The effectiveness of time dependent pricing in controlling usage incentives in wireless data network

Abstract

With advances of bandwidth-intensive mobile devices (e.g., smart phones, tablet computers), the data traffic for wireless data network has grown tremendously, and is predicted to further increase by more than 10 times in the next five years. The tremendous increase in transmission demand may cause serious congestions. This challenges network operators to find new ways to improve, or at least maintain their service quality. There are many solutions to address this problem from a technical viewpoint [1, 2]. In this paper instead, we try to solve this problem using a pricing approach. The rationales are: 1) traffic demand is highly volatile over time, so it is neither physically easy nor economically profitable to purely rely on technology to meet the extreme demand at peak time; and 2) a large amount of traffic do not have realtime requirement, or are unnecessary at all, and pricing has been proven as an effective way to shape users’ behaviors [3]. Current pricing models are not well suited for wireless applications: flat rate pricing is dominant in broadband network where bandwidth is usually sufficient, but usually causes congestions in wireless environment. For example, WeChat, a very popular mobile social application in China, uses data network to send text, voice and photos. Under the flat rate pricing model today, people often relentlessly upload photos and “short talk” of trivial errands. As a result, ISPs incur