Review on data mining techniques in wireless sensor networks

Abstract

Over the past few years, wireless sensor networks (WSN) have emerged as one of the most exciting fields in Computer Science research. A Wireless Sensor Network (WSN) is a set of sensors that are integrated with a physical environment. These sensors are small in size, and capable of sensing physical phenomenon and processing them. The main aim of deploying applications based on WSNs is to make use of the data sensed by the sensors to raise the real-time decisions. The main limitations of WSNs are characteristics of sensor nodes and nature of sensor data generated by networks. Due to these limitations, traditional data mining techniques are not suitable to WSNs. As the data generated by WSNs is highly resource-constrained, huge in volume, fast changing, it is very challenging to design suitable data mining techniques for WSNs. This inspires to explore a novel and appropriate data mining technique capable of extracting knowledge from huge volume and variety of continuously arriving data from WSNs. In this paper different existing data mining techniques adopted for WSNs are examined with different classification, evaluation approaches, Finally, some research challenges related to adopting data mining techniques in WSNs are also pointed out.

Decentralized Computation of Homology in Wireless Sensor Networks Using Spanning Trees

“…We consider two types of simplicial complexes over V . One is the Čech complexČ( (1). It has been shown that the Čech complex fully captures the homology of its underlying geometric shape [29].…”

“…This means that every hole in the coverage of network domain corresponds to a hole inČ(V ). Unfortunately, the Čech complex is not computable by our WSN since geometric information is needed, as evident from (1). The second simplicial complex we consider is the Rips complex…”

“…Wireless sensor networks (WSN) have in recent years become well-established ad-hoc networks with numerous applications in environmental and health care monitoring. Distributed nature of WSNs, random node deployment and frequent topology changes challenge the researchers to design special data mining techniques, suitable for WSNs [1]. Small sensing nodes, scattered over an area of interest, typically sample the domain in a point cloud fashion, making possible the use of computational geometry and algebraic topology to extract knowledge from the collected data [2].…”

Data Mining Method under Model-Driven Architecture (MDA)

“…For the problems of data analysis and data visualization, this paper comprehensively studies data mining [6] and data visualization technology [7,8]. It proposes the analysis and visualization framework for college information data based on MDA [9] ideas.…”