Programmable embedded systems are operated manually via wireless sensor networks (WSN), eventually known as the internet of things or IoT. IoT devices are becoming suitable supplementary of human knowledge through the training to recognize several particular ambiances using the compelling message queuing telemetry transport (MQTT) protocol. These WSN models fundamentally use attribute-based data sets, and also need local unwanted data removal, continuous tracking, monitoring, timely response, and efficient storage management. This article proposes a quality of experience-aware IoT replica. It primarily filters collected sensor data to eliminate unnecessary data traffic due to huge packet conciliation into the WSN and succeeding IoT networks. Consequently, it reduces completion time and storage complexity at perception and network levels to balance the lightweight sensor network. The recommendation module at the subscriber’s end is proficient in choosing the best alternative for each meticulous user and can suggest the best packages per user requirement. A comprehensive study with the prospect of the proposed mathematical model through suitable experiments shows that the proposed model accomplishes 94.86% correct outcome at the time of the recommendation.
|Journal||Journal of Ambient Intelligence and Humanized Computing|
|Publication status||E-pub ahead of print - 17 Jan 2022|