Recent statistics show that mobile wireless broadband penetration has exceed that of fixed broadband, with 96.2% of the world’s population subscribing to wireless services and over 40% of US citizens carry smartphones. This trend is expected to continue in the future at much faster growth rates, and spectrum scarcity becomes a problem. Dynamic spectrum allocation has been proposed as a solution to improve spectrum efficiency and increase the spectrum access opportunities for wireless devices. However, the majority of the research focus on fixed usage. In this paper, we aim at providing high Quality of Experience for mobile devices, whose services cannot be finished within one coverage. We propose an adaptive spectrum allocation framework that optimizes spectrum utilization based on spectrum users’ hierarchy, requirements, as well as mobility. The simulation results show that the proposed adaptive spectrum allocation model lead to higher spectrum utilization efficiency, shorter average waiting time and lower probability of call dropping.