Multimodal Device Clustering Using Mobile Agent for Correlation in Sensor-Based IoT
Ankita Shukla, Vishal Krishna Singh and Mala Kalra
Prolonging the overall network lifetime is an important challenge and clustering proves to be a key solution towards it. The topology of the network such as sensor-based IoT is affected by the heterogeneity of devices which results in unceasingly changing the correlation structure of the network data. To address the challenges, a clustering algorithm using Mobile Agent for correlation is proposed to recognize and classify sensor nodes into optimal groups to analyze them based on data generated. In this paper, firstly a model for correlation is designed to transmute the features of data sensed into a defined subspace and further analyzing the correlation. Finally, a novel Mobile Agent-oriented Device Clustering Algorithm (MAODCA) in sensor-based Internet of Things is proposed. The performance of the proposed algorithm is compared with Enhanced Low Energy Adaptive Clustering Hierarchy (E-LEACH), Power-Efficient Gathering in Sensor Information System Extended (PEGASIS-E), and Balanced Energy Efficient Multi-hop for mobile Wireless Sensor Networks (BEEM-M) algorithms. Substantial simulation outcomes show that the presented algorithm outperforms the existing algorithms in terms of the number of dead nodes by 4.3%, 8.76%, and 3.8% respectively, and also improves the data transmission quality by 35.8%.
Keywords: Cluster, Data Correlation, Mobile agent (MA), Multimodal, Sensor-based Internet of Things (SBIoT)