摘要
文章研究目的为探究AI增强技术如何应用于动态网络编码传输协议,以达到提升协议性能、适应性的目标。研究阶段,在对动态网络编码传输协议内涵、实现过程进行剖析后,提出基于CNN-LSTM的混合模型预测无线链路质量、基于fountain codes喷泉码构建分层编码树来动态调整编码率、冗余模式的策略以及面向FPGA硬件的编码器/解码器模块加速设计策略。期望本文可为通信网络公司提供技术借鉴、参考价值,通过对动态网络编码传输协议的优化、改进,不断提升协议的适应性,为未来大吞吐量、海量数据传输场景提供有效的技术支持。
The purpose of this paper is to explore how AI enhancement technology is applied to dynamic network coding transport protocols,so as to achieve the goal of improving protocol performance and adaptability.In the research stage,after analyzing the connotation and implementation process of dynamic network coding transmission protocol,A hybrid model based on CNN-LSTM is proposed to predict wireless link quality,a hierarchical coding tree is constructed based on fountain codes to dynamically adjust coding rate and redundancy mode,and an FPGA hardware oriented encoder/decoder module acceleration design strategy is proposed.It is expected that this paper can provide technical reference and reference value for communication network companies.By optimizing and improving the dynamic network coding transmission protocol,the adaptability of the protocol can be continuously improved,and effective technical support can be provided for the future large-throughput and massive data transmission scenarios.
作者
陈智
CHEN Zhi(Guangxi Bolian information and communication Technology Co.,LTD GuangXi Nanning 530023)
出处
《长江信息通信》
2025年第6期127-129,132,共4页
Changjiang Information & Communications
作者简介
陈智(1982-),男,广东湛江人,本科,助理工程师,研究方向:计算机技术。
