An Overview of Deep Learning Methods in the Internet of Things Technology in regular life

Abstract

Large volumes of data are generated daily as a result of the extensive usage of Internet of Things (IoT) technology in indoor daily life. dependable methods for data analysis are necessary to make efficient use of this data. Recent advances in deep learning (DL) make it easier to handle and learn from enormous amounts of IoT data, allowing for a quick and competent understanding of the fundamentals of many IoT applications in intelligent indoor settings. The current literature on the usage of DL for various indoor IoT applications is summarized in this paper. Our objective is to provide knowledge on how to apply deep learning techniques from many angles to create better two separate indoor IoT application areas: indoor localization/tracking and activity detection. One important objective is to seamlessly combine the two fields of IoT and deep learning, which will lead to a variety of creative approaches for indoor IoT applications including robots, smart home automation, health monitoring, etc. Additionally, from a comparison of technical research in the three aforementioned categories, we develop a thematic classification. To increase the effectiveness of indoor IoT applications and to encourage and inspire further advancement in this exciting field of study, we conclude by proposing and discussing various problems, difficulties, and new directions to apply deep learning.