Random Bit Generator Using Punch-Through Current Stability
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing Physical Unclonable Function (PUF) circuits are susceptible to environmental conditions and aging, leading to instability and traceability of generated random numbers, making them less secure for information protection.
Innovation Solution
A random bit generator utilizing a voltage source, bit data cells with transistors experiencing channel length variations, and a sensing control circuit to output random bits based on punch-through currents, which are stable and repeatable due to intrinsic transistor characteristics, independent of ambient conditions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If PUF circuits use minor variations in transistor performance from manufacturing process, then random bits can be generated, but the generated random numbers become unstable due to susceptibility to environmental conditions
Solution Approach 1:
The patent changes the operating parameter regime by using transistors in the velocity saturation region rather than subthreshold or linear regions. This is achieved by applying specific voltage conditions (Vgs > Vth + ΔV) that force the transistors into velocity saturation, where carrier velocity is limited by scattering mechanisms rather than electric field strength. This parameter change makes the current primarily dependent on physical dimensions (channel length, width) rather than environmental factors like temperature or voltage variations.
Solution Approach 2:
The patent employs a differential current comparison mechanism analogous to hydraulic balance. Two identical transistor pairs are configured in parallel with their sources connected together and drains connected to separate bit lines. By comparing the currents through these symmetric structures (I1 vs I2), the circuit exploits manufacturing variations while canceling out environmental effects that equally affect both transistors. The sensing circuit detects which transistor conducts more current, generating a stable random bit.
2Productivity
If PUF circuits generate random numbers by changing internal transistor structures, then random bits are produced, but the circuits suffer degradation due to aging and the random numbers may be traced
Solution Approach 1:
The patent implements a periodic enroll operation where random bits are generated on-demand rather than continuously. The circuit enters a low-power standby state between enroll operations, periodically refreshing the random bits when needed. This periodic activation reduces cumulative stress and aging effects on the transistors compared to continuous operation, extending the circuit's functional lifespan while maintaining random number generation capability.
Solution Approach 2:
The patent treats the random bit generation as a consumable resource that is periodically refreshed. Each enroll operation generates new random bits that can be used immediately, and the process can be repeated to generate fresh bits. This approach accepts that individual transistor characteristics may drift over time but compensates by periodically regenerating the random numbers, effectively making the random bit supply renewable rather than relying on permanent structural stability.
3Loss of information
If PUF circuits rely on small transistor variations, then random bits can be generated, but the random numbers become traceable through structural analysis
Solution Approach 1:
The patent makes the random number generation dynamic and state-dependent rather than static. The enroll operation requires applying specific control signals to select and activate particular transistor pairs, and the generated random bits are stored in latches or registers that can be updated. This dynamic operation means the same physical transistor variations produce different observable outputs depending on the activation sequence and timing, making structural analysis and tracing significantly more difficult compared to static PUF implementations.
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 solution provides stable and repeatable random bits suitable for security applications, resistant to environmental influences and aging, enhancing information security by generating unique and untraceable random numbers.
Implementation Method 1
A random bit generator utilizing a voltage source, bit data cells with transistors experiencing channel length variations, and a sensing control circuit to output random bits based on punch-through currents
Data Source
Figure 1
Figure 2
Figure 3
AI summary
A random bit generator (100) includes a voltage source (110), a bit data cell (120), and a sensing control circuit (130). The voltage source (110) provides a scan voltage (VS) during enroll operations. The data cell (120) includes a first transistor (122A) and a second transistor (122B). The first transistor (122A) has a first terminal coupled to a first bit line (BL1), a second terminal coupled to the voltage source (110), and a control terminal. The second transistor (122B) has a first terminal coupled to a second bit line (BL2), a second terminal coupled to the voltage source (110), and a control terminal. The sensing control circuit (130) is coupled to the first bit line (BL1) and the second bit line (BL2), and outputs a random bit data according to currents generated through the first transistor (122A) and the second transistor (122B) during an enroll operation.