A Research On Edge-enabled IoT Services Orchestration Framework And Its Key Technologies

In the current digital age,the edge-enabled Internet of Things(Io T)continuously generates and transmits data and provides a variety of services to the users.Due to resource limitation and heterogeneity constraints,end-to-end service orchestration from service providers to consumers is a key challenge for the sustainable development of the resource-constrained Io T.Edge computing provides a feasible solution for the resource limitations by bringing resources to the "edge" of the network;however,optimal resource allocation,security,and privacy challenges becomes imminent in this distributed paradigm.In addition,the heterogeneity of hardware and communication protocols brings new challenges to the application development for distributed Io T.At present,there is limited research on the overall end-to-end Io T service orchestration lifecycle,so an efficient framework is needed to optimize the allocation of resources considering the massive Io T traffic,provide an application platform for distributed Io T application development,and ensure the security and privacy of services.The model-driven Io T software platform offers essential capabilities for efficient service orchestration by providing access to Io T data streams and control for developing Io T applications.Moreover,the geographically dispersed and resource-constrained devices may become soft targets for the adversaries to attack network infrastructure,resulting in huge security and privacy challenges.As a result,the edge-enabled Io T faces a variety of security and privacy problems in the massive data generation and heterogeneous infrastructure environment,which makes the end-to-end Io T service orchestration a challenging task.Aiming at the challenges of edge-enabled Io T,this dissertation proposes a hierarchical Edge-enabled Io T Service Orchestration(EISO)framework for Io T service orchestration and studies its key technologies such as application development,security,and privacy to ensure the seamless deployment of the framework.The key technologies can ensure that the EISO framework can effectively orchestrate and coordinate the end-to-end delivery of a variety of services from data capturing to the service delivery.Specifically,the main contributions of this dissertation are as follows:(1)In order to support the seamless delivery of edge-enabled Io T services,a novel EISO framework for resource optimization is proposed to address the resource demand challenges based on delay and service delivery rate constraints.EISO is a hierarchical framework that can run under the continuous interaction of the perception layer,mist layer,edge layer,cloud layer,and application layer.Using a comprehensive analysis framework of EISO,the service provisioning is expressed as an optimization problem based on service delivery rate and delay for different latency applications.To validate the system,we perform an extensive evaluation using real-world parameters,which shows that the proposed framework optimally allocates resources,reduces system latency,and minimizes overall energy consumption for large-scale services orchestration.(2)Aiming at the challenge of heterogeneity of Io T devices,this study proposes an application development technique for customized and user-specific service orchestration,which uses the collaboration of Model-Driven Development(MDD)and attribute-driven design to develop Io T applications.This technique designs models at different levels of abstraction and provides implementable software artifacts for heterogeneous platforms.It consists of two main steps,including iterative architecture development using attribute-driven design and generating MDD metamodels to hierarchically transform the design components into the software artifacts.We evaluate the proposed technique by generating executable code for a real-world system.The experiments show that,compared with traditional methods,this technique achieves higher application development accuracy and provides technical support for the whole life cycle of Io T applications.(3)In the edge-enabled Io T services orchestration paradigm,there is a high demand for security due to the lack of security solutions caused by resource-constraints.Based on the proposed collaborative framework,this study designs a new defense mechanism to meet the security challenges of the EISO framework.The defense mechanism uses network-based traffic measurement technique to protect edgeenabled Io T services from security threats.We use the open-source mininet network simulator and floodlight controller to conduct a rigorous evaluation to simulate the attack scenario and demonstrate the effectiveness of the proposed solution.The experimental results show that the proposed scheme reduces the packet loss rate,balances the traffic on critical links,and achieves higher attack detection and mitigation accuracy.(4)In order to solve the privacy issues during end-to-end service delivery and ensure the seamless orchestration of services,this study proposes a privacy protection technique for the EISO framework.The traditional privacy protection techniques use centralized management strategies,which leads to a single point of failure,resulting in a lack of trust and privacy protection challenges including state modification,information disclosure,and identity theft.To address these challenges,we propose a solution that uses blockchain-based distributed authentication policies deployed at the edge layer of the EISO framework to enforce privacy.In the proposed scheme,the trusted edge nodes implement an authentication algorithm that uses public/private key matching to authenticate Io T transactions.The experimental evaluation shows that the proposed solution outperforms the traditional blockchain scheme on the basis of latency,bandwidth consumption,deployment time,scalability,and accuracy.Therefore it can be exploited to protect the edge-enabled Io T from privacy attacks and to ensure end-to-end Io T service orchestration.By designing the EISO framework and its key technologies of application development for customized services orchestration,security,and privacy,this dissertation provides a systematic solution for the endto-end Io T services orchestration by addressing the overall lifecycle of Io T services.