Evaluating Neural Network-Inspired Analog-to-Digital Conversion With Low-Precision RRAM
Evaluating Neural Network-Inspired Analog-to-Digital Conversion With Low-Precision RRAM
复制标题
DOI：
10.1109/tcad.2020.3013563
复制
发表时间：
2021-05
影响因子：
2.9
通讯作者：
Weidong Cao;Liu Ke;Ayan Chakrabarti;Xuan Zhang
中科院分区：
文献类型：
--
作者：
Weidong Cao;Liu Ke;Ayan Chakrabarti;Xuan Zhang
最近的工作已经证明了神经网络启发的模数转换器（NNADCs）在许多新兴应用中的巨大潜力。这些nnadc通常依赖于电阻随机存取存储器（RRAM）器件来实现基本的神经网络操作，并且通常需要高精度RRAM （6-12 b）来实现中等量化分辨率（4-8 b）。然而，这种对RRAM精度的乐观假设并没有得到大规模生产过程中实际RRAM阵列的很好支持。在本文中，我们评估了两种采用低精度RRAM器件的NNADC新设计。它们利用了传统的两阶段/流水线硬件架构和基于深度学习的定制构建块设计方法。SPICE仿真结果证明了采用4-b RRAM器件的8-b分段NNADC和采用3-b RRAM器件的14-b流水线NNADC的稳健设计。对两种nnadc的评价表明，流水线架构可以更好地使用较低精度的RRAM实现更高的分辨率。我们还对nnadc的构建块进行了设计空间探索，以实现平衡的性能权衡。综合比较显示，与最先进的NNADC和传统adc相比，流水线NNADC的功率、速度性能和竞争优势（FoMs）有所提高。此外，所提出的流水线NNADC可以支持可重构的高分辨率非线性量化，具有高转换速度和低转换能量，为近传感器处理提供智能模拟-信息接口。
Recent work has demonstrated great potentials of neural network-inspired analog-to-digital converters (NNADCs) in many emerging applications. These NNADCs often rely on resistive random-access memory (RRAM) devices to realize basic NN operations, and usually need high-precision RRAM (6–12 b) to achieve moderate quantization resolutions (4–8 b). Such an optimistic assumption of RRAM precision, however, is not well supported by practical RRAM arrays in the large-scale production process. In this article, we evaluate two new designs of NNADC with low-precision RRAM devices. They take advantage of traditional two-stage/pipelined hardware architecture and a custom deep-learning-based building block design methodology. Results obtained from SPICE simulations demonstrate a robust design of an 8-b subranging NNADC using 4-b RRAM devices, as well as a 14-b pipelined NNADC using 3-b RRAM devices. The evaluations on the two NNADCs suggest that pipelined architecture is better to achieve higher-resolution using lower precision RRAM. We also perform design space exploration on the building blocks of NNADCs to achieve a balanced performance tradeoff. Comprehensive comparisons reveal improved power, speed performance, and competitive figure of merits (FoMs) of the pipelined NNADC, compared with state-of-the-art NNADCs and traditional ADCs. In addition, the proposed pipelined NNADC can support reconfigurable high-resolution nonlinear quantization with high conversion speed and low conversion energy, enabling intelligent analog-to-information interfaces for near-sensor processing.