Maestro:面向可穿戴超声边缘计算的302 GFLOPS/W与19.8GFLOPS RISC-V向量-张量架构
Maestro: A 302 GFLOPS/W and 19.8GFLOPS RISC-V Vector-Tensor Architecture for Wearable Ultrasound Edge Computing
AI总结:
提出面向可穿戴超声边缘计算的RISC-V SoC Maestro,通过统一向量-张量单元和FFT加速器实现高能效信号处理与CNN推理,较同类SoC获5倍加速且功耗仅12mW。
摘要:
大多数可穿戴超声(WUS)设备缺乏在边缘处理信号的计算能力,转而依赖远程卸载,这带来了延迟、高功耗和隐私问题。我们提出了Maestro,一款具备统一向量-张量单元(VTU)和存储器耦合快速傅里叶变换(FFT)加速器的RISC-V SoC,专为可穿戴超声设备的边缘处理而设计,采用低成本的TSMC 65nm CMOS工艺制造。VTU在FP16下达到峰值302GFLOPS/W和19.8GFLOPS,而多精度16/32位浮点FFT加速器在FP16下提供峰值60.6GFLOPS/W和3.6GFLOPS。我们在基于超声的手势识别任务上评估了Maestro,在信号处理中实现1.62GFLOPS,能效为26.68GFLOPS/W;在卷积神经网络(CNN)工作负载中实现19.52GFLOPS,能效为298.03GFLOPS/W。与具有相似任务配置的最先进SoC相比,Maestro实现了5倍加速,功耗仅为12mW,在可穿戴超声通道预处理和基于机器学习的后处理流水线中能耗为2.5mJ。
英文摘要:
Most Wearable Ultrasound (WUS) devices lack the computational power to process signals at the edge, instead relying on remote offload, which introduces latency, high power consumption, and privacy concerns. We present Maestro, a RISC-V SoC with unified Vector-Tensor Unit (VTU) and memory-coupled Fast Fourier Transform (FFT) accelerators targeting edge processing for wearable ultrasound devices, fabricated using low-cost TSMC 65nm CMOS technology. The VTU achieves peak 302GFLOPS/W and 19.8GFLOPS at FP16, while the multi-precision 16/32-bit floating-point FFT accelerator delivers peak 60.6GFLOPS/W and 3.6GFLOPS at FP16, We evaluate Maestro on a US-based gesture recognition task, achieving 1.62GFLOPS in signal processing at 26.68GFLOPS/W, and 19.52GFLOPS in Convolutional Neural Network (CNN) workloads at 298.03GFLOPS/W. Compared to a state-of-the-art SoC with a similar mission profile, Maestro achieves a 5x speedup while consuming only 12mW, with an energy consumption of 2.5mJ in a wearable US channel preprocessing and ML-based postprocessing pipeline.
机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。