Sandwich-RAM: An Energy-Efficient In-Memory BWN Architecture with Pulse-Width Modulation

📋 论文概要

本论文提出了一种名为Sandwich-RAM的存内计算架构，针对二值权重网络(BWN)实现了高能效的乘加运算。通过将权重存储与计算集成在SRAM中，并采用脉宽调制技术，有效减少了数据移动带来的功耗损失，适用于低功耗IoT设备。

💡 主要创新点

• 提出Sandwich-RAM结构，将权重存储和乘加计算融合在同一个SRAM阵列中，避免了传统架构中特征与权重数据的大量移动。
• 引入脉宽调制(PWM)技术，通过调制脉冲宽度实现高效的二值权重乘加运算，进一步降低了计算能耗。
• 针对二值权重网络进行优化，在保持精度的同时显著提升了能效比，为边缘端CNN部署提供了新方案。

🏷 关键词

📄 原文摘要

state-of-the-art results in the field of visual perception, drastically changing the traditional computer-vision framework. However, the movement of massive amounts of data prevents CNN’s from being integrated into low-power IoT devices. The recently proposed binaryweight network (BWN) reduces the complexity of computation and amount of memory access. A conventional digital implementation, which is composed of separate feature/weight memories and a multiply-and-accumulate (MAC) unit, requires large amounts of data to be moved [3]. To reduce power the weight memory and the computations are integrated together, into an in-memory computation architecture [1,2,5]. However, feature data is still stored externally, so data movement has only been partially addressed, especially for BWN. This

👥 作者与机构

Jun Yang1, Yuyao Kong1, Zhen Wang2, Yan Liu1, Bo Wang1, Shouyi Yin3, Longxin Shi1

Southeast University, Nanjing, China Boxing Electronics, Nanjing, China 3 Tsinghua University, Beijing, China 1 2 Convolutional neural networks (CNN) achieve