18.1 A 600Gb/s DP-QAM64 Coherent Optical Transceiver Frontend with 4x105GS/s 8b ADC/DAC in 16nm CMOS

Guansheng Li,Adesh Garg,Tim He,Ullas Singh, Jiawen Zhang,Lakshmi Rao, Chang Liu, Meisam Nazari,Yang Liu, Yong Liu,Heng Zhang,Tamer Ali, Hyo Gyuem Rhew,Jiayoon Ru,Delong Cui,Ali Nazemi, Bo Zhang,Afshin Momtaz, Jun Cao

2024 IEEE International Solid-State Circuits Conference (ISSCC)(2024)

引用 0|浏览1
暂无评分
摘要
An ADC/DAC-based coherent optical transceiver [1] is a promising technique to boost the throughput of metro/long-haul links beyond 600Gb/s/λ. It leverages the powerful digital signal processors (DSP) in scaled CMOS processes to enable high-spectral-efficiency modulation formats such as dual-polarization quadrature-amplitude modulation (DP-QAM). At the DSP interface, it uses four well-synchronized DACs and ADCs to generate and capture the four streams of analog signals that are carried by the same light in DP-QAM format (Fig. 18.1.1). In 600Gb/s links that use DP-QAM64 and feedforward error correction (FEC) coding, the baud rate can go up to 70GBaud/s. This requires an ADC/DAC bandwidth of 40GHz and a sampling rate of 105GS/s to satisfy the oversampling ratio needed for spectrum shaping and clock-data recovery. Meanwhile, the ADC/DAC have to achieve low noise and high linearity to achieve the SNDR needed for DP-QAM64 at low power consumption. In this paper, we present a 600Gb/s/λ DP-QAM64 coherent optical transceiver frontend that fully integrates a 4-way 78~105GS/s 8b ADC/DAC in a DSP chip in 16nm CMOS. It supports flexible rates of 100, 200, 300, 400, 500, 600Gb/s by DP-QPSK and DP-QAM8/16/32/64 with a variable baud rate for long-haul, metro, and ZR applications.
更多
查看译文
关键词
16-nm CMOS,Systematic Errors,Low Voltage,Tree Structure,Application Rate,Low Power Consumption,Digital Signal Processing,Forward Error Correction,Datapath,High Spectral Efficiency,Quadrature Phase,Modulation Formats,Output In Order,Complex Poles,Differential Pair,Global Circuits,Massive Deployment,Noise Immunity,Loop Filter,Slight Misalignment
AI 理解论文
溯源树
样例
生成溯源树,研究论文发展脉络
Chat Paper
正在生成论文摘要