Method for forward error correction video transmission based on pjsip protocol stack

By employing a forward error correction video transmission method based on the PJSIP protocol stack, I-frames and P-frames are processed differently, which solves the problems of bandwidth waste and decoding chain interruption in low-latency video transmission in existing technologies, and achieves stability and reliability of video playback.

CN122420291APending Publication Date: 2026-07-17GUANGDONG ENVISION TECH CO LTD

Patent Information

Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
GUANGDONG ENVISION TECH CO LTD
Filing Date
2026-06-09
Publication Date
2026-07-17

AI Technical Summary

Technical Problem

Existing real-time video transmission solutions rely on round-trip latency, which is suitable for low-latency real-time video. Unified redundancy increases bandwidth and complexity, but does not necessarily improve the RTP fragment-level recovery effect. The different importance of I-frames and P-frames and the lack of differentiated protection lead to decoding chain interruption.

Method used

A forward error correction video transmission method based on the PJSIP protocol stack is adopted to differentiate I-frames and P-frames. Redundant repair packets are generated by grouping, aggregation, adding error correction headers and RS encoding to adapt to their respective characteristics, optimize bandwidth resources, and ensure stable recovery of I-frames and P-frames.

Benefits of technology

It achieves stable recovery of I-frames and P-frames under low latency conditions, avoids decoding chain interruption, optimizes bandwidth utilization, and ensures the stability and reliability of video playback.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN122420291A_ABST
    Figure CN122420291A_ABST
Patent Text Reader

Abstract

本发明涉及一种基于PJSIP协议栈的前向纠错视频传输方法,包括:采集视频并编码,获取第一RTP媒体包,判断为I帧或P帧,若为I帧则进行I帧分组,再附加第一基础公共纠错头,得到第二RTP媒体包并将其发送至接收端,同时备份第二RTP媒体包;若为P帧则进行P帧聚合,接着附加第二基础公共纠错头,再在第二基础公共纠错头上增加扩展头,得到第三RTP媒体包并将其发送至接收端,同时备份第三RTP媒体包;进行RS编码,生成冗余修复包并将其通过IP网络传输至接收端;接收端通过IP网络接收相应媒体包和冗余修复包;其对I帧和P帧都是差异化处理,适配相应帧的特性,最优利用带宽资源,不浪费冗余。
Need to check novelty before this filing date? Find Prior Art