Mirror PUF Bit Generation via Mismatch Magnitude Classification
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Bistable PUFs, such as SRAM-based PUFs, face high native bit instability across voltage and temperature conditions, leading to a significant energy overhead and high Bit Error Rate (BER), which complicates secure authentication and key generation in mobile and IoT devices.
Innovation Solution
The Mirror PUF concept post-processes preselection test results to utilize mismatch data as an uncorrelated entropy source, effectively doubling the number of stable bits without additional area or energy overhead, by classifying cells into 'low', 'medium', and 'high' mismatch categories and masking unstable bits, thereby reducing BER and increasing stability.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If preselection tests are used to measure mismatch within PUF cells, then the number of stable bits increases, but the energy overhead and complexity increase
Solution Approach 1:
The patent merges the preselection test function with the Mirror PUF generation function. The same PUF cell evaluations are used both to determine stability (for masking decisions) and to generate the Mirror PUF response bits. This combining of functions eliminates the need for separate test and operational phases, thereby reducing energy overhead while maintaining high stability.
Solution Approach 2:
The PUF cell evaluations serve multiple purposes: they determine the stability of bits for masking decisions and simultaneously generate the Mirror PUF response. This multi-functionality approach allows the system to achieve high reliability without proportionally increasing energy consumption, as the same hardware operations fulfill multiple objectives.
2Reliability
If multiple evaluations are conducted to increase stable bits, then the number of stable response bits increases, but the energy consumption increases proportionally
Solution Approach 1:
The patent combines the stability assessment process with the Mirror PUF generation process. Instead of conducting separate evaluations for stability testing and for key generation, the same PUF cell responses are used for both purposes. This merging eliminates redundant evaluations and reduces energy consumption while maintaining a high number of stable response bits.
Solution Approach 2:
The PUF cells themselves provide the stability information needed for masking decisions through their natural response variations. The system uses the inherent properties of the PUF cells (their response to challenges) to determine both stability and generate keys, without requiring external test equipment or additional energy-intensive measurement processes.
3Productivity
If all PUF cells are used to generate response bits, then the response length increases, but the Bit Error Rate increases due to unstable bits
Solution Approach 1:
The patent segments the PUF cell population into stable and unstable groups based on evaluation results. A masking mechanism is introduced that selectively includes only stable bits in the final Mirror PUF response. This segmentation approach maintains a long effective response length by including all stable bits while excluding unstable ones, thereby keeping the Bit Error Rate low.
Solution Approach 2:
The system performs preliminary evaluations of all PUF cells to determine their stability characteristics before generating the final Mirror PUF response. This preliminary action (stability assessment) allows the system to pre-identify which cells will contribute to stable responses, enabling the construction of a long, low-error response by selecting only from the stable subset of cells.
4Reliability
If Mirror PUF is implemented with additional processing, then the number of stable bits doubles, but the device complexity increases
Solution Approach 1:
The patent merges the stability assessment logic and the Mirror PUF generation logic into a unified processing framework. The same PUF cell evaluations that would normally be used for stability testing are simultaneously used to generate the Mirror PUF response bits. This merging of processing functions doubles the number of stable bits available while avoiding the complexity increase that would result from implementing completely separate test and operational systems.
Data Source
AI summary
A method for creating a physical unclonable function (PUF) bit for use with transistor circuitry includes performing a tilt test on a PUF cell of a transistor circuitry, comprising tilting the PUF cell at least once, and comparing a mismatch of a response of the PUF cell to a tilt threshold. A magnitude of the mismatch is determined. A mismatch magnitude below the tilt threshold is considered a first logic value” and a mismatch magnitude above the tilt threshold is considered a second logic value. The mismatch magnitude of the PUF cell is random. The absolute value of the mismatch magnitude is used as an entropy source to produce at least one PUF bit called a mirror PUF bit.