For femtocell downlink interference in the case of dense deployment of femtocells, a cluster-based resource allocation scheme was proposed. Firstly, Fractional Frequency Reuse (FFR) was used to divide all cells into difference spatial regions, which could suppress the co-tier interference among macrocells and cross-tier interference between a macrocell and femtocells. Secondly, all the femtocells were divided into disjoint clusters based on the knowledge of graph theory and convex optimization theory, then a channel allocation algorithm based on fairness of rate was adopted to assign sub-channel to the femtocell user equipment, for limiting the co-tier interference among femtocells. Finally, a distributed power control algorithm was used to adjust the power of femtocells dynamically to improve the performance of the system. The simulation results show that, compared with the conventional clustering method, the proposed algorithm can improve the Signal-to-Interference plus Noise Ratio (SINR) and throughput, and the probability of system throughput below 4Mb/s was reduced to 30%. At the same time, compared with the non-grouping algorithm, the fairness of the proposed algorithm was improved by 12%, which could make users get higher satisfaction.