CHEST color logo
CHEST University logos image of a line

CHEST 2021 Research Project Abstracts


P22_21: Do you Trust Your Standard Cell Library?
Trojan Resilience Untrusted Cell Library Analysis, Detection and Mitigation

Topic Areas: New Approaches to Secure On-Shore Microelectronics Design and Manufacturing
Principal investigator: Dr. Houman Homayoun, University of California Davis
PI Email


Abstract:
As of 2020, the fabless integrated circuit (IC) manufacturing model accounted for approximately one third of all integrated circuits sold by the top 50 semiconductor manufacturers. This model has consistently gained popularity over the last decade [1]. In a fabless model, designers are decoupled from the foundries which fabricate the ICs in separate (often untrusted) facilities. For designers to rapidly prepare a design for the foundry, they must be provided technology files and cell libraries to interface with electronic design automation (EDA) tools. Previous works have addressed the potential security risks of manufacturers embedding covert hardware Trojan horse devices in a manufacturer’s critical IP to actively create a catastrophic error in the design or passively spy on the design as it operates [2] [3]. However, to our knowledge, no body of work exists that has thoroughly addressed the risks of using untrusted foundry provided cell libraries. Cell libraries contain the standard cells and other macros that nearly all modern ICs are comprised of. Each cell contains many files that describe cell topography, top-level interface, and electrical characteristics. While beneficial to design flow efficiency, trusting the validity of these cell descriptions leaves ample room for malicious cell manipulation. Inaccurate cell description files can present exaggerated power characteristics to cloak extraneous devices. Cell level manipulation could be scaled to create more complex embedded hardware Trojan horses and payloads. As a defense against these threats, we propose the development of sophisticated implementations of such attacks, and a complete strategy to identify and address compromised library cells. The completion of this work will enable IC manufacturers to continue to leverage the benefits of robust cell libraries without introducing critical design security vulnerabilities.

Back