Design and Implementation of Internet of Things Monitoring System Based on Neural Network Algorithm and Cloud Edge Collaborative Architecture

Abstract

The Internet of Things (IoT) has the ability to connect things with sensing capabilities all over the world, and it mainly relies on internal base stations to process a large amount of data. In this paper, using server node resources in edge cloud environment, we study and compare the impact of loose coupling and scheduler support on the performance of IoT applications, and introduce three customized offload strategies to meet the different requirements of computingintensive and delay-sensitive IoT applications. Experiments show that computing resources have a greater impact on the performance of IoT applications than communication resources, but when the system is expanded to accommodate more IoT devices, communication bandwidth becomes the dominant resource and directly affects the overall performance of IoT applications. The research on the application of neural network algorithm based on cloud computing in the Internet of Things, firstly summarizes the Internet of Things and neural network algorithm, and then explores the specific application of cloud computing neural network algorithm in the Internet of Things.