In this paper, we study the resource allocation problem of two-cell cognitive radio systems, each cell includes multiple secondary users (SUs). The aim is to maximize the sum capacity of the two cells with constraints. For comparison, two types of multi-carrier modulation, conventional orthogonal frequency division multiplexing (OFDM) and filter bank based multi-carrier (FBMC), are used to evaluate the algorithm's performance. In case of two cells with multiple users per cell, distributed allocation scheme is valid. Due to the mixed integer optimization problem of the single cell, the multiple access channel (MAC) helps to solve each cell's problem to ensure the distributed game converging to equilibrium through transferring the original problem into concave function with linear constraints. Another issue in resource allocation problem is the channel state information (CSI). Because of the uncertainty of CSI, an estimation prescribed by the outage probability of primary system is considered in this work. The combination of single cell's optimization and convergence of two cells solves the problem of resource allocation. Some conclusions can be drawn from simulation results and show the proposed algorithm performs well in two cells resource allocation problem. 1