On Network User Time Perference And Time-Variant Pricing Mechanism

With the development of network technology and the rise of mobile Internet, it is possible for human beings to enjoy network service anywhere and anytime. However, it will bring the explosion of network traffic and will make network traffic vary drastically over time. The operator’s networks are confronted with great pressure of expansion. Meanwhile, the explosive emergence of new mobile Internet services has been eroding the operator’s traditional revenue from voice/message services. Both the network regulators and operators are challenged to increase the network revenue which will contribute to network expansion and guarantee the healthy development of networks. The thesis finds ways to describe the user’s time-preferences of the behavior model of network users, and gives the optimal pricing mechanism to maximize the network revenue. It also analyses how the mechanism will influence the networks’ social welfare.The thesis first introduces some basic knowledge on network pricing, including basic models, pricing theory and pricing methods for multiple services. Then, with the help of previous study, we manage to describe the user’s time preferences of the behavior model. Based on the behavior model, an optimal pricing mechanism is provided, which will help us to find proper prices to maximize the network revenue with the given model parameters. Through analysis, we conclude that the impact of price variation near the optimal price on the network revenue turns out to be small, in other words, flat rate in periods where users have similar behaviors brings little revenue loss. In order to make the mechanism easy to be implemented, we have derived a half time-variant pricing mechanism which brings a customized tradeoff between complexity and revenue. According to the mechanism, network operator can find their own optimal prices according to the user’s time preferences and the tolerability of revenue loss.After that, the thesis delivers the model of pricing mechanism under the influence of user’s time preferences and the existence of Nash equilibrium. We analyze respectively how the revenue-maximized pricing mechanism will influence the network’s social welfare when the network operator knows complete and partial user information. If the operator has complete information about the user’s time preferences (specifically, the network utility over different users and time), according to analysis, we conclude that the operator can maximize the revenue and social welfare of the network at the same time. While if the operator only know the network utility over time, revenue maximization can result in some loss of network’s social welfare. And if the high-utility users are in the majority, the social warfare of revenue maximization is not less than2/3of the optimal value; network regulators who are concerned with the social welfare can relax the restriction on the operator’s behaviors to maximize their network revenvue.