RECOD: reliable detection protocol for large-scale and dynamic continuous objects in wireless sensor networks |
| |
Authors: | Yim Yongbin Park Soochang Lee Euisin Nam Ki-Dong Kim Cheonyong Kim Sang-Ha |
| |
Affiliation: | 1.Department of Computer Engineering, Chungnam National University, Daejeon, Korea ;2.Department of Computer Engineering, Chungbuk National University, Cheongju, Korea ;3.School of Information and Communication Engineering, Chungbuk National University, Cheongju, Korea ;4.Electronics and Telecommunications Research Institute (ETRI), Daejeon, Korea ; |
| |
Abstract: | Whether individual objects such as enemy tanks or intruders have been reliably detected typically depends on the number of data reports successfully delivered to a sink node from the sensor nodes surrounding the object. When the number of data reports exceeds a required threshold, the sink recognizes the object that is detected by sensor nodes. Thus, previous studies exploited this framework for reliable detection as event reliability for individual objects, and proposed event-to-sink reliable-transport mechanisms that can reach a required threshold. Recently, in wireless sensor networks, research has focused on coverage detection for large-scale phenomena such as biochemical material and wild fires. Such phenomena are known as continuous objects because they generally cover wide areas and frequently change their shape as a result of wind or geographical features. Since continuous objects are large-scale and alterable, they present new challenges for the event reliability. In this paper, we first define new criteria for measuring the event reliability of large-scale phenomena. Then, we propose a novel event-to-sink transport protocol that is reliable, even when excessive data is generated from many sensor nodes detecting these phenomena. Analysis and simulation results demonstrate the event reliability of our protocol. |
| |
Keywords: | |
本文献已被 SpringerLink 等数据库收录! |
|