A dynamic pricing model for data services in GPRS networks

GPRS data services are subject to severe performance degradation due to limited network resources. The technical approaches taken to alleviate the problem are not sufficient to encounter a complete range of incoming scenarios. In this paper, we define and study analytically a performance model for GPRS data services that takes into account prices that change dynamically, depending on the congestion in the system. The model is based on a Markov modulated Poisson process (MMPP), where the arrival rates are affected by a demand function. Performance metrics are also calculated. We believe that the proposed model captures the effect of dynamic pricing on the performance of GPRS data services in a realistic manner.