arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2409.02689physics.app-phcs.ET
收藏

基于单个片上非线性声波器件的频域并行计算

Frequency-domain Parallel Computing Using Single On-Chip Nonlinear Acoustic-wave Device

Jun Ji, Zichen Xi, Bernadeta R. Srijanto, Ivan I. Kravchenko, Ming Jin, Wenjie Xiong, Linbo Shao

AI总结:

本研究提出利用GHz声波非线性的频域并行计算方案,单个铌酸锂片上器件可同时实现十亿次运算,算力密度与能效优异,可支持矩阵乘法、卷积成像等任务,适用于近传感与边缘计算场景。

摘要:

乘累加(MAC)是信号处理、数值模拟和机器学习领域的核心计算操作。本研究提出了一种可扩展、可编程的频域并行计算方案,利用吉赫兹(GHz)频段声波的非线性效应实现计算。通过将数据编码在频域中,单个非线性声波器件可同时执行十亿次算术运算。在铌酸锂(LN)上制备的单个器件占片面积仅0.03 mm²,可实现0.0144太次每秒浮点运算(TFLOPS)的算力,对应的计算面积密度为0.48 TFLOPS/mm²,核心能效为0.14 TFLOPS/瓦。作为应用验证,我们演示了两个16×16矩阵的乘法运算,以及128×128像素图像的卷积成像处理。该技术有望在近传感器信号处理和边缘计算领域实现广泛应用。

英文摘要:

Multiply-accumulation (MAC) is a crucial computing operation in signal processing, numerical simulations, and machine learning. This work presents a scalable, programmable, frequency-domain parallel computing leveraging gigahertz (GHz)-frequency acoustic-wave nonlinearities. By encoding data in the frequency domain, a single nonlinear acoustic-wave device can perform a billion arithmetic operations simultaneously. A single device with a footprint of 0.03 mm$^2$ on lithium niobate (LN) achieves 0.0144 tera floating-point operations per second (TFLOPS), leading to a computing area density of 0.48 TFLOPS/mm$^2$ and a core power efficiency of 0.14 TFLOPS/Watt. As applications, we demonstrate multiplications of two 16-by-16 matrices and convolutional imaging processing of 128-by-128-pixel photos. Our technology could find versatile applications in near-sensor signal processing and edge computing.

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。

↑