An Improved Dendritic Cell Algorithm Based Intrusion Detection System for Wireless Sensor Networks

Base on the similarities between WSNs intrusion detection and artificial immune system. This paper utilizes the benefits of one of the Danger Theory based AIS intrusion detection algorithms, namely the Dendritic Cell Algorithm (DCA) to design a distributed and hierarchical intrusion detection model for WSN. Beside, because of the shortcomings of DCA about its Dendritic Cell (DC) evaluation mechanism and lymph decision mechanism, the false alarm rate is high. Therefore, this paper puts forward an improved Dendritic Cell Algorithm reference to the idea of data fusion theory. In this algorithm, we firstly define a scoring function to evaluate the DC context in DC part. And then in lymph part, we fuse multiple DCs’ evaluation through Dempster rule to make global decision. The experimental results show that the improved DCA based IDS proposed can effectively improve the accuracy rate without depending on the MCAV threshold and show advantages in flexibility and adaptability.