Research On Interconnection Architecture And Key Technologies Of Information-Centric Internet Of Things

As a new generation of information infrastructure,the Internet of Things(IoT)interconnects and integrates the traditionally separated physical world and cyberspace,which has greatly promoted the transformation of the infor-mation industry and brought a new round of technological revolution.Howev-er,with the rapid development of the IoT,the traditional network architecture based on host address and end-to-end communication model is difficult to meet the needs of users and network development.Its limited address resources,low support for mobility,data distribution,and security are even more obvious.Information-centric networking(ICN),as one of the main research directions of future network architecture,adopts the content-centric architecture,name-based data retrieval,and data-centric security mechanism.Therefore,ICN is regarded as a promising solution for IoT,i.e.,ICN-IoT.However,IoT's specific characteristics such as the massive number of de-vices,strong network heterogeneity,and high security requirements,make the existing ICN still face many challenges in solving the interconnection problem of IoT.Therefore,in this article,we consider the characteristics of IoT environ-ment and combines the advantages of ICN.We first investigate the design of ICN-IoT interconnection framework.Then,we further research two key tech-nologies:in-network caching algorithm and access control for IoT data and devices.Specifically,the main contributions of this article are as follows:(1)Sensory data centric interconnection framework.The sensing layer is the foundation of the IoT networking.Interconnecting heterogeneous sens-ing networks and Internet backbone is the base of achieve IoT's vision of in-terconnecting everything.To achieve this,we first design a Spatio-temporal relationship-based naming scheme for sensing networks.Based on this nam-ing scheme,we propose an efficient scope matching mechanism,name-based node/network discovery methods,and Spatio-temporal relationship-based for-warding and routing strategies.These mechanisms and strategies construct the interconnection framework called Sensory Data-Centric Networking(SDCN),which ensures efficient data collecting and heterogeneous network intercon-necting.(2)Data-driven popularity modeling and caching strategy.The in-network caching mechanism is a key technologies for ICN-IoT to reduce redundan-t traffic,relieve network pressure,and improve data distribution efficiency.To this end,we propose a data-driven popularity modeling and caching algorithm.Through the in-depth analysis of the changing regularity and distribution char-acteristics of IoT data's popularity,a popularity evolving model(PEM)and two modeling methods:a modeling method based on statistics and deep neu-ral networks,are proposed.Finally,the popularity-based cache replacement and the prefetch algorithm are proposed to improve the cache performance and interconnection efficiency of ICN-IoT.(3)Chaotic hierarchical key tree-based access control mechanism.The limited resources,isolated systems,and heterogeneous network characteristics of IoT make the access control of IoT data and devices particularly impor-tant.We propose a chaotic hierarchical key tree-based access control system(HAC),which uses a distributed and cooperative security architecture based on the characteristics of the IoT.By using the hierarchical naming scheme of ICN,a chaotic hierarchical key tree is proposed.Combining with the name-based access policy retrieval and the attribute-based localize command verifica-tion mechanism,HAC ensures secure and efficient access control in ICN-IoT's many-to-many communication and data distribution scenarios.(4)The ICN-IoT system prototype.Base on the proposed interconnection framework and key technologies,we implement the prototype of ICN-IoT pro-totype,i.e.,SDCN.Specifically,the system includes the implementations for ICN-IoT backbone and two typical sensing networks(i.e.,Wireless Sensor Net-work and Mobile Sensing Network)combing with the related naming,match-ing,discovering,switching,forwarding,and routing algorithms/strategies,and further be deployed in our campus and the Great Wall for environmental moni-toring project.To evaluate the proposed interconnection framework and the key technologies,we conducted a series of experiments based on the prototype.The results show that our design can greatly improve the interconnection efficiency and cache performance,and also protect the data security while consuming less network resource compared with the state-of-arts.In summary,this article explores the interconnection and key technology of the ICN-IoT from architecture,cache algorithm,and access control.A se-ries of framework,algorithms,and strategies are proposed and verified through the theoretical analysis and the prototype system.The content in the article provides the theoretical basis and technical support for the interconnection net-work architecture and the related services of ICN-IoT.