论文部分内容阅读
在差错信道中传输IP分组时,信道误码容易导致数据包丢失。在无反馈信道的情况下,无法使用反馈重发对数据包进行恢复。因此,该文研究了分组协议中报头字段的差错恢复问题。分析了TCP/UDP/IP等协议中的关键字段,将用于区分传输链路中各路会话的协议字段统称为复用ID。建立了复用ID估计的理论模型,分析了已有文献中基于校验方法对活动复用ID进行检测的缺点,提出利用TCP序列号间关联性的全新检测方法,并针对五元组(源/目的 IP地址,协议号,源/目的端口)提出了估计算法。实验结果表明该方法大大降低了五元组的差错概率,提高了最终的信源还原效果。最后对实验结果进行了分析。
When transmitting IP packets in an error channel, channel error easily results in packet loss. In the absence of a feedback channel, data packets can not be recovered using feedback retransmission. Therefore, this paper studies the error recovery of header fields in packet protocols. The key fields in the TCP / UDP / IP protocol are analyzed. The protocol fields used to distinguish each session in the transmission link are collectively referred to as the multiplexing ID. The theoretical model of reused ID estimation is established, the shortcomings of existing recursive ID checking methods based on checking methods are analyzed, a new detection method using association between TCP sequence numbers is proposed, / Destination IP address, protocol number, source / destination port) proposes an estimation algorithm. The experimental results show that this method can greatly reduce the error probability of quintuplet and improve the final source reduction effect. Finally, the experimental results were analyzed.