ICIC (inter-cell interference coordination) techniques are effectively applied into dense frequency reuse networks such as LTE (long term evolution). While mitigating the interference produced by nearby cells, researches only concentrate on maximum SINR or proportional fairness, without a consideration of various throughput demands. An adaptive resource allocation based on satisfaction (ARAS) algorithm is proposed not only to mitigate interferences but also to satisfy throughput demands adaptively. It utilizes eNBs (evolved-NodeBs) communications via X2 interface to allocate resources between cells. Each cell is divided into two zones: cell-edge and cell-center, of which a satisfaction is always tracked. The scheduler adjusts the resources allocation in a distributed manner related to the zone satisfaction states. Simulation results show that the proposed algorithm achieves a better satisfaction, meets different throughput demands and increases throughput fairness compared with reuse-1 model, FFR (fractional frequency reuse) and SFR (soft frequency reuse).