PUF Value Generation Using Anti-Fuse Array
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing Physical Unclonable Function (PUF) systems face challenges in generating consistently random binary values that are unique to each chip, making them vulnerable to duplication even if the chip layout is reverse-engineered.
Innovation Solution
A PUF value generation system utilizing a pair of dielectric breakdown-based anti-fuse memory cells and a current limiting circuit, where one anti-fuse memory cell randomly programs first, creating a conductive link while inhibiting the other from programming, leveraging the random nature of dielectric breakdown for unique and reliable random bit generation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Manufacturing precision
If a PUF system uses dielectric breakdown-based anti-fuse memory cells to generate random values, then the randomness and uniqueness of the generated codes are improved, but the reliability of consistent code generation across voltage and temperature variations deteriorates
Solution Approach 1:
The patent uses dielectric breakdown as a physical phenomenon that occurs at specific voltage thresholds. By applying a programming voltage that exceeds the breakdown threshold, the system creates a permanent conductive link in the anti-fuse memory cell. This parameter-based approach ensures that once breakdown occurs, the cell state remains stable across subsequent voltage and temperature variations, resolving the contradiction between randomness generation and reliable code consistency.
2Productivity
If the system applies high voltage to program anti-fuse memory cells simultaneously, then the programming speed is improved, but the risk of both cells in a pair programming (reducing randomness) increases
Solution Approach 1:
The patent applies programming voltage to multiple anti-fuse memory cells simultaneously in a pair, but relies on the preliminary random variation in dielectric breakdown timing inherent in the manufacturing process. The system does not control which cell breaks down first, but rather exploits the natural random timing difference. This preliminary action approach allows simultaneous programming while maintaining randomness, as the breakdown event in each cell is predetermined by manufacturing variations rather than controlled by the system.
Solution Approach 2:
The patent incorporates a read-verify operation that checks the state of anti-fuse memory cells after programming. This feedback mechanism detects whether one or both cells in a pair have programmed, and if both have programmed (indicating loss of randomness), the system can identify and handle such cases. The feedback loop ensures that only pairs with exactly one programmed cell are used for PUF code generation, maintaining the required randomness.
3Reliability
If the system uses complex circuitry to ensure one cell programs before the other, then the reliability of random bit generation is improved, but the device complexity increases
Solution Approach 1:
The patent exploits the natural random timing difference in dielectric breakdown between cells in a pair, without requiring additional control circuitry to enforce the sequencing. The physical phenomenon of dielectric breakdown inherently provides the random ordering needed, and the system simply observes and utilizes this self-generated randomness. This self-service approach maintains high reliability of random bit generation while avoiding the complexity of additional control mechanisms.
Solution Approach 2:
The patent uses identical anti-fuse memory cell structures in pairs, with both cells having the same design and connectivity. This copying approach simplifies the overall system design, as the randomness emerges from the natural variations in identical structures rather than requiring differentiated complex circuitry. The uniform cell design reduces device complexity while maintaining reliable random bit generation through the inherent variations in breakdown timing.
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
The system ensures reliable and repeatable generation of random values, resistant to duplication, with high randomness and uniqueness, maintaining stability across various conditions like temperature and voltage.
Implementation Method 1
dielectric breakdown-based anti-fuse memory cells
Data Source
AI summary
A method and system are used to generate random values for Physical Unclonable Function (PUF) for use in cryptographic applications. A PUF value generation apparatus comprises two dielectric breakdown based anti-fuses and at least one current limiting circuit connected between anti-fuses and power rails. Two anti-fuses are connected in parallel for value generation in programming by applying high voltage to both anti-fuses at the same time. Time for dielectric breakdown under high voltage stress is of random nature and therefore unique for each anti-fuse cell. Therefore the random time to breakdown causes one cell to break before another, causing high breakdown current through the broken cell. Once high breakdown current through one broken or programmed cell is established, a voltage drop across a current limiting circuit leads to decreased voltage across both cells, thereby slowing the time dependent breakdown process in the second cell and preventing it from breakage under programming conditions.


