Shielding Wireless Sensor Network Using Markovian IDS

• Main Authors: Kuen-Tzung Chen, 陳坤宗
• Other Authors: 廖宜恩
• Format: Others
• Language: zh-TW
• Published: 2009
• Online Access: http://ndltd.ncl.edu.tw/handle/48928649467113044110

碩士 === 國立中興大學 === 資訊科學與工程學系所 === 97 === Wireless sensor node is congenitally limited by its insufficient resources of hardware such as memory size and battery duration; these factors not only affect the lifespan of wireless sensor network but also impose great challenges on adding security mechanism on sensor nodes. However, protecting sensor nodes from malicious attacks becomes more and more important as the applications of wireless sensor networks increase rapidly. In this thesis, we propose a new intrusion detection system (IDS) called Markovian IDS for protecting sensor nodes from malicious attacks. Markovian IDS incorporates game theory with anomaly detection and misuse detection for finding the best defense strategy of sensor nodes, and it employs Markov Decision Process (MDP) to predict attack patterns and then take appropriate defense strategies. Experimental results show that the proposed Markovian IDS has better successful defense rate compared to game theory only or MDP only methods.