A Machine Learning and Fuzzy-Based Reliable Data Collection and Communication in AioT—Fog Computing Environment

The IoT-based sensor devices that are used to supply information for the proper operation of life in the modern world are currently in great demand. These IoT sensor devices transmit information to the cloud, where it is processed and used. When combined with the Internet of Things, cloud computing poses few risks (IoT). The vast amount of data that the cloud needs to analyze will take more time to process, which will have a negative influence on the quality of service (QoS), the performance of the IoT apps, and the overall network. Fog computing is used to overcome these risks. Since different IoT sensors have diverse functionalities and the majority of them face issues like latency, security, performance, etc., these devices are becoming less effective in real-time. The sending of IoT device's sensor data to the cloud and cloud data retrieval are two potential causes of the issues. While sending or receiving sensor data, there may be network traffic, network disconnections, or cloud location issues are some reasons. The solution to these challenges lies in fog computing. Data analysis at the fog node (FN) used to transferring it to a cloud server is the main goal of fog computing. This data is generated by IoT devices. The risks associated with cloud computing can be lessened by keeping data in fog nodes. IoT to increase productivity, effectiveness, latency, and highly secure services, fog computing could be a better option. It also aids in achieving high caliber and a quick reaction time. It makes it possible for data to be processed by Internet of Things (IoT) devices and sent to the fog node, from where it can be connected to the cloud data center. Here, we make an effort to provide a choice of computing models, features of fog computing, and a reference fog architecture with many levels, Utilizing descriptive and inferential analysis techniques to examine data from various sensor devices to derive insightful determination and a comprehensive review of fog with IoT, ML algorithms used in fog, challenges in fog, and applications of fog computing in the physical domain.