System-level chip security design method based on untrusted IP core

Abstract

The invention relates to a system-level chip security design method based on an untrusted IP core. The method comprises the following steps: firstly, completing security design of an RTL by adopting ahigh-order synthesis technology; converting the RTL file of the chip into a gate-level netlist through a series of logic synthesis; analyzing the logic structure of the gate-level netlist, and extracting the characteristics of the standard netlist; dividing the collected standard netlist feature samples of the gate-level circuit into a training data set and a test data set, and training a gradient lifting algorithm by adopting the training data set to obtain a hardware Trojan horse classifier based on gradient lifting; and finally, inputting the test data set into the trained gradient liftingTrojan classifier through a cross validation method to obtain a hardware Trojan prediction result based on the model. According to the invention, an SoC security design architecture with strong adaptability, high reusability, wide expansibility, rapidness and intelligence is designed from the source of an IP supply chain.

Description

technical field [0001] The invention relates to the field of integrated circuit information security, in particular to a system-level chip security design method based on an untrustworthy IP core. Background technique [0002] With the development of wireless communication, sensor technology, embedded system and microelectronic technology, the issue of integrated circuit information security is facing unprecedented challenges. [0003] At present, under the trend of globalization of the semiconductor supply chain, the integrated circuit industry needs to use third-party Intellectual Property (3PIP) IP cores designed / manufactured in different countries or regions to complete the System-on-Chip (System-on-Chip, 3PIP) IP core. : SoC) production. The traditional strategy of treating the underlying hardware as a safe and reliable security protection has failed, and untrusted suppliers can directly or indirectly implant malicious circuits into various stages of IC production. Do...

AI Technical Summary

Problems solved by technology

But the disadvantage is that the mixed use of multiple IP cores will increase additional technical overhead, and the design of Trojan horse isolation and detection also increases production costs, and this method fails to consider information leakage Trojan horse detection

[0007] Due to the wide ap

Images