arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2512.09155eess.SPcs.ARcs.MS
收藏

一种用于FPGA上高精度算术的混合余数浮点数值架构

A Hybrid Residue Floating Numerical Architecture for High Precision Arithmetic on FPGAs

École de technologie supérieure (ÉTS)(高等技术学院);Evolution Optiks R&D Inc.(进化光学研发有限公司)

Mostafa Darvishi

AI总结:

本文提出了一种混合余数浮点数值架构,通过结合无进位余数通道和轻量级浮点数缩放因子,在FPGA上实现高精度算术的高效计算。

摘要:

浮点运算在FPGA平台上仍然昂贵,由于宽数据路径和归一化逻辑,这促使了替代表示法的出现,这些表示法在较低成本下保持动态范围。本文介绍了一种混合余数浮点数值架构(HRFNA),这是一种统一的算术系统,结合了无进位余数通道和轻量级浮点数缩放因子。我们开发了完整的数学框架,推导出有界误差归一化规则,并展示了针对模运算、指数管理和混合重建的FPGA优化微架构。HRFNA在Xilinx ZCU104上实现,通过Vitis模拟、RTL综合和芯片内ILA跟踪确认了精确周期正确性。该架构在IEEE 754单精度基准上实现了超过2.1倍的吞吐量提升,并在保持长迭代序列的数值稳定性的同时,将LUT减少38-52%。这些结果表明,HRFNA为现代FPGA设备上的浮点运算提供了一种高效且可扩展的替代方案。

英文摘要:

Floating point arithmetic remains expensive on FPGA platforms due to wide datapaths and normalization logic, motivating alternative representations that preserve dynamic range at lower cost. This work introduces the Hybrid Residue Floating Numerical Architecture (HRFNA), a unified arithmetic system that combines carry free residue channels with a lightweight floating point scaling factor. We develop the full mathematical framework, derive bounded error normalization rules, and present FPGA optimized microarchitectures for modular multiplication, exponent management, and hybrid reconstruction. HRFNA is implemented on a Xilinx ZCU104, with Vitis simulation, RTL synthesis, and on chip ILA traces confirming cycle accurate correctness. The architecture achieves over 2.1 times throughput improvement and 38-52 percent LUT reduction compared to IEEE 754 single precision baselines while maintaining numerical stability across long iterative sequences. These results demonstrate that HRFNA offers an efficient and scalable alternative to floating point computation on modern FPGA devices.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑