Big Data Using Efficient Expectation–Maximization Algorithm in Wireless Sensor Networks

As of late, the big data rose as a hotly debated issue because of the tremendous development of the
information and communication technology. One of the profoundly anticipated key contributors of the big
data later on networks is the distributed wireless sensor networks (WSNs). In spite of the fact that the data
created by an individual sensor may not seem, by all accounts, to be significant, the general data created
across numerous sensors in the densely distributed WSNs can deliver a significant part of the big data.
Energy-efficient big data gathering in the densely distributed sensor networks is, consequently, a
challenging research area. A standout amongst the most effective solutions to address this test is to use the
sink node's mobility to encourage the data gathering. While this technique can diminish energy
consumption of the sensor nodes, the use of versatile sink presents extra challenges such as deciding the
sink node's trajectory and cluster arrangement preceding data gathering. In this paper, we propose another
versatile sink directing and data gathering strategy through system clustering based on altered desire
expansion technique. Also, we determine an ideal number of clusters to limit the energy consumption. The
effectiveness of our proposal is confirmed through numerical results.