Radiation source detection from mobile sensor networks using principal component analysis

Zhao, Jifu

Permalink

https://hdl.handle.net/2142/92861 Copy

Description

Title

Radiation source detection from mobile sensor networks using principal component analysis

Author(s)

Zhao, Jifu

Issue Date

2016-07-18

Director of Research (if dissertation) or Advisor (if thesis)

Sullivan, Clair J.

Committee Member(s)

Mohaghegh, Zahra

Department of Study

Nuclear, Plasma, & Rad Engr

Discipline

Nuclear, Plasma, Radiolgc Engr

Degree Granting Institution

University of Illinois at Urbana-Champaign

Degree Name

M.S.

Degree Level

Thesis

Keyword(s)

• Radiation detection
• mobile sensor networks
• principal component analysis

Abstract

Detecting the presence of possible illicit radioactive materials in large areas is challenging because of changing background radiation, shielding effects and short collection time, especially when the radioactive materials are moving. The concept of mobile sensor networks is put forward to solve this problem. In this thesis, a small mobile sensor network is established using commercially available radiation detectors and cell phones. A spectrum decomposition and reconstruction method based on Principal Component Analysis (PCA) is proposed to work with mobile sensor networks. Two experiments are designed to test this method's performance on real-world data. The PCA-based method's performance is analyzed using receiver operating characteristic, or ROC curves. Further study finds that although the PCA-based method doesn't work well on current mobile sensor networks, its performance can be improved by increasing the radiation spectral quality.

Graduation Semester

2016-08

Type of Resource

text

Permalink

http://hdl.handle.net/2142/92861

Copyright and License Information

Copyright 2016 Jifu Zhao

Owning Collections