17.5 CASE STUDY II: GN FOR THREAT DETECTION IN WSN

This case study is based on the work done by Baig et al. [6] for developing an online security scheme for the WSN. In this study, the simple GN algorithm for pattern-recognition has been used to detect the DDoS attack in WSNs. In this scheme, each network node also acts as the GN node within the WSN. The scheme shows that the GN algorithm provides an energy-efficient mechanism for attack pattern detection within the WSN.

17.5.1 The GN Approach for WSN

In this approach, the GN enables the network to process its internal traffic flow patterns in real-time through a process similar to human introspection using very little energy. The GN algorithm is implemented using an energy-efficient scheme, which uses partial updating of patterns within the network for energy conservation. The threat detection scheme is implemented in five stages and is designed for DDoS, which commonly occur within the WSNs:

  1. Initialization: At this stage, the GN nodes (i.e., the wireless nodes) would be initialized with the DDoS attack patterns. Therefore, all the GN nodes will store the attack patterns for later detection.
  2. Observation: All GN nodes would continuously monitor for attack pattern using the simple GN algorithm described in Section 17.1. The monitoring involves observing the threshold values obtained from the network neighborhood [6].
  3. Communication: Each GN node would communicate with its adjacent nodes by exchanging its finding with the others. ...

Get Mobile Intelligence now with the O’Reilly learning platform.

O’Reilly members experience books, live events, courses curated by job role, and more from O’Reilly and nearly 200 top publishers.