Search

Reducing learning energy consumption is critical to edge-artificial intelligence (AI) processors with on-chip learning since on-chip learning energy dominates energy consumption, especially for applications that require long-term learning. To achieve this goal, we optimize a neuromorphic learning algorithm and propose random target window (TW) selection, hierarchical update skip (HUS), and asynchronous time step acceleration (ATSA) to reduce the on-chip learning power consumption. Our approach results in a 28-nm 1.25-mm2 asynchronous neuromorphic processor (ANP-I) with on-chip learning energy per sample less than 15% of inference energy per sample. With all weights randomly initialized, this processor enables on-chip learning for edge-AI tasks such as gesture recognition, keyword spotting, and image classification, consuming sub- $0.1 ~\mu \text{J}$ of learning energy per sample at 0.56 V and 40-MHz frequency while maintaining >92% accuracy for all tasks.

This paper presents a chopper-stabilized three-stage operational amplifier (OpAmp) with a unity gain bandwidth of 69 MHz and an input referred noise density of 3 nV $/\surd {Hz}$ . The proposed design achieves a stable unity gain by proposing a new pole and zero scheme with very low power consumption, drawing only 3.3 mA from a 1.8 V power supply while driving a load capacitor as large as 100pF. To achieve rail-to-rail input swing, the design uses both NMOS and PMOS differential pairs at the input and biases them in the subthreshold region to provide an identical net trans-conductance over the rail-to-rail input common mode. Furthermore, an adaptive biasing is employed and the current sources are kept ON during large signal transitions at the input, thus eliminating crossover distortion and providing a high slew rate of 40 V/ $\mu \text{s}$ at a 100 pF load capacitor. The design employs chopping at 2.5 MHz and is enhanced with a local ripple reduction loop, making the OpAmp suitable for high gain and wide bandwidth applications with less filtering required. The design also reduces the input bias current significantly from 500 nA to 1.5 nA by buffering the input and applying it to the modified bootstrap switches. The proposed OpAmp, fabricated in a $0.18~\mu \text{m}$ CMOS process, exhibits a maximum offset of $4.5~\mu \text{V}$ , a flicker noise corner frequency of 246 Hz, a DC gain of 146 dB, a power supply rejection ratio of 123 dB, and a common mode rejection ratio of 116 dB.

Nonvolatile near-memory-compute (nvNMC) macros are promising candidates for edge artificial intelligence (AI) devices requiring high energy efficiency, short wakeup-to-compute latency, and robust inference accuracy with high precision of inputs (IN), weights ( $W$ ), and outputs (OUT). Nonetheless, the practical application of nvNMC macros is hindered by inherent design challenges: 1) high energy consumption in reading repetitious weight data, 2) low energy efficiency due to high bitstream toggling rate in digital multiply-and-accumulate (MAC) circuits, 3) narrow signal margin and high memory readout latency, and 4) redundant MAC computation in digital MAC circuits under input-stationary flow. In addressing these challenges, we developed a number of schemes based on the concept of system-circuit co-design, including 1) a weight-feature aware read (WFAR) scheme; 2) a toggling-aware weight-tuning (TAWT) scheme; 3) a differential charge-accumulating margin-enhanced voltage-sensing amplifier (DCME-VSA); and 4) an input-sparsity aware pre-computing unit (ISAPU). An 8-Mb spin-transfer torque magnetic random access memory (STT-MRAM) nvNMC macro fabricated using foundry-provided 22 nm STT-MRAM achieved read bandwidth of 436 GB/s, MAC computing latency of 20 ns, and energy efficiency of 53.6–190.2 TOPS/W when performing eight-bit input and eight-bit weight MAC operations with 26-bit output and 576 accumulations.

On-chip non-volatile compute-in-memory (nvCIM) enables artificial intelligence (AI)-edge processors to perform multiply-and-accumulate (MAC) operations while enabling the non-volatile storage of weight data in power-off mode to enhance energy efficiency. However, the design challenges of nvCIM-based AI-edge processors include: 1) lack of a nvCIM-friendly computing flow; 2) a tradeoff between usage of memory devices versus process variations, computing yield and area overhead; 3) long computing latency and low energy efficiency; and 4) small-signal margin and large bitline current. This article presents an nvCIM-friendly AI-edge processor that uses a hybrid-mode resistive random access memory nvCIM (hmRe-nvCIM) macro to overcome the abovementioned challenges by three processor-level schemes: 1) a multimode nvCIM engine controller (mmCIM-EC); 2) a bitwise-input-sparsity and place-value-aware dynamic accumulation (BIS-PVA-DA); and 3) a bitwise weight column inversion (BWCI) and two macro-level schemes: 1) a dynamic-accumulation-aware current quantization (DACQ) and 2) a current–voltage-hybrid analog-to-digital converter (CVH-ADC). The proposed AI-edge processor fabricated using 22-nm technology achieved 51.4 TOPS/W and 472.7- $\mu \text{s}$ wake-up to response time, while the hmRe-nvCIM macro achieved 67.2 TOPS/W under 8-bit input, 8-bit weight, and 22- or 24-bit output precision.

Advanced artificial intelligence edge devices are expected to support floating-point (FP) multiply and accumulation operations while ensuring high energy efficiency and high inference accuracy. This work presents an FP compute-in-memory (CIM) macro that exploits the advantages of computing in the time, digital, and analog-voltage domain for high energy efficiency and accuracy. This work employs: 1) a hybrid-domain macrostructure to enable the computation of both the exponent and mantissa within the same CIM macro; 2) a time-domain computing scheme for energy-efficient exponent computation; 3) a product-exponent-based input-mantissa alignment scheme to enable the accumulation of the product mantissa in the same column; and 4) a place-value-dependent digital–analog-hybrid computing scheme to enable energy-efficient mantissa computations of sufficient accuracy. A 22-nm 832-kB FP-CIM macro fabricated using foundry-provided compact 6T-static random access memory (SRAM) cells achieved a high energy efficiency of 72.14 tera-floating-point operations per second (TFLOPS)/W while performing FP-multiply-and-accumulate (MAC) operations involving BF16-input, BF16-weight, FP32-output, and 128 accumulations.

This article presents a novel static random access memory computing-in-memory (SRAM-CIM) structure designed for high-precision multiply-and-accumulate (MAC) operations with high energy efficiency (EF), high readout accuracy, and short compute latency. The proposed device employs 1) a time-domain incremental-accumulation (TDIA) scheme to enable high-accumulation MAC operations while maintaining a large signal margin across MAC values (MACVs), 2) a dynamic differential-reference (D2REF) scheme based on software-hardware co-design to reduce read energy consumption, and 3) a low-dMACV-aware recursive time-to-digital converter (LMAR-TDC) for implementation with the D2REF scheme to further suppress readout energy consumption. A 28 nm 1 Mb SRAM-CIM macro fabricated using foundry-provided compact 6T-SRAM cells achieved EF of 39.31 TOPS/W and compute latency of 6.6 ns for 8b-MAC operations with 64 accumulations per cycle and near-full output precision (22b).

Convolutional neural networks (CNNs) play a key role in many deep learning applications; however, these networks are resource intensive. The parallel computing ability of computing-in-memory (CIM) enables high energy efficiency in artificial intelligence accelerators. When implementing a CNN in CIM, quantization and pruning are indispensable for reducing the calculation complexity and improving the efficiency of hardware calculations. Mixed-precision quantization with flexible bit widths provides a better efficiency-accuracy tradeoff than fixed-precision quantization. However, CIM calculations for mixed-precision models are inefficient because the fixed capacity of CIM macros is redundant for hybrid precision distributions. To address this, we propose a software and hardware co-design static random-access memory (SRAM)-based CIM architecture called SUN, including a CIM-adaptive mixed precision joint pruning quantization algorithm and dynamic hybrid precision CNN accelerator. Three techniques are implemented in this architecture: 1) a mixed precision joint pruning algorithm for reducing the memory access and removing the redundant computing; 2) a CIM-adaptive filter-wise and paired mixed-precision quantization for improving CIM macro utilization; and 3) an SRAM-based CIM CNN accelerator in which the SRAM CIM macro is used as the processing element to support sparse and mixed-precision CNN computation with high CIM macro utilization. This architecture achieves a system area efficiency of 428.2 TOPS/mm 2 and throughput of 792.2 GOPS on the CIFAR-10 dataset.

The tutorial explores key security and functional safety challenges for Artificial Intelligence (AI) in embedded automotive systems, including aspects from adversarial attacks, long life cycles of products, and limited energy resources of automotive platforms within safety-critical environments in diverse use cases. It provides a set of recommendations for how the security and safety engineering of machine learning can address these challenges. It also provides an overview of contemporary security and functional safety engineering practices, encompassing up-to-date legislative and technical prerequisites. Finally, we identify the role of AI edge processing in enhancing security and functional safety within embedded automotive systems.