| Summary: | 碩士 === 銘傳大學 === 資訊管理學系碩士班 === 99 === A sensor network is composed of a lot of tiny sensors, where each sensor can sense the environmental information such as temperature, humidity, pressure, etc. The collected information is forwarded through a specific routing architecture to the base station for further analysis. Applications of sensor networks include environmental detection, military operations, e-commerce etc. However, security mechanisms must be established first in the network to defend against malicious attacks when using in a hostile environment. In this paper, we are concerned with new security mechanisms for a serious attack called the black hole.
Earlier, Ngai et al. have proposed a security mechanism for the black hole attack, based on analyzing the statistic data collected from sensors. That is, a region of the network will return fewer packets to the base station when it contains a black hole, and the base station may detect the black hole from that data anomaly. However, we point that their mechanism has three drawbacks: 1) The data anomaly observed from the statistical data may be caused by the random distribution of sensors but not any black hole. 2) The data anomaly may be caused by the quick change of the sensing environment. 3) The mechanism does not offer any real-time response upon detecting a black hole. As a result, we propose a new mechanism to overcome the above weaknesses by employing special type of sensors called the mobile node. Then, we observe that certain environments may not support mobile nodes, so we propose other mechanisms to defend against the black hole black by using the multiple-path routing architecture. Finally, we make a comparison between our mechanisms and mechanisms proposed earlier.
|
|---|
