基于融合深度学习的物联网传感器应用雾大数据分析

Fog big data analysis of IoT sensor application based on fusion deep learning

物联网传感器的应用数量急剧增长,产生了大量数据,需要非常有效的数据分析程序。然而,不同的物联网基础设施和物联网传感器设备在发送和接收消息方面存在协议限制,给智能物联网传感器应用的开发带来了障碍,使得现有物联网传感器无法适应其他物联网传感器应用。文中研究和分析了物联网传感器如何为大数据分析生成数据,并强调了智能解决方案的现有挑战。物联网传感器应用需要在雾计算环境中进行大数据分类和分析。提出的雾大数据分析模型和基于反向传播神经网络分析模型相结合的融合模型为潜在的机器通信实践带来了新的挑战。将提出的模型应用于智能城市数据集,取得了较好结果,并在不同的物联网应用环境中比较了不同的深度学习和机器学习算法的性能。

The number of applications of IoT sensors is growing rapidly,generating large amounts of data and requiring very effective data analysis programs.However,different IoT infrastructures and IoT sensor devices have protocol limitations in sending and receiving messages,which poses obstacles to the development of smart IoT sensor applications,making existing iot sensors unable to adapt to other IoT sensor applications.This paper studies and analyzes how IoT sensors generate data for big data analytics and highlights the existing challenges of intelligent solutions.IoT sensor applications require big data classification and analysis in fog computing environments.The proposed fog big data analysis model and the back propagation neural network analysis model of IoT sensor application based on fusion model bring new challenges to the potential machine communication practice.Improved results were achieved by applying the proposed model to the most important fog applications developed on smart city datasets,and the performance of different deep learning and machine learning algorithms was compared in different IoT application environments.