A machine learning-assisted data aggregation and offloading system for cloud–IoT communication

Abstract

Data aggregation and dissemination in cloud-based internet of things (IoT) are main issues because of interoperability problems in communication. In an IoT environment, data handling and offloading are constant processes that avoid communication failure and increase service utilization levels. This paper introduces a machine learning (ML)-assisted data aggregation and offloading (ML-DAO) system to improve the reliability of cloud–IoT communication. The method introduced helps reduce the response time and routing cost errors in data aggregation and improve the data service rate. The data handling rate is also enhanced using the IoT assisted by fog elements that maximize edge-level communication. Cloud–IoT communication quality is measured on the basis of time and service attributes; ML techniques are designed to enhance the level’s precision while aggregating the data. To achieve optimum communication quality, the proposed ML-DAO operates on certain measurable functional metrics. The performance of the system is assessed using the following metrics: route cost error, processing time, aggregation delay, service utilization rate, failure probability, and response time. Experimental results prove the consistency of the proposed scheme, as the metrics are optimized with lesser unallocated data chunks.