PUF Random Code Generation Using Random Dopant Implant Patterns
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
PUF-based random number generators face challenges in consistently generating random binary numbers due to manufacturing process variations and environmental conditions like temperature and voltage changes, which affect the performance parameters of devices in the array.
Innovation Solution
A semiconductor layer with an array of devices, where some devices are within a predetermined performance parameter range and others are outside it, achieved through randomly patterned dopant implant regions, ensuring consistent and stable random number generation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Manufacturing precision
If manufacturing process variations are minimized, then device performance consistency is improved, but random number generation capability deteriorates
Solution Approach 1:
The patent intentionally introduces dopant implant regions that change the performance parameters (threshold voltage, resistance) of selected devices outside the normal manufacturing specification range. This creates deliberate parameter variations that enable random number generation while maintaining manufacturing control.
Solution Approach 2:
The patent applies dopant implantation selectively to specific devices within the array, creating local variations in device characteristics. Some devices remain within specification while others are intentionally modified, creating the necessary diversity for PUF functionality without compromising overall manufacturing quality.
2Stability of the object's composition
If process variations are reduced, then device uniformity is improved, but random code generation reliability deteriorates
Solution Approach 1:
The patent creates local variations by selectively implanting dopants in specific devices, making them non-uniform compared to standard devices. This local non-uniformity is intentional and provides the basis for reliable random code generation while the rest of the array maintains uniformity.
Solution Approach 2:
The patent deliberately changes performance parameters of selected devices through dopant implantation, creating a controlled non-uniformity that enables reliable random number generation. The modified devices have threshold voltages or resistance values outside the normal range, providing stable binary states.
3Manufacturing precision
If performance parameter range is narrowed, then manufacturing precision is improved, but random number generation stability deteriorates
Solution Approach 1:
The patent intentionally expands the performance parameter range for selected devices through dopant implantation. Devices with dopant regions have threshold voltages or resistance values outside the normal manufacturing range, creating stable binary states (0 or 1) that are less sensitive to environmental variations.
Solution Approach 2:
The patent applies parameter changes locally to specific devices rather than uniformly across the array. This creates a mixed population of devices with different characteristic ranges, where modified devices provide stable random states while unmodified devices maintain normal operation.
4Ease of manufacture
If devices are made identical, then manufacturing complexity is reduced, but random code generation capability deteriorates
Solution Approach 1:
The patent maintains identical fabrication processes for all devices but introduces local variations through selective dopant implantation. The base manufacturing process remains simple and uniform, while random code generation capability is achieved through localized modifications to specific devices in the array.
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 random distribution of devices with significantly different performance parameters ensures consistent and stable random multi-bit binary number generation, even under varying conditions, suitable for cryptographic applications.
Implementation Method 1
Random distribution of the first and second devices can be achieved in the disclosed embodiment of the structure by, for example, including randomly patterned dopant implant regions in the semiconductor layer
Data Source
AI summary
Disclosed is a structure for implementing a Physically Unclonable Function (PUF)-based random number generator and a method for forming the structure. The structure includes same-type, same-design devices in a semiconductor layer. While values of a performance parameter exhibited by some devices (i.e., first devices) are within a range established based on the design, values of the same performance parameter exhibited by other devices (i.e., second devices) is outside that range. A random distribution of the first and second devices is achieved by including randomly patterned dopant implant regions in the semiconductor layer. Each first device is separated from the dopant implant regions such that its performance parameter value is within the range and each second device has a junction with dopant implant region(s) such that its performance parameter value is outside the range or vice versa. A random number generator can be operably connected to the devices to generate a PUF-based random number.