This paper introduces a novel authentication scheme for advanced metering infrastructure of smart grid using ring oscillator physically unclonable functions (ROPUFs). The scope of the design covers the communication between the utility company and the smart meter network. The scheme is based on hardware-oriented security and can be implemented on existing smart meters. Authentication keys of 64 bits are used for security level 1 communication. The size of keys can increase up to 1024 bits for security level 5. The cryptographic keys used in this scheme are extremely difficult to model by adversaries as each key is used only once. These authentication keys are generated from ROPUF response bits using the Hamming code. The design of the scheme ensures fault tolerance as ring oscillator comparison pairs with high frequency differences are selected to prevent bit flips and only 10% discrepancy is tolerated in parity bits. The proposed scheme is highly secure and efficient in terms of latency and data storage. As a proof of concept, two different machine learning algorithms, support vector machine and multigene genetic programming, are used to show that the scheme cannot be modeled by adversaries.
Lightweight physical unclonable functions (LPUFs) exploit manufacturing process variations of semiconductor integrated circuits (ICs) to protect IoT-based electronic and smart devices from new cyberattacks. This paper proposes two novel security techniques to enhance the robustness of LPUFs using configurable-based ring oscillator PUFs (CF-ROPUFs). These techniques are the intra-die frequency aware (IFA) approach to improve PUF reliability and the logarithmic gamma function ( $Ln_{\gamma }$ ) technique to enhance PUF randomness. The lightweight CF-ROPUF design is realized on hardware, and data samples are collected under varying temperatures and supply voltages over a population of 30 Spartan-3E FPGAs. Experimental results of the IFA technique in terms of average Hamming Weight (HM) demonstrate that the percentage of the reliable RO sample frequencies PUF output is 98.5%. For the analysis, PUF reliability is evaluated in terms of accuracy, repeatability, and reproducibility, the international organization for Standardization (ISO) standards. The results indicate that the RO samples are accurately measured from the CF-ROPUFs mapped in all the chips. After using the proposed 1-out-of-r coding algorithm, the results demonstrate high average repeatability of 98.2% and a magnified average reproducibility of 99.63%. It is also shown that our CF-ROPUF design is immune from accelerated aging impacts reliability issues. Statistical results show that Ln( $Ln_{\gamma }$ ) enhances the normality and mitigate the negative impacts of the systematic process variations on RO sample frequencies. Randomness results show that CF-ROPUF binary response bits can successfully pass the 15 NIST test suites for true randomness with an enhanced percentage, 93.3%, with the application of the 1-out-of-r coding.
The field programmable gate array (FPGA) supply chain is vulnerable to security issues from untrusted participants involved, resulting in the significant research being conducted in this domain. These challenges include, but are not limited to intellectual property theft, counterfeiting, bitstream tampering, etc. Such FPGAs are not reliable and can lead to significant security threats to users. To combat the distrust in the FPGA supply chain and secure its integrity, we propose a zero trust framework to secure the FPGA supply chain by leveraging blockchain and ring oscillator physical unclonable functions (ROPUFs). The policies for architecture draw inspiration from zero trust tenets outlined by the National Institute of Standards and Technology. To evaluate the successful implementation of the proposed technique, a case study of FPGA supply chain is presented. For this case study, a typical FPGA supply chain is considered which includes all the participants. This scenario is successfully simulated on the Ganache framework provided by Truffle suite. The experiment is performed on Artix 7 Xilinx FPGAs mounted on Nexys 4 Digilent boards. Finally, each tenet of the implemented zero trust architecture is evaluated by discussing an attack scenario, thus demonstrating the efficiency of the proposed zero trust framework to mitigate potential security risks. To the best of our knowledge, the work presented in this paper is the first to include zero trust principles for protecting the FPGA supply chain using blockchain and hardware oriented security.
Physical Unclonable Functions (PUFs) are used for authentication and generation of secure cryptographic keys. However, recent research work has shown that PUFs, in general, are vulnerable to machine learning modeling attacks. From a subset of Challenge-Response Pairs (CRPs), the remaining CRPs can be effectively predicted using different machine learning algorithms. In this work, Artificial Neural Networks (ANNs) using swarm intelligence-based modeling attacks are used against different silicon-based PUFs to test their resiliency to these attacks. Amongst the swarm intelligence algorithms, the Gravitational Search Algorithm (GSA), Cuckoo Search Algorithm (CS), Particle Swarm Optimizer (PSO) and the Grey Wolf Optimizer (GWO) are used. The attacks are extensively performed on six different types of PUFs; namely, Configurable Ring Oscillator, Inverter Ring Oscillator, XOR-Inverter Ring Oscillator, Arbiter, Modified XOR-Inverter Ring Oscillator, and Hybrid Delay Based PUF. From the results, it can be concluded that the first four PUFs under study are vulnerable to ANN swarm intelligence-based models, and their responses can be predicted with an average accuracy of 71.1% to 88.3 % for the different models. However, for the Hybrid Delay Based PUF and the Modified XOR-Inverter Ring Oscillator PUF, which are especially designed to thwart machine learning attacks, the prediction accuracy is much lower and in the range of 9.8 % to 14.5 %.
Silicon Physical Unclonable Functions (sPUFs) are one of the security primitives and state-of-the-art topics in hardware-oriented security and trust research. This paper presents an efficient and dynamic ring oscillator PUFs (d-ROPUFs) technique to improve sPUFs security against modeling attacks. In addition to enhancing the Entropy of weak ROPUF design, experimental results show that the proposed d-ROPUF technique allows the generation of larger and updated challenge-response pairs (CRP space) compared with simple ROPUF. Additionally, an innovative hardware-oriented security algorithm, namely, the Optimal Time Delay Algorithm (OTDA), is proposed. It is demonstrated that the OTDA algorithm significantly improves PUF reliability under varying operating conditions. Further, it is shown that the OTDA further efficiently enhances the d-ROPUF capability to generate a considerably large set of reliable secret keys to protect the PUF structure from new cyber-attacks, including machine learning and modeling attacks.
Silicon Physical Unclonable Functions (SPUFs) are used for extracting unique and random binary signatures from semiconductor chips. This paper introduces the first asynchronous LFSR based PUF; namely, the LFSR-PUF, that can be specifically implemented on a FPGA. The proposed ALFSR-PUF is derived from an asynchronous LFSR that uses D-latch/flip-flops and LUTs on an FPGA to generate a random clock by exploiting the inherent manufacturing process variations of the design primitives. The LFSR-PUF efficiently uses FPGA basic building blocks in each CLB to optimally occupy a very small area (two slices) in the FPGA. Besides design efficiency and area optimization, the challenge-response space is also increased. Consequently, an instantiation of more than one PUF instance is possible within each CLB. Experimental results on Xilinx Spartan-3E FPGA (90 nm) show that the generated signatures are both unique and reliable under varying temperature and voltage conditions.