Adaptive Multi-Sensor Grids with Dynamic Threshold Optimization for IoT-Enabled Smart City Development

Abstract

This study presents a flexible smart city model which we have developed using IoT for in-depth urban monitoring and control, which we achieve via many sensor integrations and dynamic decision making. We put forth an intelligent system which uses low-cost hardware and many sensors, which in turn enable services like smart lighting, waste management, traffic control and disaster alerts. We present our framework, which gives us better response and accuracy through the use of adaptive threshold values and hybrid edgeto-cloud processing, which we do instead of the traditional static models. Also, we report that we achieved very low power consumption (approx. 28mW per node) and very fast response times ~1.75 s) in our experiments. The put forth approach is a solution that is energy and resource-efficient as well as scalable to the issues which today's urban settings present. Also, it is designed to improve resiliency via the addition of system capacity, which in turn allows us to add more sensor platforms that in turn act as redundant units for fault tolerance.