Despite the rising prevalence of deep neural networks (DNNs) in cyber-physical systems, their vulnerability to adversarial bit-flip attacks (BFAs) is a noteworthy concern. This paper proposes B3FA, a semi-black-box BFA-based parameter attack on DNNs, assuming the adversary has limited knowledge about the model. We consider practical scenarios often feature a more restricted threat model for real-world systems, contrasting with the typical BFA models that presuppose the adversary's full access to a network's inputs and parameters. The introduced bit-flip approach utilizes a magnitude-based ranking method and a statistical reconstruction technique to identify the vulnerable bits. We demonstrate the effectiveness of B3FA on several DNN models in a semi-black-box setting. For example, B3FA could drop the accuracy of a MobileNetV2 from 69.84% to 9% with only 20 bit-flips in a real-world setting.

Low-precision weights and activations in deep neural networks (DNNs) outperform their full-precision counterparts in terms of hardware efficiency. When implemented with low-precision operations, specifically in the extreme case where network parameters are binarized (i.e. BNNs), the two most frequently mentioned benefits of quantization are reduced memory consumption and a faster inference process. In this paper, we introduce a third advantage of very low-precision neural networks: improved fault-tolerance attribute. We investigate the impact of memory faults on state-of-the-art binary neural networks (BNNs) through comprehensive analysis. Despite the inclusion of floating-point parameters in BNN architectures to improve accuracy, our findings reveal that BNNs are highly sensitive to deviations in these parameters caused by memory faults. In light of this crucial finding, we propose a technique to improve BNN dependability by restricting the range of float parameters through a novel deliberately uniform quantization. The introduced quantization technique results in a reduction in the proportion of floating-point parameters utilized in the BNN, without incurring any additional computational overheads during the inference stage. The extensive experimental fault simulation on the proposed BNN architecture (i.e. ZOBNN) reveal a remarkable 5X enhancement in robustness compared to conventional floating-point DNN. Notably, this improvement is achieved without incurring any computational overhead. Crucially, this enhancement comes without computational overhead. ZOBNN excels in critical edge applications characterized by limited computational resources, prioritizing both dependability and real-time performance.

Deep neural networks (DNNs) are increasingly being deployed in safety-critical systems such as personal healthcare devices and self-driving cars. In such DNN-based systems, error resilience is a top priority since faults in DNN inference could lead to mispredictions and safety hazards. For latency-critical DNN inference on resource-constrained edge devices, it is nontrivial to apply conventional redundancy-based fault tolerance techniques. In this paper, we propose FitAct, a low-cost approach to enhance the error resilience of DNNs by deploying fine-grained post-trainable activation functions. The main idea is to precisely bound the activation value of each individual neuron via neuron-wise bounded activation functions, so that it could prevent the fault propagation in the network. To avoid complex DNN model re-training, we propose to decouple the accuracy training and resilience training, and develop a lightweight post-training phase to learn these activation functions with precise bound values. Experimental results on widely used DNN models such as AlexNet, VGG16, and ResNet50 demonstrate that FitAct outperform state-of-the-art studies such as Clip-Act and Ranger in enhancing the DNN error resilience for a wide range of fault rates, while adding manageable runtime and memory space overheads.

The most advanced ASIC-based approximate adders are focused on gate or transistor level approximating structures. However, due to architectural differences between ASIC and FPGA, comparable performance gains for FPGA-based approximate adders cannot be obtained using ASIC-based approximation ones. In this paper, we propose a method for designing a low-error approximate adder that effectively deploys the modern FPGA structure. We introduce an FPGA-based approximate adder, named as Majority Approximate Adder (MAA), with less error than the advanced approximate adders. MAA is constructed using an approximate part and an accurate one; i.e. the accurate part is based on a smaller carry-chain compared with the carry-chain of the corresponding accurate adder. In addition, approximate part is designed to use FPGA resources efficiently with a low mean error distance (MED). Experimental results based on Monte-Carlo simulation demonstrates that a 16-bit MAA has a 49.92% lower MED than the state of the art FPGA-based approximate adder. MAA also takes up less area and consumes less power than other FPGA-based approximate adders in the literature.

Because of their high accuracy, deep neural net-works (DNNs) have achieved amazing success in security-critical systems such as medical devices. It has recently been demon-strated that Adversarial Bit Flip Attacks (BFAs) against DNN hardware by flipping a very small number of bits can result in catastrophic accuracy loss. The reliance on test data, however, is a significant drawback of previous state-of-the-art bit-flip attack methods. This is frequently not possible with applications containing sensitive or proprietary data. In this paper, we propose Blind Data Adversarial Bit-flip Attack (BDFA), a novel technique to enable BFA against DNN hardware without any access to the training or testing data. This is achieved by optimizing for a synthetic dataset, which is engineered to match the statistics of batch normalization across different layers of the network and the targeted label. Experimental results show that BDFA could decrease the accuracy of ResNet50 significantly from 75.96% to 13.94% with only 4 bits flips.

Transistor aging raises a vital lifetime reliability challenge for FPGA devices in advanced technology nodes. In this paper, we design a tool called LEAP to enable the aging-aware FPGA architecture exploration. The core idea of LEAP is to efficiently model the aging-induced delay degradation at the coarse-grained FPGA basic block level using deep neural networks (DNNs), while achieving almost the same accuracy as the transistor-level simulation. For each type of the FPGA basic block such as LUT and DSP, we first characterize its accurate delay degradation via transistor-level SPICE simulation under a versatile set of aging factors from the FPGA fabric and in-field operation. Then we train one DNN model for each block type to learn the relation between its delay degradation and aging factors. Moreover, we integrate our DNN models into the widely used Verilog-to-Routing (VTR 8) toolflow and generate the aging-aware FPGA architecture file. Experimental results demonstrate that our proposed flow can predict the delay degradation of FPGA blocks more than 104x to 107x faster than transistor-level SPICE simulation, with the maximum prediction error of less than 0.7%. Therefore, FPGA architects can leverage LEAP to explore better aging-aware FPGA architectures.

In this paper, we develop a framework called MAPLE to enable the aging-aware FPGA architecture exploration. The core idea is to efficiently model the aging-induced delay degradation at the coarse-grained FPGA basic block level using deep neural networks (DNNs). For each type of the FPGA basic block such as LUT and DSP, we first characterize its accurate delay degradation via transistor-level SPICE simulation under a versatile set of aging factors from the FPGA fabric and in-field operation. Then we train one DNN model for each block type to quickly and accurately predict the complex relation between its delay degradation and comprehensive aging factors. Moreover, we integrate our DNN models into the widely used Verilog-to-Routing toolflow (VTR 8) to support analyzing the impact of aging-induced delay degradation on the entire large-scale FPGA architecture. Experimental results demonstrate that MAPLE can predict the delay degradation of FPGA blocks 104to 107times faster than transistor-level SPICE simulation, with a prediction error less than 0.7%. Our case study demonstrates that FPGA architects can effectively leverage MAPLE to explore better aging-aware FPGA architectures.

With the CMOS technology scaling, transistor aging has become one major issue affecting circuit reliability and lifetime. There are two major classes of existing studies that model the aging effects in the circuit delay. One is at transistor-level, which is highly accurate but very slow. The other is at gate-level, which is faster but less accurate. Moreover, most prior studies only consider a limited subset or limited value ranges of aging factors.

Traditionally, increasing logical masking probability has been used to improve the circuit reliability against single-event transients (SETs). As the very first work, this paper presents a new approach to increase the reliability of digital circuits against soft errors caused by multiple event transients (METs) by taking advantages of circuit partitioning and local logical restructuring techniques. In the proposed approach, the circuit is partitioned into various subcircuits and, then, several structures of each subcircuits which satisfy the area constraints are extracted by using a graph-based procedure. In order to select the suitable alternative between various subcircuit structures, we introduce a novel metric named global failure probability in the presence of METs (GFPM). This parameter provides an evaluation of each subcircuits contribution in the soft error rate (SER) of the given circuit making it possible to estimate the impacts of changing the structure of the subcircuits on the circuit SER. Hence, it prevents from repeatedly calculating the SER of the circuit that is very time-consuming leading to significant improvements in the optimization runtime. Experimental studies on ISCAS benchmark circuits show that the proposed approach, on average, achieves 18.4% SER reduction with 11.9% area overhead and 8.2% delay overhead comparing to the original circuit while the global SET-based SER mitigation approach and the global MET-based SER mitigation approach achieve 8.46% and 21.8% SER reduction, respectively. Besides, the proposed technique is about $580 \times $ faster than the global MET-based method.

In this paper, a novel image encryption algorithm is proposed based on the combination of the chaos sequence and the modified AES algorithm. In this method, the encryption key is generated by Arnold chaos sequence. Then, the original image is encrypted using the modified AES algorithm and by implementing the round keys produced by the chaos system. The proposed approach not only reduces the time complexity of the algorithm but also adds the diffusion ability to the proposed algorithm, which make the encrypted images by the proposed algorithm resistant to the differential attacks. The key space of the proposed method is large enough to resist the brute-force attacks. This method is so sensitive to the initial values and input image so that the small changes in these values can lead to significant changes in the encrypted image. Using statistical analyses, we show that this approach can protect the image against the statistical attacks. The entropy test results illustrate that the entropy values are close to the ideal, and hence, the proposed algorithm is secure against the entropy attacks. The simulation results clarify that the small changes in the original image and key result in the significant changes in the encrypted image and the original image cannot be accessed.

Current technology scaling trends aggressively increases the susceptibility of combinational circuit reliability to radiation-induced transient faults (which also known as soft errors). Various gate sizing techniques have been used to reduce soft error rate (SER) in the past, but their main drawback is that they are expensive in term of run time. These methods require changes to adapt to the large scale circuits. In this paper, an efficient circuit partitioning-based gate sizing method is presented, which significantly speeds up the gate sizing optimization process. In the proposed method, the circuit is divided into the topologically levelized small subcircuits by cone structures. Then, the subcircuits which are located in the same level are resized individually and independently. The subcircuit error probability (SEP) metric is introduced to evaluate the contribution of each subcircuit into the total circuit SER. The key idea of the proposed method is to evaluate the effects of each gate sizing on circuit reliability locally using SEP instead of global evaluation by the total circuit SER. Such evaluation results in speeding up the gate sizing optimization process. Experimental results show that the proposed approach is about 280× orders of magnitude faster than the sensitivity-based gate sizing approach [R. R. Rao, D. Blaauw, and D. Sylvester, “Soft error reduction in combinational logic using gate resizing and flipflop selection,” in Proc. IEEE/ACM Int. Conf. Comput.-Aided Des., 2006, pp. 502-509] while it can achieve up to 45% reduction in circuit SER with less than 17% area overhead. This level of speed and efficiency makes the proposed approach a viable solution to mitigate the SER of very large combinational circuits used in industry.

Soft errors in combinational logic circuits are emerging as a significant reliability concern for nanoscale VLSI designs. This paper presents a novel sensitivity-based gate sizing methodology to reduce the soft error rate (SER) of combinational circuits in the presence of process variations. The proposed method is based on modeling the statistics of SER of the circuit gates as a random variable to formulate a statistical optimization problem. A backward traversing algorithm with capability for incremental analysis is developed for computing the distribution of circuit gates of SER random variables. We present a gate resizing algorithm in which the gates with the most contribution to the circuit SER are selected in a candidate set using a statistical ordering approach. The proposed algorithm trades off SER reduction and area overheads. The experimental results show that using the proposed methodology, the circuit statistical SER can be reduced by up to 56.4% compared with the 14.8% SER reduction of a circuit obtained using the worst case methodology at the expense of 10% area overhead under 10% process variation ratio. The results also show that the proposed method achieves about 40% more SER reduction compared with that obtained using closed-form analysis for statistical soft error rate estimation (CASSER), the most recently published similar work, in the same experimental conditions. Comparing the runtime of the proposed optimization algorithm with the optimization based on CASSER, it is observed that the proposed method is two orders of magnitude faster than CASSER due to its incremental analysis property.

This paper presents a novel circuit optimization technique to reduce soft error rates (SER) of combinational logic circuits in the presence of process variations. We take advantage of gate sizing technique which has been shown to be one of the most effective methods for SER mitigation in digital circuits. A statistical SER (SSER) estimation approach is proposed to be used to prune the circuit graph into a smaller set of candidate gates. Then, we perform incremental statistical sensitivity computations to determine the resizing step that are the largest improvement to circuit SER. The proposed algorithm trades off SER reduction and area overhead. Experimental results on a variety of benchmarks show SER reductions of 67.3% with gate sizing approach, with 5.5% area overheads and delay improvement of 3.2%, on average. The runtimes for the optimization algorithms are on the order of 10 minutes.

Soft error rate (SER) estimation is becoming more and more important since nanometer digital integrated circuits are getting increasingly vulnerable to soft errors. In this paper, a novel approach is proposed for soft error rate analysis of digital combinational circuit considering all masking factors. We introduce a concept called Probabilistic Vulnerability Window (PVW) which is an inference of the necessary conditions for a Single Event Transient (SET) to cause observable errors in the circuit. A computation model is proposed to calculate PVW's for all circuit gate outputs. Using the computation model, the proposed method estimates the soft error rate of the circuit by computing the probabilistic vulnerability windows in a backward approach. Experimental results show that the proposed method increases the SER computation speed by 1000X, with less than 10% accuracy loss when compared to the Monte-Carlo based fault injection methods. The results also show than the proposed approach keeps its efficiency when it is applied for estimating the soft error rate considering various SET's with different initial widths while the runtime of traditional SER estimation methods increases rapidly in such cases.

Carbon Nanotube Field Effect Transistors (CNFETs) show great promise as extensions to silicon CMOS. However, CNFET-based circuits will face great fabrication challenges that will translate into important parameter variations and decreased reliability. Hence, asynchronous logic, which is intrinsically more robust to variability, seems an ideal and perhaps unavoidable choice for digital circuits in CNFET technology. This article presents the results on the design and analysis of a CNFET-based implementation of an asynchronous circuit primitive: the Muller C-element. Using a CNFET SPICE model, we evaluate the robustness of CNFET-based C-element in the presence of CNT fabrication-related nonidealities. We investigate a quantitative evaluation of how timing variability impacts the functionality of a C-element and then, extract the necessary delay constraints of the C-element circuit from the signal transition graph specification. Considering the large degrees of spatial correlation observed between the CNFETs fabricated on directionally grown CNTs, a layout technique is exploited to overcome the robustness challenges of a CNFET-based C-element. Extensive Monte Carlo simulations on the proposed technique have demonstrated the effectiveness of the proposed CNFET-based C-element by improving approximately 50X in its robustness in expense of 65% area, 47% delay, and 56% power consumption overheads. Experimental results indicate that implementation of some CNFET-based Quasi Delay Insensitive (QDI) benchmark circuits using the proposed C-element results in significant robustness improvement with negligible power and throughput overheads. As a promising step toward CNFET-based giga-scale integrated circuits, this article shows that the asynchronous logic is an effective approach to design robust integrated circuits in CNFET technology with inherent extreme physical variations.

Carbon nanotube field effect transistors (CNFETs) show great promise as extensions to silicon CMOS. However, imperfections, which are mainly related to carbon nanotubes (CNTs) growth process, result in metallic and nonuniform CNTs leading to significant functional yield reduction. This paper presents a comprehensive technique for statistical functional yield estimation and enhancement of CNFET-based VLSI circuits. Based on experimental data extracted from aligned CNTs, we propose a compact statistical model to estimate the failure probability of a CNFET. Using the proposed failure model, we show that enhancing the CNT synthesis process alone cannot achieve acceptable functional yield for upcoming CNFET-based VLSI circuits. We propose a technique which is based on replacing each transistor by series-parallel transistor structures to reduce the failure probability of CNFETs in the presence of metallic and nonuniform CNTs. The technique is adapted to use single directional independence, which is inherent in aligned CNTs, to enhance the functional yield as validated by theoretical analysis and simulation results. Tradeoffs between failure probability reduction and design overheads such as area and current drive are explored. As demonstrated by extensive simulation results, the proposed technique achieves 80% functional yield in CNFET technology at the cost of 7.5X area and 34% current drive overheads if the CNT density and the fraction of semiconducting CNTs are improved to 200 CNTs per μm and 99.99%, respectively.

Minimizing power consumption is one of the most important objectives in VLSI design. This paper introduces a methodology for the optimization of total power consumption of template based asynchronous circuits via dual Vddassignment, dual Vthassignment and template sizing while maintaining performance requirements. The utilized circuit model is a Timed Petri-net which captures the dynamic behavior of the circuit. These three power reduction techniques are properly encoded in a quantum genetic algorithm and evaluated simultaneously. Experimental results are given for a number of 65 nm related benchmark circuits and show that this method reduces the total power by close to an order of magnitude, with no or negligible performance penalty. From the experimental results, we show that the combination of asynchronous design and three low power techniques is an effective way to achieve low power and high performance circuits.

This paper introduces a framework for the minimization of leakage power consumption of asynchronous circuits via using dual threshold voltages technique. The utilized circuit model is an extended Timed Petri-Net which captures the dynamic behavior of the circuit. We propose a heuristic method based on quantum genetic algorithm which finds the optimal high and low threshold voltage assignment. Experimental results are given for a number of 90 nm ISCAS benchmark circuits. From the experimental results, we show that the combination of asynchronous and multiple threshold voltage design techniques is an effective way to achieve low leakage power budget in high performance asynchronous circuits.

Replacing functional units of an extensible processor with reconfigurable fnctional units enhances performance and flexibility ofprocessors to execute custom instructions. That is due to the ability ofreconfigurable fnctional units to perform computations in hardware to increase performance, while retaining much of the flexibility of a software solution. In this paper, we develop a heterogeneous architecture for the reconfigurable fnctional unit of an extensible processor. To verify the efficiency of our architecture, we applied it to 8 applications of Mibench. Our experiments show that compared to the similar architectures, ours supports a wide range of custom instructions. In addition, use of the new architecture improves execution time of custom instructions by 20% to 30% on average. Moreover, compared with the previous architecture, area is reduced by 15%.