Internet of Things-Enabled Optimal Data Aggregation Approach for the Intelligent Surveillance Systems

The Internet of Things (IoT)-based intelligent surveillance systems in smart cities are a challenging issue as various devices capture the data. These devices, deployed close to the underlined phenomenon, such as cameras, are duplicated or redundant as accuracy is the main requirement of these systems. For this purpose, sensor nodes are deployed to provide 24/7 monitoring of a smart city, which minimizes the security risks and enables quick response in case of any disaster. However, due to a large number of devices, huge data are generated; thus, controlling traffic congestion in case of undesirable circumstances, for example, in case of accident or intention blockage of the road, is desperately needed. Numerous data aggregation mechanisms were reported in the literature to address this issue with smart city surveillance systems. However, these approaches were designed for either specific application environments or complex environments, making the implementation process hard. In this article, we have developed an Internet of Things-enabled optimal data aggregation approach, specifically designed for the intelligent surveillance systems in smart cities, to convert raw data values into the refined ones with minimum possible data loss ratio. Moreover, the proposed scheme bounds every server to perform or carry out the data refinement process to maintain the expected ratio of accuracy and precision. In this approach, ordinary devices are forced to capture and forward data in raw form, preferably without any or minimum possible processing. This reduces the load on ordinary devices in intelligent surveillance systems. Additionally, we have developed a novel approach to eliminate or reduce (if elimination is not possible) noisy data or outliers. It is implemented along with existing state-of-the-art techniques to verify the exceptional performance of the proposed data aggregation approach. These algorithms were compared using various performance evaluation metrics such as refinement ratio, data loss ratio, energy efficiency, and lifetime. The simulation results have verified that the proposed scheme's performance is better than the existing approaches.