A Survey on Data Quality Analysis in Wireless Sensor Networks

PhD Qualifying Examination


Title: "A Survey on Data Quality Analysis in Wireless Sensor Networks"

by

Mr. Xin MIAO


Abstract:

Wireless sensor networks (WSNs) have been widely used in many fields such 
as environmental surveillance, emergency navigation, traffic monitoring, 
and industrial control. Data collected from WSNs, however, often contain 
incomplete, inaccurate, incorrect or inconsistent parts. Users may draw 
false conclusions from data and then make wrong decisions, which can 
severely hamper the usage of WSNs. Therefore, how to clean and repair 
sensor data is becoming an important task in recent years.

This survey provides an overview of state-of-art techniques to analyze and 
improve the quality of sensor data. In general, existing approaches can be 
divided into two categories: anomaly detection and data repairing. In the 
first category, abnormal data (a.k.a., outliers) are detected using 
statistical models, while in the second category, contradictions in the 
data are detected and fixed with editing rules or master data.

Among all these techniques, no one is a clear favorite since they address 
the problem from different aspects. This survey elaborates these 
approaches in depth and also compares their design tradeoffs, advantages 
and disadvantages. Moreover, the unsolved issues and future research 
directions in this open area are also discussed.


Date:                   Tuesday, 11 January 2011

Time:                   10:00am - 12:00noon

Venue:                  Room 3501
                         lifts 25/26

Committee Members:	Dr. Yunhao Liu (Supervisor)
 			Prof. Dimitris Papadias (Supervisor)
                         Dr. Lei Chen (Chairperson)
 			Dr. Lin Gu
 			Prof. Lionel Ni


**** ALL are Welcome ****