Adaptive service function for system reward maximization under elastic traffic model

We propose an adaptive service model that maximizes the amount of service that spectrum users (SUs) achieve from accessing DSA systems. The proposed model allows SUs to utilize available spectrum efficiently by enabling them to locate spectrum opportunities in a distributed manner, thereby maximizing the long-term rewards that SUs receive. In this model, SUs adapt their required level of service with time depending on the amount of service they received so far. This proposed model is suitable for and can be used by existing objective functions. It leads to the maximization of the amount of service that SUs receive through DSA. Our simulation results show that the proposed adaptive model is very scalable by performing well regardless of the number of users in the system, and allows users to achieve high service rewards by quickly locating spectrum opportunities in the system.