为扩展现实感知设计高吞吐混合精度神经处理芯片，显著降低功耗与面积。
XR-NPE: High-Throughput Mixed-precision SIMD Neural Processing Engine for Extended Reality Perception Workloads
- 支持多种低精度格式，动态适配计算层，大幅减少内存带宽
- 实测比顶尖方案面积小42%、功耗降38%，能量效率提升23%
- 适合资源受限的XR设备，代码开源便于复用
本文提出XR-NPE，一种面向扩展现实（XR）感知任务（如视觉惯性里程计、物体分类、眼动追踪）的高吞吐混合精度SIMD神经处理引擎。首次支持FP4、Posit(4,1)、Posit(8,0)和Posit(16,1)格式，采用层自适应混合算法实现超低比特精度，显著降低内存带宽需求，并配合量化感知训练实现极小精度损失。提出的可重构尾数乘法与指数处理电路（RMMEC）减少SIMD MAC计算单元中的暗硅问题，结合选择性电源门控，使算术强度提升2.85倍。XR-NPE在28nm CMOS工艺下最高工作频率达1.72 GHz，面积仅0.016 mm²，算术强度为14 pJ，相比现有最优MAC方案面积减少42%、功耗降低38%。基于XR-NPE的AXI接口矩阵乘法协处理器，在VCU129上相较当前最优加速器减少1.4倍LUT、1.77倍FF，能效提升1.2倍；在VIO任务中能效提升23%、计算密度提升4%。该设计具备可扩展性与精度自适应能力，适用于未来资源受限的XR设备。完整代码与复现工具已公开于GitHub：https://github.com/mukullokhande99/XR-NPE。
原文摘要 · Abstract (English)
This work proposes XR-NPE, a high-throughput Mixed-precision SIMD Neural Processing Engine, designed for extended reality (XR) perception workloads like visual inertial odometry (VIO), object classification, and eye gaze extraction. XR-NPE is first to support FP4, Posit (4,1), Posit (8,0), and Posit (16,1) formats, with layer adaptive hybrid-algorithmic implementation supporting ultra-low bit precision to significantly reduce memory bandwidth requirements, and accompanied by quantization-aware training for minimal accuracy loss. The proposed Reconfigurable Mantissa Multiplication and Exponent processing Circuitry (RMMEC) reduces dark silicon in the SIMD MAC compute engine, assisted by selective power gating to reduce energy consumption, providing 2.85x improved arithmetic intensity. XR-NPE achieves a maximum operating frequency of 1.72 GHz, area 0.016 mm2 , and arithmetic intensity 14 pJ at CMOS 28nm, reducing 42% area, 38% power compared to the best of state-of-the-art MAC approaches. The proposed XR-NPE based AXI-enabled Matrix-multiplication co-processor consumes 1.4x fewer LUTs, 1.77x fewer FFs, and provides 1.2x better energy efficiency compared to SoTA accelerators on VCU129. The proposed co-processor provides 23% better energy efficiency and 4% better compute density for VIO workloads. XR-NPE establishes itself as a scalable, precision-adaptive compute engine for future resource-constrained XR devices. The complete set for codes for results reproducibility are released publicly, enabling designers and researchers to readily adopt and build upon them. https://github.com/mukullokhande99/XR-NPE.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。