Incentive Based Approach to Find Selfish Nodes in Mobile P2P Networks

Authors: 
Hemanth Meka, Sanjay Kumar Madria, Mark Linderman
Keywords: 
http://ieeexplore.ieee.org/xpl/articleDetails.jsp?arnumber=6407777
Abstract: 

In a Mobile P2P (M-P2P) network, selfish nodes can drop packets to affect the efficiency of the whole network. Therefore, we propose a simple and efficient mechanism using virtual currency to identify selfish nodes in the network. Each node issues a receipt to its broker to prove that it has provided a forwarding service. Based on receipts received each broker provides well-behaved nodes some virtual currency whereas selfish nodes are punished on dropping packets to make the network more efficient. When compared to other known approaches, our scheme provides real-time detection of selfish nodes and does not require/assume the presence of any tamper proof hardware at each node and can work with any routing algorithm. Simulation results show that our scheme is efficient in terms of (i) time to detect selfish nodes, (ii) the number of false positives and (iii) the number of packets exchanged when compared with others.

Publish Date: 
Monday, December 3, 2012
Venue: 
Performance Computing and Communications Conference (IPCCC), 2012 IEEE 31st International, Austin, TX