Home // UBICOMM 2015, The Ninth International Conference on Mobile Ubiquitous Computing, Systems, Services and Technologies // View article

Collaborative Detection with Uncertain Signal Distributions in Wireless Sensor Networks

Authors:
Tai-Lin Chin
Jiun-Hao Chen
Cheng-Chia Huang

Keywords: Sensor networks, Target Detection, Data Fusion, Constant False Alarm Rate

Abstract:
Sensor networks are envisioned to have the capability to detect the presence of an event or target in a monitored region. Sensors can collect measurements about the target and make local decisions about the presence or absence of the target. To reduce probability of false alarms, collaborative detection is usually exploited, where the local decisions are fused to arrive at a consensus about the target presence. In general, the performance of a sensor network can be evaluated in terms of detection probability and false alarm probability. This paper adopts the Constant False Alarm Rate (CFAR) detector for sensors to make local decisions. The distributions of the target signal and noise are assumed unknown a priori. Simple and effective methods are proposed to estimate the distributions of sensor measurements. The AND and OR fusion methods are exploited in making the final decisions. Simulations are conducted to verify the analytic results to the simulated results. The best selection of sensors to participate the fusion in order to protect a particular location in the monitored region is also shown by experiments. Essentially, the paper analyzes the approximated detection probability and false alarm probability based on the estimated distributions of the unknown target signal and noise. Through simulations, it is shown that those approximated results could be close to the true values.

Pages: 114 to 119

Copyright: Copyright (c) IARIA, 2015

Publication date: July 19, 2015

Published in: conference

ISSN: 2308-4278

ISBN: 978-1-61208-418-3

Location: Nice, France

Dates: from July 19, 2015 to July 24, 2015