Learn to Sense vs. Sense to Learn: A System Self-Integration Approach

Abstract

The diffusion of Internet of Things (IoT) devices has opened up new opportunities for decentralized data analytics. In this context, data transmission can be affected by both network issues and distance between devices and receivers. These factors can affect the ability to aggregate and analyze data from multiple IoT devices, resulting in noisy, partial, or incorrect information. To this end, self-healing techniques pursue corrective actions when information acquired from sensors is not reliable. In this paper, we propose a new self-integration approach to improve the performance of decentralized self-healing techniques.