An Efficient Clustering and Deep Learning Based Resource Scheduling for Edge Computing to Integrate Cloud-IoT

Internet of things (IoT) gains wide attention in every domain due to its potential ability to connect large number of device that generates a huge amount of data. IoT applications require high speed data transfer and less latency due to its less data storage and energy limitations. Data transfer provides space for new data and less latency reduces the computation complexities. To obtain such an IoT system, Cloud computing is incorporated with IoT applications. Cloud offers wide unlimited storage and processing services virtually that overcomes the issues in IoT data management process. Integration of cloud with IoT provides better data storage and processing facilities, however, the centralized computation, and networking introduces delay due to long distance data transfer from IoT devices to cloud data centers. To overcome this issue, this research work introduces a concept of edge computing which processes the data in the edge devices and transfers it to the cloud which reduces the latency and increases the efficiency of the system. To achieve this hybrid data clustering and deep learning based resource scheduling are introduced in the proposed work to reduce the computational complexities. The performance of the proposed integration approach is evaluated in terms of latency, efficiency, computation time and compared with conventional clustering approaches and cloud IoT systems without edge computing services. Experimental results validate that the proposed approach exhibits less latency and improved efficiency than the traditional cloud IoT system.