Optimization For Earth-lunar DTN Network Transmission Protocol Via Network Coding And Multi-source Bundle Aggregation

In the past 50 years,the Lunar has been the cornerstone of human exploration into deep space.It not only contributes to the scientific cause of mankind,but also has important strategic significance at the national level.Nowadays,with the continuous exploration of the Lunar,the hidden resources of the Lunar poles and the back of the Lunar have attracted the attention of all countries.The manned mission to the Lunar has shifted from the surface of the Lunar to the back of the Lunar.In order to ensure the control of the soft landing spacecraft on the back of the Lunar,the return of vehicle information,and the emergency communication of astronauts,the relay satellite orbit and reliable communication protocol need to be solved.To this end,the United States and Europe have theoretically discovered the feasibility of the Earth-Lunar(L2)Halo orbit as a relay orbit.Combined with foreign theories,China successfully launched the world's first Earth-Lunar L2 point relay satellite in 2018 after several flight demonstrations.At the same time,relying on Delay Disrupt-Tolerant Networking(DTN)to ensure reliable end-to-end transmission capability between heterogeneous networks,a DTN network is constructed.At present,the distance among the three-dimensional Halo orbit,the Earth and the Lunar varies greatly,and it is difficult for the relay satellite to maintain stable high throughput.On the other hand,the increase in traffic volume has put pressure on the relay store-and-forward of multi-source and multi-priority data.In this paper,based on the multi-source and single-relay DTN network configuration,a multi-source bundle aggregation algorithm is proposed.Then,we establish the distance model.The Markov characteristic of the Earth-Lunar link is used to obtain the end-to-end optimal bundle size set of the DTN network through the Markov decision algorithm.Based on this,the relay satellite aggregates two bundles from different sources,and reduces the number of transmission bundles while ensuring the size.From the simulation results,we obtained the optimal bundle size set,and the optimal size decreases with increasing distance.At the same time,the multi-source bundle aggregation algorithm has better performance in reducing the delay of large file transmission than the traditional slice algorithm.The network coding and forwarding strategy are designed for the multi-source and high-priority service forwarding requirements in the Earth-Lunar DTN network.The traditional high-priority forwarding methodaffects the transmission process.This paper combines the network coding with the DTN bundle protocol(BP)to XOR the high-priority bundles with the low-priority bundles in the process.In addition,we design a new retransmission mechanism to ensure that the destination node's encoding packets can obtain the information of the high priority bundles in real time.At the same time,according to the particularity of network coding,the bundle format is modified,which simplifies the offset of the source node and the destination node in the packet header.This method provides guarantee for the original packet to be restored to the original file after decoding.The experimental results show that the network coding scheme needs to sacrifice the accuracy of certain low-priority bundles.However,from the perspective of file size and high-priority bundles ratio,the network coding forwarding scheme can improve the throughput of the overall file and burst high-priority bundles.