Understanding the Effects of Quota Trading on Mobile Usage Dynamics

We consider a time-based model for quota dynamics in mobile data networks. In particular we present a dynamic program to characterize the behavior of mobile users when they have the option to trade data during the month. The user will decide how much data to consume at each time period based on the utility that can be gained, the amount of quota remaining and the price available for trading. In contrast to past work on quota trading, our model explicitly takes into account the time remaining in the billing period when users make trading decisions. In addition, we propose a VCG formulation to ensure that users truthfully reveal the value they assign to traded data. We present a number of variants of the basic model that differ based on how much foresight users are assumed to have with respect to future price estimates and their utility needs. We use our models of quota dynamics to estimate the gain in user utility when quota trading is introduced. This in turn translates into an increased price that the operator can charge for each level of quota. We also compare the utility benefits of quota trading with an arguably simpler scheme in which end users buy and sell unused or extra quota with the operator itself.