Tunable stochastic memristors for energy-efficient encryption and computing.

Area of Science:

• Materials Science
• Computer Engineering
• Information Security

Background:

• Post-digital computation faces challenges in balancing security (unpredictability) and computing (predictability).
• Current solutions require separate, power-intensive silicon-based hardware for these opposing needs.

Purpose of the Study:

• To develop a single, flexible hardware solution for both information security and computing.
• To demonstrate a memristor-based system capable of cryptographic and logical operations.

Main Methods:

• Fabrication and characterization of Cu0.3Te0.7/HfO2 (CuTeHO) ion-migration-driven memristors.
• Operation of CuTeHO memristors under varying biases to achieve random number generation and Boolean logic.
• Experimental demonstration of an integrated system for cryptographic key generation, logic operations, and encryption/decryption.

Main Results:

• CuTeHO memristors exhibit dual functionality: generating physically unclonable functions (randomness) and performing universal Boolean logic.
• A single system integrating these memristors successfully performs cryptographic key generation, logic operations, and encryption/decryption.
• Circuit-level analysis indicates significant energy and latency advantages over conventional hardware.

Conclusions:

• Memristors offer functional flexibility for implementing diverse computational requirements.
• The demonstrated CuTeHO memristors provide an energy-efficient, integrated solution for advanced information security and computing.
• This work paves the way for novel hardware architectures in post-digital computation.