ABSTRACT
Named Data Networking performance depends on In-network caching. The effective cache management in NDN involves cache placement, replacement and eviction policies. This paper presents optimized NDN caching system using game theory. Many approaches based on game theory utilizes bandwidth, time and power as resources, but our work pricing and profit model based on coverage and gain derivations. NDN routers are modeled as Poisson point processes using stochastic geometry. Then with derived probability, caching scheme develops profit model and formulates profit gained by ISP and producers from caching in intermediate nodes. For jointly maximizing average profit of both ISP and producers, a Stakelberg game is proposed. Using this game theoretic model, we evaluate uniform pricing scheme which reduces backbone access cost and maximizes sum profit gained by ISP and producer. It is observed that the optimal solution directly relates to available storage size in each NDN router and popularity distribution of requested content by consumer.
