This work presents NeuDW-CIM, a highly efficient neuromorphic Compute-in-Memory (CIM) macro for Spiking Neural Networks (SNNs) implemented in 65 nm CMOS. The design introduces a custom twin 9T bit-cell for ternary in-puts/weights and a reconfigurable non-linear In-Memory ADC (IMA). The macro supports two specialized modes: 1) Nonlinear Dendrite (NLD) mode, which utilizes reconfigurable IMA to emulate biological dendritic functions, achieving measured accuracies of 97.2% on N-MNIST and 95.5% on DVS Gesture; and 2) Top-K Winner (KWN) mode, featuring an early-stopping mechanism that reduces IMA conversion latency by 30% and digital LIF latency by 10x. Benefiting from the sparse update in KWN mode, NeuDW-CIM achieves a measured energy efficiency (EE) of 0.8 pJ/SOP (1.6x improvement).

翻译：本文提出了NeuDW-CIM，一款基于65纳米CMOS工艺实现的高效神经形态存算一体（CIM）宏单元，专用于脉冲神经网络（SNN）。该设计采用定制的双管9T（twin 9T）存储单元以支持三值输入/权重，并集成了可重构的非线性存内模数转换器（IMA）。该宏单元支持两种专用模式：1）非线性树突（NLD）模式，利用可重构IMA模拟生物树突功能，在N-MNIST和DVS Gesture数据集上分别达到97.2%和95.5%的实测准确率；2）Top-K胜出者（KWN）模式，该模式采用早停机制，使IMA转换延迟降低30%，数字LIF延迟降低10倍。得益于KWN模式下的稀疏更新特性，NeuDW-CIM实现了0.8 pJ/SOP的实测能效（EE），较同类方案提升1.6倍。