A Correlation based Approach to Differentiate between an Event and Noise in Internet of Things
Journal Title: International Journal of Advanced Computer Science & Applications - Year 2018, Vol 9, Issue 12
Abstract
Internet of Things (IoT) is considered a huge enhancement in the field of information technology. IoT is the integration of physical devices which are embedded with electronics, software, sensors, and connectivity that allow them to interact and exchange data. IoT is still in its beginning so it faces a lot of obstacles ranging from data management to security concerns. Regarding data management, sensors generate huge amounts of data that need to be handled efficiently to have successful employment of IoT applications. Detection of data anomalies is a great challenge that faces the IoT environment because, the notion of anomaly in IoT is domain dependent. Also, the IoT environment is susceptible to a high noise rate. Actually, there are two main sources of anomalies, namely: an event and noise. An event refers to a certain incident which occurred at a specific time, whereas noise denotes an error. Both event and noise are considered anomalies as they deviate from the remaining data points, but actually they have two different interpretations. To the best of our knowledge, no research exists addressing the question of how to differentiate between an event and noise in IoT. As a result, in this paper, an algorithm is proposed to differentiate between an event and noise in the IoT environment. At first, anomalies are detected using exponential moving average technique, then the proposed algorithm is applied to differentiate between an event and noise. The algorithm uses the sensors’ values and correlation existence between sensors to detect whether the anomaly is an event or noise. Moreover, the algorithm was applied on a real traffic dataset of size 5000 records to evaluate its effectiveness and the experiments showed promising results.
Authors and Affiliations
Dina ElMenshawy, Waleed Helmy
Keywords
Related Articles
- EP ID EP429120
- DOI 10.14569/IJACSA.2018.091212
- Views 70
- Downloads 0
Training Difficulties in Deductive Methods of Verification and Synthesis of Program
The article analyzes the difficulties which Bachelor Degree in Informatics and Computer Sciences students encounter in the process of being trained in applying deductive methods of verification and synthesis of procedura...
Convolutional Neural Network Hyper-Parameters Optimization based on Genetic Algorithms
In machine learning for computer vision based applications, Convolutional Neural Network (CNN) is the most widely used technique for image classification. Despite these deep neural networks efficiency, choosing their opt...
A New Particle Swarm Optimization Based Stock Market Prediction Technique
Over the last years, the average person's interest in the stock market has grown dramatically. This demand has doubled with the advancement of technology that has opened in the International stock market, so that nowaday...
Instruction Design Model for Self-Paced ICT System E-Learning in an Organization
Adopting an Information Communication and Technology (ICT) system in an organization is somewhat challenging. User diversity, heavy workload, and different skill gap make the ICT adoption process slower. This research st...
Polarimetric SAR Image Classification with High Frequency Component Derived from Wavelet Multi Resolution Analysis: MRA
A method for polarimetric Synthetic Aperture Radar: SAR image classification with high frequency component derived from wavelet Multi-Resolution Analysis: MRA is proposed. Although it is well known that polarization sign...