Audio and Image Cross-Modal Intelligence via a 10TOPS/W 22nm SoC with Back-Propagation and Dynamic Power Gating
Audio and Image Cross-Modal Intelligence via a 10TOPS/W 22nm SoC with Back-Propagation and Dynamic Power Gating
复制标题
通过具有反向传播和动态功率门控功能的 10TOPS/W 22nm SoC 实现音频和图像跨模态智能
DOI：
--
复制
发表时间：
2022
期刊：
影响因子：
--
通讯作者：
D. Sylvester
中科院分区：
文献类型：
--
作者：
Zichen Fan;Hyochan An;Qirui Zhang;Boxun Xu;Li Xu;Chien;Yimai Peng;Ang Cao;Bowen Liu;Changwoo Lee;Zhehong Wang;Fanghao Liu;Guanru Wang;S. Jiang;Hun;D. Blaauw;D. Sylvester
我们提出了一种超低功耗多媒体信号处理器（MMSP）SoC，它集成了多功能深度神经网络（DNN）引擎与音频和图像信号处理加速器，用于跨模态物联网智能。所提出的MMSP具有2 MB MRAM，用于在片上存储所有DNN权重，并使用MRAM缓存和动态功率门控的节能型EEPROM。SoC实现了高达3-10 TOPS/W的峰值能效，功耗仅为0.25-3.84 mW。作为第一个在单个加速器SoC上演示CNN、GAN和反向传播（BP）进行跨模态融合的公司，它的能效比最先进的DNN处理器高出1.4 - 4.5倍。
We present an ultra-low-power multimedia signal processor (MMSP) SoC that integrates a versatile deep neural network (DNN) engine with audio and image signal processing accelerators for cross-modal IoT intelligence. The proposed MMSP features 2MB MRAM to store all DNN weights on-chip with an energy-efficient dataflow using an MRAM-cache and dynamic power gating. The SoC achieves up to 3-10 TOPS/W peak energy efficiency and consumes only 0.25-3.84 mW. Being the first to demonstrate CNN, GAN, and back-propagation (BP) on a single accelerator SoC for cross-modal fusion, it outperforms state-of-the-art DNN processors by 1.4 - 4.5× in energy efficiency.