Time-Dependent Pricing for Mitigating Fluctuation of Data Traffic Demand

摘要

Fluctuation of capacity demand throughout a day has long been a problem for ISPs. During peak time, ISPs face significant pressure to provide all on-going traffic with enough bandwidth and to satisfy their QoS requirements; during off-peak time, most of the network resources are left idle, resulting in a waste of capacity. In this paper, pricing is employed as an incentive mechanism to shift demand from traffic peak time to off-peak time. Users pay a flat rate monthly, and receive charge discounts if they voluntarily delay or discard their usage during peak time. The discount adopted in each slot is completely time-dependent and varies according to the real-time network congestion level. Specifically, the discount of each time slot is obtained by solving a revenue maximization problem.