Design And Implementation Of A Distributed Io T Platform For Remotely Developing And Experimenting IoT Applications Based On Heterogeneous Devices

In recent years,the Internet of Things(IoT)has developed rapidly and been widely used in people's daily life.The development of IoT brings two problems.First,with the emergence of new devices and operating systems,the heterogeneity of IoT devices is becoming stronger and stronger.Second,IoT applications are becoming more and more complex.Most IoT applications contain multiple heterogeneous devices.In the traditional IoT application developing and testing process,developers need to select and purchase IoT devices.And then,to develop IoT applications,developers have to install the corresponding drivers and Integrated Development Environment(IDE)for specific IoT devices.Once the applications are developed,developers need to deploy devices to actual environment and test the performance of applications.Based on the experimental result of applications,developers can improve the performance of applications.In order to overcome the shortcomings of existing systems,we design and implement a distributed remote IoT developing and testing platform called LinkLab.The LinkLab platform adopts a distributed architecture that supports access to a variety of heterogeneous IoT devices,providing developers with cloud compilation,online burning,and application testing.Based on the LinkLab platform,developers do not need to configure the development environment locally.Code writing,application compilation,device burning,and application testing are all done in the cloud,simplifying the developing and testing process of IoT applications.The contributions of this thesis are summarized as follows.(1)This thesis proposes a distributed,remote IoT platform architecture that supports heterogeneous IoT devices.Based on this architecture,deployment and expansion of IoT devices can be easily realized.At present,seven types of IoT devices have been supported by LinkLab.(2)This thesis proposes an online compilation method for IoT applications based on multi-user cache.Experiments show that the average compilation time can be reduced by 68.9% based on this method.(3)This thesis presents a fine-grained device allocation algorithm.Based on this algorithm,the influence of wireless signals among tasks can be reduced.Simulation results show that the algorithm can reduce the job waiting time by 64.0%.(4)This thesis implements and deploys the LinkLab system.Furthermore,the LinkLab platform is used in the IoT course.The result shows that the average time required for students to complete the developing and testing of an IoT application is only 1.05 hours based on the LinkLab platform.Based on the above contributions,this thesis implements the LinkLab platform.Currently,the compilation of 4 types of IoT applications is supported by LinkLab.166 IoT devices are connected to LinkLab using 18 Raspberry Pi and one desktop computer.The experimental results show that the platform supports multiple heterogeneous devices,provides remote developing and testing functions for IoT developers,reduces the difficulty of IoT application development.