IoT-enabled Knowledge Extraction and Edge Device Sustainability in Smart Cities

Internet of Things (IoT) deployments are becoming the backbone of all future Smart City (SC) environments. They can, therefore, act as massive crowd-sourced data aggregators, driven by device-to-device interactions with SC users' mobile devices and their wireless interfaces. Provided that, our research focuses on developing probabilistic and machine learning models to (a) enable knowledge discovery from passive user interactions with the wireless IoT infrastructure and (b) apply the collected intelligence to increase the energy-efficiency and resiliency of the Smart City's IoT network. In this extended abstract we elaborate on the motivation behind our work, and the related challenges, while pointing to the solutions developed so far.

I. Introduction and Motivation

Until now the realization of the “Smart City” concept by urban planners and city governments has been relying on plain digitalization and automated implementations that often overlook the whole equation's human aspect. The inclusion of those who work and reside in the Smart City (SC) environment into the design process is becoming highly significant especially since the current urban population represents 55% of the world's population, a number that is expected to reach 68% by 2050. In addition, more and more municipalities and governments actively participate to discuss and address challenges with EU's Smart City Initiative and UN's United Smart City Initiative being two cases in point.