An Overview of Outlier Detection Technique with Support Vector Machine Developed for Wireless Sensor Networks
-
Abstract
Wireless Sensor networks (WSNs) is an efficient and emerging area of Computer Science Engineering which has been currently employed in various fields of engineering particularly in communication system to make it effective and reliable. It is important to maintain the basic security level for different types of attacks like both external and internal for successful application of WSNs. Outliers in wireless sensor networks are measurements that deviate from the normal model of sensed data and result from errors, events or malicious attacks on the network. WSNs are more likely to generate outlier due to their special characteristics like constrained available with the resources causing frequent physical failure and harsh deployment area. The dynamic nature of sensor data and the specificity of the wireless sensor network make traditional outlier detection techniques unsuitable for direct application in such contexts so it is essential to select and adapt appropriate techniques to implement in wireless sensor networks for better sensing quality and more reliable system. This paper provides a comprehensive overview of existing outlier detection techniques specifically developed for the wireless sensor networks. Additionally, it presents a technique used to select data type, outlier type and outlier degree.We also investigate applicability of event detection technique for outlier detection. Through experimental study, we evaluate performance of our outlier detection technique to detect outliers and classify them as local or global based on real sensors.
Abstract
Wireless Sensor networks (WSNs) is an efficient and emerging area of Computer Science Engineering which has been currently employed in various fields of engineering particularly in communication system to make it effective and reliable. It is important to maintain the basic security level for different types of attacks like both external and internal for successful application of WSNs. Outliers in wireless sensor networks are measurements that deviate from the normal model of sensed data and result from errors, events or malicious attacks on the network. WSNs are more likely to generate outlier due to their special characteristics like constrained available with the resources causing frequent physical failure and harsh deployment area. The dynamic nature of sensor data and the specificity of the wireless sensor network make traditional outlier detection techniques unsuitable for direct application in such contexts so it is essential to select and adapt appropriate techniques to implement in wireless sensor networks for better sensing quality and more reliable system. This paper provides a comprehensive overview of existing outlier detection techniques specifically developed for the wireless sensor networks. Additionally, it presents a technique used to select data type, outlier type and outlier degree.We also investigate applicability of event detection technique for outlier detection. Through experimental study, we evaluate performance of our outlier detection technique to detect outliers and classify them as local or global based on real sensors.
Chapters in this book
- Frontmatter I
- Preface of the Editors V
- Advances in Systems, Signals and Devices VII
- Editorial Board Members VIII
- Advances in Systems, Signals and Devices XI
- Contents XIII
- Digital Inclinometer for Range of Motion Measurements in All Anatomical Planes 1
- Accurate Reduced-Order Modeling of MEMS and NEMS Microactuators under Dynamic Electrostatic Loading and Large Strokes 21
- Dynamic Performance of a Narrow Frequency Band Acoustic Microsensor 35
- A Novel Multi-Input Single-Output Mixed-Mode Universal Filter Employing Second Generation Current Conveyor Circuit 53
- High Performance Second Generation Current Conveyor Circuit and Multiplier Applications 65
- Schottky Barrier Carbon Nanotube Transistors Op-Amp Circuit 81
- Validation of a SaaS-based Platform for Mobile Health Applications 93
- GPRS-Based Remote Sensing and Teleoperation of a Mobile Robot 113
- A Hardware Implementation of Genetic Algorithms using FPGA Technology 129
- An Overview of Outlier Detection Technique with Support Vector Machine Developed for Wireless Sensor Networks 145
- Electronic Nose System and Principal Component Analysis Technique for Gases Identification 167
- Multisensor Data Fusion in Fatigue Detection Using Wearable Devices 181
- Application of a Trajectory Piecewise Linearization Approach on a Nonlinear MEMS Gyroscope 197
- Monolithic Integrated CMOS Ambient Light Sensor 217
Chapters in this book
- Frontmatter I
- Preface of the Editors V
- Advances in Systems, Signals and Devices VII
- Editorial Board Members VIII
- Advances in Systems, Signals and Devices XI
- Contents XIII
- Digital Inclinometer for Range of Motion Measurements in All Anatomical Planes 1
- Accurate Reduced-Order Modeling of MEMS and NEMS Microactuators under Dynamic Electrostatic Loading and Large Strokes 21
- Dynamic Performance of a Narrow Frequency Band Acoustic Microsensor 35
- A Novel Multi-Input Single-Output Mixed-Mode Universal Filter Employing Second Generation Current Conveyor Circuit 53
- High Performance Second Generation Current Conveyor Circuit and Multiplier Applications 65
- Schottky Barrier Carbon Nanotube Transistors Op-Amp Circuit 81
- Validation of a SaaS-based Platform for Mobile Health Applications 93
- GPRS-Based Remote Sensing and Teleoperation of a Mobile Robot 113
- A Hardware Implementation of Genetic Algorithms using FPGA Technology 129
- An Overview of Outlier Detection Technique with Support Vector Machine Developed for Wireless Sensor Networks 145
- Electronic Nose System and Principal Component Analysis Technique for Gases Identification 167
- Multisensor Data Fusion in Fatigue Detection Using Wearable Devices 181
- Application of a Trajectory Piecewise Linearization Approach on a Nonlinear MEMS Gyroscope 197
- Monolithic Integrated CMOS Ambient Light Sensor 217
