Integrating Edge Computing with Internet of Things Systems for Smart Homes

The fast development of the Internet of Things (IoT) has dramatically transformed dwelling space, leading to the development of smart homes smart systems of connected devices that can perceive, compute and communicate. They enable automation, increase energy efficiency, safety and security, and improve user comfort etc. In the recent years, smart home systems were built upon centralized cloud computing architecture for data processing and decision-making. Even though the cloud brings in the ability of scalability and the centralized management of resources; but it also imposes some important challenges, including high end-to-end delay, high consumption, and an increase in vulnerability to data exposure since confidential information is moved to the other servers. Edge computing has been proposed as a promising paradigm to mitigate these issues by distributing computation and allowing data to be processed at, or near, the data source. This paper study the deployment of edge computing support for IoT systems in smart homes and perform a comparative study of two system architectures, the traditional cloud-based and the proposed edge-enabled architecture. The performance is evaluated through extensive simulations in OMNET++ using real smart home scenarios and traffic patterns. System efficiency is evaluated considering key performance parameters— end-to-end latency, bandwidth usage, CPU utilization and packet delivery ratio. It is shown that the edge-enabled architecture significantly reduces the latency, efficiently utilize the network resources, balance the processing load, and enhance the reliability of the packet delivery compared to cloud-only model