Research On CCN Caching Strategy Based On Ocean Bottom Observatory Network

The ocean accounts for more than 70% of the earth's area,but,as human being,the cognition of the ocean is very limited.With the continuous development of marine science and marine economy,some scholars put forward the necessity of building marine Internet.Although the development of the land Internet has been very mature,it can't be directly copied to the ocean in the actual deployment.As an important means of ocean observation,the undersea observation network adopts the method of extending the land network to the ocean.In terms of deployment,the optical cable is generally used as the backbone network in the near shore and the wireless communication is used as the extension in the far ocean.The two are seamlessly connected to form a threedimensional undersea observation system.At present,the undersea observation network has completed large-scale deployment in many countries,and has made significant contributions to marine data collection,ecological monitoring,disaster prevention and other undertakings.Based on the above-mentioned idea of learning from the land in the sea,we can refer to the technology of the land Internet in the process of building the marine Internet and further optimize the marine adaptability;however,some scholars think that the TCP / IP protocol stack of the land Internet has encountered the bottleneck of improvement,so it should not be optimized by patching,but should be reexamined with an innovative idea.In recent years,new network architecture has been proposed,one of which is CCN: content center network.However,the CCN basic network model can't be directly applied to the ocean,and further theoretical analysis and architecture improvement are needed.Therefore,through the investigation and analysis of CCN and undersea observation network,this paper summarizes the adaptability performance of CCN network architecture to undersea observation system and the advantages compared with TCP / IP network,and draws the conclusion that CCN is more suitable for undersea observation system than traditional network model,and has the hope to become the network architecture of the future marine Internet.On the basis of the above,this paper proposes an improvement of the CCN architecture based on the undersea observation network: the network reminder mechanism is added to the basic CCN architecture,and the implementation of this mechanism mainly depends on the new data structure "Information Package" and "Request Package" and "Data Package" in the original structure are improved.Through the reminding and confirmation mechanism for long link and large content requests,it gives the requested party certain defense ability and initiative,and to a certain extent,it plays a role in limiting abnormal requests and traffic control.On the basis of such data structure,in order to design a CCN cache strategy that is more suitable for undersea observation network,this paper makes a further investigation on the cache strategy,disassembles and analyzes the optimal model of the CCN cache placement algorithm,and proposes an improved probability cache placement strategy based on the undersea observation network.This cache placement strategy judges the hop number of the source and the current node determines the probability of data caching on the routing node,so that the content of high-frequency requests can be pushed around the requester at a faster speed.In the network structure of undersea observation system,especially in the wireless extension part,due to the unstable link state and high bit error rate,the cost of network transmission is relatively large,so we should try to reduce the number of routing and forwarding hops as much as possible through the cache mechanism;at the same time,due to the limited cache space and energy,we should try to reduce the redundancy of the cache.Therefore,this paper proves through experiments that: because the algorithm proposed in this paper takes into account the diversity of cache in the network and the effectiveness of cache,compared with the fixed probability cache algorithm such as pro0.8,pro0.2 and the classic LCE,LCD algorithm,this algorithm has better performance under the uniform distribution and homogeneous distribution of content requests,and the size of different node cache space,Mainly reflected in the average cache hit rate and the average route hop ratio.It is concluded that the cache algorithm proposed in this paper is more suitable for the bottom observation network based on CCN.