate the encrypted machine code. The system represents a first step to show the capabilities of homomorphic encryption in software and system architecture.