Coding and Decoding Method of Transmission Control Protocol Based on Network Coding

Abstract

The invention discloses a coding and decoding method of a transmission control protocol based on network coding. The method optimizes the coding method at the network coding layer, distinguishes between normal data packets and redundant packets, and uses different data packets for different types of data packets. Coding method, for the normal coded data packet, the direct XOR method is used to realize the coding; for the redundant data message, the coding coefficient is randomly selected from a certain row in the Vandermonde matrix, and then the redundant data is obtained by using the linear coding method according to the coding coefficient. Encode data packets to reduce encoding complexity; optimize the design of encoded data packet headers at the network layer, fully consider the encoding coefficients and parameters required for decoding, shorten the encoding header overhead, and improve transmission efficiency; optimize at the network encoding layer The decoding algorithm is improved, which makes the Gaussian elimination process simpler, thereby reducing the decoding complexity.

Description

technical field [0001] The invention relates to the field of transmission control protocol (TCP, Transmission Control Protocol) of network coding, in particular to a coding and decoding method of transmission control protocol based on network coding. Background technique [0002] The Transmission Control Protocol (TCP) is a connection-oriented, reliable, byte-stream-based transport layer communication protocol. Because it can provide very reliable transmission services in wired network scenarios, it has become the basic transport protocol of the Internet. . At the beginning of the design, the TCP protocol was specially designed for wired networks. The TCP protocol considered all packet loss in the network to be caused by network congestion. However, in actual wireless network applications, packet loss in the network is not only caused by congestion, but more likely due to random packet loss caused by unsatisfactory wireless transmission characteristics. In a wireless netwo...

AI Technical Summary

Problems solved by technology

[0008] 1) In the network coding layer, the original data is encoded by random linear coding. Considering that the TCP protocol is a byte stream-based transmission protocol, its coding coefficient is generally in the finite field GF(2 8 ), assuming that the current value of each encoding window is W, and the length of the TCP data message is L bytes, then to generate a new encoded data message, WL times of finite field multiplication and (W- 1) L times of finite field addition operations, which increases the computational complexity of the sending end

[0009] 2) Due to the use of random linear coding, it is necessary to add coding coefficients to the header of the coded data packet, resulting in a more complex coded header, more redundant information, and reduced transmission efficiency

[0010] 3) When the random linear coding method is used, the receiving end will perform a Gaussian elimination every time it receives an encoded data packet, which greatly increases the decoding complexity of the receiving end

[0011] The above problems make the implementation of transmission control protocol based on network coding more complicated, which is not conducive to the actual implementation

Images