Competition Analysis of Adaptive Incentive Protocols for P2P Networks

Incentive protocols are crucial for enhancing contribution and cooperation among nodes in P2P networks. In this paper, we consider an analytical framework wherein peers learn the gains of different strategies and evolve adaptively towards the better performance strategy. While a unitary incentive policy in the system may exhibit its incentive capability on itself, the competition among different incentive policies are still need to be investigated when they are co-existing optional strategies for peers. To illustrate the expected performance gain and system robustness under the incentive protocol competition, we study three incentive policies and two learning (or adaptive) models and analyze their performance under different conditions. We also provide simulation results to validate our analysis and present robustness characteristics of incentive policies under competition.