Research On Network Coding Technology Based On Packet Importance

Network coding technology can significantly improve the data throughput of the network,and has received extensive attention from many scholars.However,the current network coding technology regards all data packets as equal importance,and does not consider the difference in importance of data packets.In some actual communication environments,the importance of different data packets is not consistent,such as safety in the Internet of Vehicles.Information data packets have a higher level of importance than entertainment information data packets.The video stream is divided into base layer and enhancement layer to send different data layers for users with different channel transmission conditions.Therefore,this paper will study the network coding technology according to the difference in the importance of data packets.The main work is as follows:This paper proposes a sparse network coding(Sparse Network Coding,SNC)method based on packet importance.Since sparse network coding has partial decoding properties and the average decoding delay is low,this paper deduces the model value of the lower bound of the number of transmissions required for partial decoding of important data packets through the Markov absorption chain model.The model values are compared and the validity of the model is confirmed.this paper proposes an overall sequential decoding delay index for important data packets,and adjusts the sparsity to weigh the average decoding delay and the sequential decoding delay of important data packets enables the network system not only to receive important information preferentially during transmission,but also to maintain a low average decoding delay.A sliding-window Random Linear Network Coding(RLNC)method based on packet importance is proposed.By setting appropriate redundancy for important data packets to ensure the probability of successful decoding in the lost link,and using a cyclic shift method to store the coding coefficients,the transmission system does not need to use feedback to automatically update the coding window.The simulation results show that,the proposed method can improve the decoding probability of partial data packets and reduce the sequential decoding delay.