谷歌浏览器插件
订阅小程序
在清言上使用

A Fast and Resource-efficient GSOP Implementation for Optical Coherent Receivers

2022 IEEE 16th International Conference on Solid-State & Integrated Circuit Technology (ICSICT)(2022)

引用 0|浏览15
暂无评分
摘要
In-phase/quadrature(I/Q) imbalance compensation is an important procedure in optical coherent systems, and the Gram-Schmidt orthogonalization procedure (GSOP) is one of the most widely used compensation algorithms. However, the division and square-root operations consume a lot of hardware resources, limiting the speed of FPGA implementation. This paper proposes a fast and resource-efficient GSOP implementation by converting square-root and division operations into linear fitting. We introduce the Newton iteration method to compensate for the accuracy degradation. The FPGA implementation shows that the throughput of the 128-way-parallel GSOP is up to 100Gbps at the frequency of 392MHz.
更多
查看译文
关键词
128-way-parallel GSOP,bit rate 100 Gbit/s,compensation algorithms,division operations,fast GSOP Implementation,FPGA implementation,frequency 392.0 MHz,Gram-Schmidt orthogonalization procedure,hardware resources,I/Q imbalance compensation,in-phase/quadrature imbalance compensation,linear fitting,Newton iteration method,optical coherent receivers,optical coherent systems,resource-efficient GSOP implementation,square-root operations
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