The secure hash algorithm 2 (SHA-2) family, including the SHA-224/256/384/512 hash functions, is widely adopted in many modern domains, ranging from Internet of Things devices to cryptocurrency. SHA-2 functions are often implemented on hardware to optimize performance and power. In addition to the high-performance and low-cost requirements, the hardware for SHA-2 must be highly flexible for many applications. This paper proposes an SHA-2 hardware architecture named the multimode SHA-2 accelerator (MSA), which has high performance and flexibility at the system-on-chip level. To achieve high performance and flexibility, our accelerator applies three optimal techniques. First, a multimode processing element architecture is proposed to enable the accelerator to compute various SHA-2 functions for many applications. Second, a three-stage arithmetic logic unit pipeline architecture is proposed to reduce the critical paths and hardware resources. Finally, nonce generator and nonce validator architectures are proposed to reduce memory access and maximize the performance of the proposed MSA for blockchain mining applications. The MSA accuracy is tested on a real hardware platform (the Xilinx Alveo U280 FPGA). The experimental results on the field programmable gate array (FPGA) prove that the proposed MSA achieves significantly better performance, hardware efficiency, and flexibility than previous works. The evaluation results for energy efficiency show that the proposed MSA achieves up to 38.05 Mhps/W, which is 543.6 and 29 times better than the state-of-the-art Intel i9-10940X CPU and RTX 3090 GPU, respectively.
The Advanced Encryption Standard (AES) is a fundamental cryptographic algorithm widely used to secure data in embedded systems, IoT devices, and cloud computing platforms. However, recent research on AES hardware accelerators face challenges in achieving high performance and hardware efficiency, particularly when supporting multiple modes and key sizes. To address these limitations, this paper proposes a hardware-efficient RISC-V accelerator with low-latency AES instruction extension (AES-RV), designed to enhance both processing speed and energy efficiency across various AES configurations. Specifically, AES-RV incorporates three key optimizations: high-bandwidth internal buffers for continuous data processing, a specialized AES unit with low-latency custom instructions, and system pipelining with a ping-pong memory transfer mechanism. The AES-RV accelerator is implemented and evaluated on a real-time Xilinx ZCU102 FPGA system-on-chip (SoC), utilizing 29,608 FFs, 32,483 LUTs, and 12 BRAMs. Performance comparisons against a baseline RISC-V implementation for multiple AES modes and key sizes demonstrate latency improvements ranging from 195.5 times to 255.97 times. Additionally, evaluations against powerful CPUs and GPUs in real-time AES executions reveal energy efficiency gains of 9.92 times to 453.04 times. Compared to state-of-the-art AES hardware accelerators, AES-RV achieves throughput improvements of 13.56 times to 33.52 times, energy efficiency enhancements of 2.36 times to 58.76 times, and area efficiency gains of 91.42 times to 638.8 times.
3 April 2025 in IEEE Transactions on Circuits and Systems I: Regular Papers
Classification is a crucial aspect of cardiovascular-related challenges, requiring thorough research and optimization to develop effective solutions for both patients and doctors. Recently, rapid advancements in artificial intelligence, particularly Convolutional Neural Networks (CNNs), have introduced numerous effective methods, significantly improving disease classification in Electrocardiogram (ECG) analysis. However, existing CNN-based accelerators often encounter challenges such as high parameter counts, limited flexibility in handling diverse CNN configurations, and inefficient hardware utilization. To address these issues, this paper proposes the Mini InceptionNet Accelerator (MINA), a hardware-efficient and flexible accelerator designed specifically for one-dimensional (1-D) CNN-based ECG classification. First, a novel 1-D CNN model, Mini InceptionNet, reduces the parameter count by 41.6% compared to the smallest existing 1-D CNN, minimizing memory requirements while maintaining high classification accuracy. Second, a flexible Processing Element Array (PEA) is designed with a Sharing Buffer Allocator (SBA) to support dynamic data coordination across various network topology parameters. Third, each Processing Element (PE) is equipped with four Local Data Memories (LDMs) and an ALU, enabling efficient intermediate data storage and versatile operations for modern CNN models. To demonstrate its effectiveness, MINA has been successfully implemented and verified on the ZCU102 FPGA at the system-on-chip level. FPGA evaluations show that MINA achieves $1.3\times - 2.9\times $ higher energy efficiency (GOP/s/MeLUT) than state-of-the-art 2-D CNN accelerators. Compared to existing 1-D CNN accelerators, MINA achieves at least $1.53\times $ improvement in the area-delay product (ADP). Additionally, weight pruning is discussed as a supporting strategy, achieving up to $3\times $ faster inference time and a $2.13\times $ improvement in ADP at 70% sparsity.
With rapid advances in artificial intelligence applications, computational demands have grown in size and complexity, necessitating flexible and highly efficient hardware architectures. Although traditional hardware solutions have demonstrated high performance and efficiency for specific applications, they still face challenges in terms of flexibility and energy efficiency. To overcome these shortcomings, we propose a new architecture based on coarse-grained linear array (CGLA), called "IMAX." IMAX adopts a linear structure in which cache memory and processing units are alternately arranged, absorbing irregular memory access latencies and achieving superior performance and energy efficiency. IMAX3 introduces further optimizations in communication performance, double buffering techniques, and improved sparse matrix multiplication (SpGEMM), achieving even higher performance. Real-time evaluation results on the Xilinx VPK180 SoC demonstrate that IMAX3 achieves significant improvement over conventional graphics processing units (GPUs) and other architectures. In executing Fast Fourier Transforms (FFT), IMAX3 achieves up to 10 times the energy efficiency of Jetson AGX Orin. For SpGEMM, IMAX3 delivers up to 503 times the speed of GTX 1080Ti and up to 162 times that of Jetson AGX Orin. In comparison with related architecture, IMAX3 also exhibits significantly superior performance in calculations such as matrix multiplication and FFT. In matrix multiplication, IMAX3 achieved approximately 23 times faster execution than STRELA and 61 times faster than RipTide, with substantial improvements in energy efficiency. These results confirm IMAX3 as a flexible and efficient hardware platform, outperforming traditional GPUs and other architectures in both computational performance and energy efficiency, particularly in real-time processing and low-power environments.
This article proposes a flexible and scalable BLAKE/BLAKE2 coprocessor aiming for high flexibility, high performance, and low power for blockchain-based IoT applications.
1 August 2024 in IEEE Transactions on Circuits and Systems I: Regular Papers
Emerging modern internet-of-things (IoT) systems require hardware development to support multiple 8/32/64-bit lightweight cryptographic (LWC) algorithms with high speed and energy efficiency to ensure diverse security requirements. Accordingly, a coarse-grained reconfigurable array (CGRA) is considered the most effective architecture for achieving high speed, low power, and high flexibility for implementing LWC algorithms. However, existing CGRA designs for cryptography focus only on improvements to outdated 8/32-bit algorithms, suffer from large area requirements, and have long compilation times. To address these issues, this paper proposes a new CGRA-based accelerator named LiCryptor to support various 8/32/64-bit LWC algorithms with high speed and small area. Three innovative ideas are proposed to enable LiCryptor to achieve these goals: a compact multi-grained processing element array (M-PEA), a shared 8/32/64-bit arithmetic logic unit (ALU), and an assembly-like inline directive (AID) mapping method. The LiCryptor has been successfully implemented and verified on the Xilinx ZCU102 FPGA. Real-time performance evaluation across various LWC algorithms on FPGA shows that LiCryptor is 1.33 to 4 times better in execution time and 3.4 to 153 times better in power-delay products (PDP) compared to today’s most powerful CPUs. Notably, evaluation of AID mapping on the ARM Cortex-A53 CPU of the ZCU102 FPGA shows that its compilation time is less than 1.5 ms for most LWC algorithms, at least 2,333 times faster than CFG mapping in current CGRAs. Moreover, experimental results on 45nm ASIC technology show that the LiCryptor significantly outperforms existing CGRAs and other reconfigurable designs in terms of throughput and area efficiency.