Data Publish/Subscribe with Network Coding in DTN

Xia Yi,Deng Guanghong

doi:10.3991/ijoe.v12i06.5512

Abstract

A data Publish/Subscribe method with network coding (DPS-NC) used in delay tolerant network is proposed. Based on the publish/subscribe mechanism, the topic data were encoded and transmit with random network coding method, which can take full advantage of the network capacity to multicast. At the same time, for working out the problem of large data redundancy and vast invalid de-liveries in the network coding broadcast process, a packet routing scheme based on ant colony algorithm was brought. In the end, some simulation experiments are carried out and results show that the DPS-NC method can get better delivery performance.

Highlights

As a new transmission mode in Delay Tolerant Networks (DTN) [1, 2] network, network coding has been proved that it can improve network utilization, reduce transmission delay and enhance network security
Some simulation experiments are carried out and the results show that the data Publish/Subscribe method with network coding (DPS-NC) method can get better delivery performance
In the DTN environment, the flood broadcast mechanism of network coding will bring out some problems such as information redundancy, invalid transmission, and data congestion [5]

Summary

INTRODUCTION

As a new transmission mode in Delay Tolerant Networks (DTN) [1, 2] network, network coding has been proved that it can improve network utilization, reduce transmission delay and enhance network security. The data publish/subscribe mechanism [6] was brought out for improving the data delivery efficiency and reducing information redundancy. It establish efficient and loose coupling of information distribution channels between data producers and consumers, to improve data transmission efficiency [7]. Direct at the demand of data distribution in DTN network, a network coding-based data publish/subscribe method (DPS-NC) is proposed. It introduce the network coding method into data transmission process to improve data delivery performance and reduce data redundancy. According to the linear equations solvable conditions, if the coding vector matrix composed by gij is full rank, the data from source node can be decoded at sink

Data publication and subscription

SIMULATION EXPERIMENTS

Findings

CONCLUSIONS S