Quality of Service Enhancement for IoT-Based Smart Office System Using Ad Hoc On-Demand Distance Vector-Smart Control Ration Control Algorithm (AODV-SRCA)

The performance of an application in IoT settings heavily relies on the quality of service (QoS) parameters, which play a crucial role. Due to their diverse nature, attaining optimal QoS parameters presents a significant challenge. This study presents an innovative approach to enhance QoS parameters in IoT by leveraging machine learning (ML) and Blockchain methodologies. A simulation of the IoT network integrated with Blockchain technology is conducted using an NS2 simulator to improve QoS. Various QoS parameters, including delay, throughput, packet delivery ratio, and packet drop, are thoroughly examined. Subsequently, the QoS data is analysed using different ML algorithms such as Naive Bayes (NB), Decision Tree (DT), and Ensemble learning techniques. The findings indicate that the Ensemble classifier achieves the highest classification accuracy of 83.74% compared to NB and DT classifiers.