Cache management in mobile ad hoc networks

Adversary compromises the sensor in sensor networks deployed in a hostile region, and create many replicas of the captured node using multiple identities to perform malicious activities creating node replication attack. It is difficult to differentiate the genuine node from the malicious node.


Sequential Probability Ratio Test (SPRT)

Sequential Probability Ratio Test (SPRT), a mobile replica detection scheme works on the basis of the speed of the node following the rule of no single node should move at the speed greater than system configuration speed. Replica attack can be easily detected with the location of the node as two nodes with same ID located in different location shows like the mobility of the node but with speed greater than configured speed.

Drawbacks in SPRT

In SPRT, the detection of replica accuracy could significantly drop because the genuine node (the replica) may be falsely regarded as the replica (genuine node). In SPRT, localization and time synchronization errors affect the detection accuracy.

Localized algorithms

Localized algorithms overcome the drawback in SPRT using two algorithms, namely, extremely Efficient Detection (XED) and Efficient Distributed Detection (EDD).

Solution in NS2

  • Localized algorithm and SPRT is implemented in ns2 to detect the replicas from the genuine nodes.

  • Performance evaluation is carried out to calculate detection accuracy, detection time, storage overhead, computation overhead, and communication overheads in the network.

Related Titles:

  • J.Ho,M.Wright, and S. K. Das, “Fast detection of replica node attacks in mobile sensor networks using sequential analysis,” in Proc. IEEE Int. Conf. Computer Communications (INFOCOM), Brazil, 2009, pp. 1773–1781.

  • “Localized Algorithms for Detection of Node Replication Attacks in Mobile Sensor Networks” IEEE TRANSACTIONS ON INFORMATION FORENSICS AND SECURITY, VOL. 8, NO. 5, MAY 2013”.