A PUF device, chip, and PUF source code generation method

Abstract

This invention discloses a PUF device, chip, and source language generation method. The PUF device includes a single-electron transistor used to generate a random Coulomb oscillation curve when an excitation voltage signal is applied, which is then converted into a binary numerical sequence as the PUF source language. This invention introduces randomness from the principles of quantum / quasi-quantum devices, resulting in a higher level of security. It does not rely on the symmetry of the device array or comparisons between devices; a single device can realize the PUF function, resulting in an extremely simple structure. A single device can output a multi-bit random digital response under a single excitation, thus possessing natural multi-bit output, increasing source language density, and reducing array size requirements. The randomness originates from the non-periodic Coulomb oscillation induced by random defect states in the single-electron transistor, rather than the initial state of the device or thermal noise. Therefore, the randomness is stable and does not depend on the startup state or easily drifting physical quantities. This lays the device foundation for low-power, low-error PUF.

Description

Technical Field

[0001] This invention relates to the field of information security technology, and in particular to a PUF device, chip, and source language generation method. Background Technology

[0002] With the rapid development of the Internet of Things, cloud computing, artificial intelligence, and embedded systems, the deployment scale and application scenarios of integrated circuits in complex systems are constantly expanding. Device authentication, key generation, and secure communication have become fundamental issues in information system design. Traditional information security solutions mainly rely on software algorithm random number generators and key storage units. These solutions have inherent weaknesses when facing physical intrusion, side-channel attacks, and algorithm modeling attacks.

[0003] To overcome the aforementioned problems, Physically Unclonable Function (PUF) technology has been proposed and received widespread attention. PUF utilizes unavoidable process variations during chip manufacturing to create unique and uncopyable characteristic responses for each chip at the physical level. These characteristics can be used to generate chip identification, device authentication information, and encryption keys. Because PUF does not require long-term storage of keys within the chip, it offers significant advantages in resistance to physical attacks and copying.

[0004] Current PUF (Programmable Array of Elements) technologies are primarily based on Static Random Access Memory (SRAM) or Resistive Random Access Memory (RRAM) arrays. The randomness of these PUFs mainly stems from differences in the startup or resistive states of the memory cells. However, these technologies generally suffer from high raw bit error rates, require complex error correction modules, consume significant power, lack stable and repeatable physical reconstruction, and have relatively fixed excitation-response relationships. Furthermore, with advancements in data acquisition capabilities and machine learning algorithms, the risk of modeling attacks targeting these PUFs is gradually increasing.

[0005] Therefore, introducing new sources of randomness from the perspective of device physics principles to construct low-power, low-error-rate, and highly fuzzy and difficult-to-model PUF devices with excitation-response relationships is a technical problem that urgently needs to be solved in the field of hardware security. Summary of the Invention

[0006] To address the shortcomings of existing PUF technologies, this invention provides a PUF device, chip, and source language generation method. It introduces a new source of randomness from the perspective of device physics principles, thereby constructing a PUF technology with low power consumption, low bit error rate, and highly fuzzy and difficult-to-model stimulus-response relationship.

[0007] In one aspect, a PUF device is provided, including a single-electron transistor for generating a random coulombic oscillation curve when an excitation voltage signal is applied, to be converted into a binary numerical sequence as a PUF source.

[0008] Furthermore, the single-electron transistor includes: Electric island; The source and drain are located on opposite sides of the charged island, and are both electrically connected to the charged island through a tunnel junction to provide electron injection and output channels.

[0009] At least one gate, coupled to the charged island capacitor, is used to control single-electron transport behavior.

[0010] Furthermore, the at least one gate is one of a back gate, a partial gate, and a dual gate.

[0011] Furthermore, the single-electron transistor is fabricated using two-dimensional materials or nano-semiconductor materials.

[0012] Furthermore, it includes an array composed of multiple single-electro-transistors, which, when an excitation voltage signal is applied to the multiple single-electro-transistors, converts the random oscillation curves generated by the multiple single-electro-transistors into a binary numerical matrix as a PUF source.

[0013] Secondly, a chip is provided, including the PUF device as described above.

[0014] Thirdly, a PUF source language generation method is provided, based on the PUF device described above, including the following steps: An excitation voltage signal is applied to the gate of a single-electron transistor, and the random coulombic oscillation curve generated by the single-electron transistor is obtained. The random coulomb oscillation curve is sampled and converted into a binary numerical sequence based on a preset threshold, which is then used as the source code for PUF.

[0015] Furthermore, it also includes: By applying a preset perturbation to a single-electron transistor, and then re-executing the process of applying an excitation voltage signal, obtaining a random coulomb oscillation curve, and performing a binary conversion, the resulting multiple sets of binary numerical sequences are combined to form a numerical matrix as the source language of the PUF.

[0016] Furthermore, the preset disturbance includes at least one of the following: changing the source-drain bias voltage that does not affect the single-electron tunneling transport mechanism of the device, applying electromagnetic wave irradiation such as light or microwave to the single-electron transistor through an optical fiber or microwave conductor.

[0017] Furthermore, for a PUF device comprising multiple single-electron transistors, an excitation voltage signal is applied to each single-electron transistor, a random coulomb oscillation curve is obtained, and a binary conversion process is performed. The resulting multiple sets of binary numerical sequences are combined to form a numerical matrix as the PUF source code.

[0018] This invention proposes a PUF device, chip, and source language generation method, which generates PUF source languages ​​based on the random coulomb oscillation of a single-electron transistor, and has at least the following beneficial effects: (1) Introducing randomness from the principle of quantum / quasi-quantum devices results in a higher level of security; (2) It does not depend on the symmetry of the device array or the comparison between devices. A single device can realize the PUF function, and the structure is extremely simple. (3) A single device can output a multi-bit random digital response under a single excitation, thus possessing a natural multi-bit output, increasing source density, and reducing array size requirements; (4) The randomness originates from the non-periodic coulombic oscillation induced by random defect states in a single-electron transistor, rather than the initial state of the device or thermal noise. Therefore, the randomness is stable and does not depend on the start-up state or easily drifting physical quantities. (5) It lays the foundation for low power consumption and low bit error rate PUF. Attached Figure Description

[0019] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0020] Figure 1 This is an equivalent circuit diagram of a single monad transistor provided in an embodiment of the present invention; Figure 2 This is a schematic diagram illustrating the process of converting a random Coulomb oscillation curve into a binary numerical sequence, as provided in an embodiment of the present invention. V g Gate voltage, I ds This is the source leakage current.

[0021] In the diagram, 1 is the charged island, 2 is the source, 3 is the drain, 4 is the gate, and 5 is the tunnel junction. Detailed Implementation

[0022] To make the objectives, technical solutions, and advantages of this invention clearer, the technical solutions of this invention will be described in detail below. Obviously, the described embodiments are merely some embodiments of this invention, and not all embodiments. Based on the embodiments of this invention, all other implementation methods obtained by those skilled in the art without creative effort are within the scope of protection of this invention.

[0023] To address the limitations of existing PUF (Physically Unclonable Function) technologies in terms of randomness sources, safety margins, and power consumption, this paper proposes a novel physically unclonable function implementation scheme based on a single-electron transistor. A single-electron transistor is a type of electronic device operating at the quasi-quantum scale, characterized by electron tunneling transport as individual electrons under Coulomb blocking conditions. When the device size is reduced to the nanoscale and the corresponding capacitance and tunneling resistance conditions are met, the single-electron transistor exhibits Coulomb oscillation characteristics where the current varies with the gate voltage during gate voltage scanning.

[0024] Ideally, the Coulomb oscillation of a single-electron transistor exhibits good periodicity, with its period primarily determined by the device's geometry and capacitance parameters. However, in the actual fabrication of nanodevices, factors such as boundary roughness, etching damage, interface defects, and impurity atoms inevitably introduce localized defect states randomly into the charged islands and their neighborhoods. These defect states alter the localization behavior of electrons within the charged islands and modulate the single-electron transport behavior. Due to the randomness of the spatial location and energy distribution of these defect states, the device's equivalent capacitance parameter exhibits non-uniform characteristics with varying gate voltage, resulting in a non-periodic random distribution of the position, spacing, and shape of the Coulomb oscillation peaks, forming random Coulomb oscillation characteristics. This technique, based on these physical facts, utilizes the transport modulation effect introduced by random defect states to transform the Coulomb oscillation of a single-electron transistor from ideal periodic behavior into a stable but non-periodic random Coulomb oscillation. This random Coulomb oscillation exhibits good repeatability within the same device but shows significant differences between different devices, thus constituting a high-quality physical source suitable for PUFs.

[0025] Based on this, embodiments of the present invention provide a PUF device, including a single electronic transistor, for generating a random coulombic oscillation curve when an excitation voltage signal is applied, so as to convert it into a binary numerical sequence as a PUF source language.

[0026] Specifically, Figure 1 The diagram shown is an equivalent circuit diagram of a single-electron transistor, which includes: Charged island 1, formed of nanoscale conductive material, is used to locally confine a single or a few electrons, and its size satisfies the condition that the charging energy of a single electron is greater than the thermal energy. Source 2 and drain 3 are located on both sides of the charged island 1, and are electrically connected to the charged island 1 through tunnel junction 5 to provide electron injection and output channels. Tunnel junction 5 is located between source 2 / drain 3 and charged island 1. Its tunneling resistance and capacitance parameters satisfy the single electron transport condition. At least one gate 4 is capacitively coupled to the charged island 1 to regulate the number of electrons and chemical potential in the charged island 1, thereby controlling the single-electron transport behavior.

[0027] In specific implementation, the single-electron transistor is fabricated using two-dimensional materials or semiconductor thin-film materials, preferably using single-layer or few-layer graphene materials to form the charged islands and conductive channels. The size of the single-electron transistor should be smaller than the size that induces the quantum confinement effect, so that the electronic energy levels within the charged islands are discretized; the single-electron transistor operates under low-temperature conditions to ensure significant Coulomb blocking and Coulomb oscillation characteristics.

[0028] In this embodiment, the at least one gate 4 is one of a back gate, a partial gate, and a dual gate.

[0029] In some embodiments, the PUF device includes an array of multiple single-electro-transistors, which, when an excitation voltage signal is applied to the multiple single-electro-transistors, converts the random oscillation curves generated by the multiple single-electro-transistors into a binary numerical matrix as a PUF source language.

[0030] The PUF function of this PUF device is implemented as follows: Under a fixed source-drain bias, a preset gate voltage excitation is applied to a single-electron transistor, and multiple current characteristic points in its Coulomb oscillation curve are sampled. The sampled source-drain current signals are mapped to binary digital states according to a preset threshold, thereby obtaining a set of digital outputs. Since the Coulomb oscillation is a non-periodic random distribution, the correlation between different sampling points is low, resulting in a digital sequence with good randomness and uniqueness.

[0031] The PUF device provided in the above embodiments has the following technical features and effects: 1. Novel source of randomness: Randomness originates directly from random coulombic oscillations caused by defect state modulation during single-electron transport, rather than relying on the initial state of the storage cell or thermal noise; 2. High physical non-reproducibility: Random defect states are naturally introduced by the manufacturing process and are difficult to reproduce through reverse engineering or replication processes; 3. Low power consumption: The single-electron transistor has a small operating current, making it suitable for power-sensitive security applications. 4. Stability and repeatability: Under the same excitation conditions, the device can repeatedly output a consistent random response, meeting the reliability requirements of PUF; 5. Good scalability: This technical solution can be combined with array structure, key generation algorithm and reconstruction mechanism to form a PUF system with a higher level of security.

[0032] This invention also provides a chip, including the PUF device as described above.

[0033] Based on the PUF device disclosed in the foregoing embodiments, this invention also provides a PUF source language generation method, including the following steps: Under fixed source-drain bias conditions, an excitation voltage signal is applied to the gate of a single-electron transistor, and the random coulomb oscillation curve generated by the single-electron transistor is obtained. The random Coulomb oscillation curve is sampled, and the sampled point values ​​are compared with a preset threshold. Sampled points that are not less than the preset threshold are mapped to 1, and sampled points that are less than the preset threshold are mapped to 0. This converts the sampled point sequence into a multi-digit output of a binary numerical sequence, which can be used as the source code of PUF. Figure 2 The diagram shows the process of converting a random Coulomb oscillation curve into a binary numerical sequence.

[0034] To obtain richer PUF source terms, multiple sets of binary numerical sequences can be obtained by applying perturbations, thereby constructing a binary numerical matrix as the PUF source terms. Therefore, in some embodiments, the following is also included: By applying a preset perturbation to a single-electron transistor, and then re-executing the process of applying an excitation voltage signal, acquiring a random coulomb oscillation curve, and performing binary conversion, multiple sets of binary numerical sequences are combined to form a numerical matrix as the PUF source. The preset perturbation includes at least one of the following: changing the source-drain bias voltage (which does not affect the single-electron tunneling transport mechanism of the device), applying electromagnetic radiation such as light or microwaves to the single-electron transistor through an optical fiber or microwave conductor. It should be noted that the conditions of the preset perturbation should remain consistent during key generation and decryption to ensure the repeated output of a stable random numerical sequence under the same environmental and excitation conditions.

[0035] For a PUF device comprising multiple single-electron transistors, when generating the source code, an excitation voltage signal is applied to each single-electron transistor, a random coulomb oscillation curve is obtained, and a binary conversion process is performed. The resulting multiple sets of binary numerical sequences are combined to form a numerical matrix as the PUF source code.

[0036] Therefore, the present invention has the following beneficial effects: (1) Introducing randomness from the principle of quantum / quasi-quantum devices results in a higher level of security; (2) It does not depend on the symmetry of the device array or the comparison between devices. A single device can realize the PUF function, and the structure is extremely simple. (3) A single device can output a multi-bit random digital response under a single excitation, thus possessing a natural multi-bit output, increasing source density, and reducing array size requirements; (4) The randomness originates from the non-periodic coulombic oscillation induced by the defect state of the single-electron transistor, rather than the initial state of the device or thermal noise. Therefore, the randomness is stable and does not depend on the start-up state or easily drifting physical quantities. (5) It lays the foundation for low power consumption and low bit error rate PUF.

[0037] This invention can be applied to, but is not limited to, the following scenarios: chip-level identity authentication and anti-counterfeiting, secure boot and key generation, hardware security of IoT terminals and embedded systems, and security enhancement modules for data storage and computing systems.

[0038] It is understood that the same or similar parts in the above embodiments can be referred to each other, and the contents not described in detail in some embodiments can be referred to the same or similar contents in other embodiments.

[0039] Although embodiments of the present invention have been shown and described above, it is understood that the above embodiments are exemplary and should not be construed as limiting the present invention. Those skilled in the art can make changes, modifications, substitutions and variations to the above embodiments within the scope of the present invention.

Claims

1. A PUF device, characterized in that, This includes single-electron transistors, whose charged islands and their neighborhoods are randomly distributed with localized defect states naturally introduced during the fabrication process. These are used to generate non-periodic random Coulomb oscillation curves induced by random defect states when an excitation voltage signal is applied, which are then converted into binary numerical sequences as PUF source code.

2. The PUF device according to claim 1, characterized in that, The single-electron transistor includes: Electric island; The source and drain are located on opposite sides of the charged island, and are both electrically connected to the charged island through a tunnel junction to provide electron injection and output channels. At least one gate, coupled to the charged island capacitor, is used to control single-electron transport behavior.

3. The PUF device according to claim 2, characterized in that, The at least one gate is one of a back gate, a partial gate, and a dual gate.

4. The PUF device according to claim 2, characterized in that, The single-electron transistor is fabricated using two-dimensional materials or nano-semiconductor materials.

5. The PUF device according to any one of claims 1 to 4, characterized in that, It includes an array composed of multiple single electronic transistors, which, when an excitation voltage signal is applied to the multiple single electronic transistors, converts the random oscillation curves generated by the multiple single electronic transistors into a binary numerical matrix as a source language for PUF.

6. A chip, characterized in that, Includes the PUF device as described in any one of claims 1 to 5.

7. A method for generating PUF source code, characterized in that, Based on the PUF device as described in any one of claims 1 to 5, the implementation includes the following steps: An excitation voltage signal is applied to the gate of a single-electron transistor, and the random coulombic oscillation curve generated by the single-electron transistor is obtained. The random coulomb oscillation curve is sampled and converted into a binary numerical sequence based on a preset threshold, which is then used as the source code for PUF.

8. The PUF source code generation method according to claim 7, characterized in that, Also includes: By applying a preset perturbation to a single-electron transistor, and then re-executing the process of applying an excitation voltage signal, obtaining a random coulomb oscillation curve, and performing a binary conversion, the resulting multiple sets of binary numerical sequences are combined to form a numerical matrix as the source language of the PUF.

9. The PUF source code generation method according to claim 8, characterized in that, The preset disturbance includes at least one of the following: changing the source-drain bias voltage that does not affect the single-electron tunneling transport mechanism of the device, applying light to the single-electron transistor, and electromagnetic wave irradiation.

10. The PUF source code generation method according to claim 7, characterized in that, For a PUF device comprising multiple single-electron transistors, an excitation voltage signal is applied to each single-electron transistor, a random coulomb oscillation curve is obtained, and a binary conversion process is performed. The resulting multiple sets of binary numerical sequences are combined to form a numerical matrix as the PUF source code.

Images