Affiliations: [a] Department of Computer Science and Engineering, Panimalar Engineering College, Tamilnadu, India
| [b] Department of Artificial Intelligent, Adi Shankara Institute of Engineering and Technology, Kerala, India
Correspondence:
[*]
Corresponding author. N. Pughazendi, Department of Computer Science and Engineering, Panimalar Engineering College, Tamilnadu, India. E-mail: pughazendi@gmail.com.
Abstract: Internet of Things (IoT) devices installed in hospital direct data unceasingly; in this manner, energy usage augments with the number of broadcasts too. In this paper, Reliable Cluster based Data Collection Framework (RCDCF) for IoT-Big Data Healthcare Applications (HA) is developed. During clustering process, the connected IoT devices are grouped into clusters. In clustering technique, the available IoT devices are gathered into groups. The device with high battery capacity and processing ability is selected as a cluster head (CH). Each member of the cluster is allocated multiple slots by applying a general function pooled by the Fog node and the entire devices. To perceive and eliminate outliers from the sensor data, Density-based spatial clustering of applications with noise (DBSCAN) method is utilized. To forecast the objective and subjective behaviours of the equipments, a Random Forest Deep Neural Network (RF-DNN) based classification model is utilized. By experimental results, it has been shown that RCDCF achieves 19% and 20% reduced energy consumption at Cloud and Fog centers, respectively. Moreover, RCDCF has 2.1% and 1.3% increased correctness of data at Cloud and Fog data centers, respectively, when compared to the existing framework.
Keywords: Internet of Things (IoT), big data, cloud, clustering, health care solution, slot allocation, Random Forest Deep Neural Network (RF-DNN), categorization
