当前位置:首页 > 报告详情

DC25_PAPER_Track9_LowLatencySpeculativeErrorCorrection_Nir_V10.pdf

上传人: S** 编号:1240815 2026-05-16 25页 2.30MB

1、 Information Classification:General Low latency speculative error correction using simplified ML detector for 64Gbps wireline transceiver Ehud Nir,Mansi Mehrotra,Amin Karami,Chris Holdenried,Robert Wang Cadence Design Systems,Toronto,Canada Information Classification:General Abstract Wireline interc

2、onnect data-rates have been doubling consistently every 3-4 years due to increasing demand for higher connectivity speed in datacenters.With limited improvement of the electrical channels,MLSE has emerged as a key digital technique for boosting signal power in wireline links,at the price of increase

3、d latency and die area.This paper presents an ultra-low latency speculative error correction(SEC)engine embedded within the DFE.It uses a simplified ML detector to correct single symbol errors before they propagate into a burst.Significant reduction of the trellis size is achieved by using soft info

4、rmation together with assumptions on the error distribution.The design was customized for an existing Gen6 PCIe PHY in 5nm CMOS technology.Simulated and measured SNR gain exhibits comparable performance at a fraction of the latency and area of full MLSE.Authors Biography Ehud Nir Ehud Nir received t

5、he B.Sc.and M.Sc.degrees in Electronics Engineering from Tel Aviv University,Tel Aviv,Israel,in 2001 and 2013,respectively.He has over 25 years of experience in custom digital and DSP circuit design.He is currently a director of digital engineering at Cadence Design Systems Inc.,developing high-spee

6、d digital circuits for wireline communication.Mansi Mehrotra Mansi Mehrotra received the Bachelor of Engineering degree in Electronics Engineering from Indian Institute of Technology,BHU(IIT BHU),India,in 2012.She has over 11 years of experience in digital design of Controllers and PHY Subsystems.Sh

word格式文档无特别注明外均可编辑修改,预览文件经过压缩,下载原文更清晰!
三个皮匠报告文库所有资源均是客户上传分享,仅供网友学习交流,未经上传用户书面授权,请勿作商用。
1. **技术目标**:提出超低延迟推测错误校正(SEC)引擎,嵌入DFE中,通过简化ML检测器在错误传播前纠正单符号错误,提升64Gbps PCIe Gen6 PHY性能。 2. **核心创新**:结合软信息与错误分布假设,大幅缩减网格大小;SEC延迟仅1ns(约1周期),面积4500μm²(5nm CMOS)。 3. **性能数据**:SNR增益达全MLSE的99.97%(仅差0.03dB),SER较传统DFE提升15-100倍,接近MLSE水平(SER=1e-6时SNR=19.67dB)。 4. **验证结果**:通过36dB信道实测,SEC-PUDFE在SER<1e-4时较DFE提升15-60倍,错误数与MLSE偏差±5%。
SEC如何提升性能? MLSE为何复杂? SEC如何降低延迟?
客服
商务合作
小程序
服务号
折叠