A distributed channel allocation scheme aware of interference in cognitive radio networks is proposed. The objective of the proposed scheme is to minimize the number of dominant interference signals over a threshold instead of the total interference power. The proposed channel allocation can be treated by using game theory and proved to be a potential game, thus it is guaranteed to converge to a Nash equilibrium. In addition, the proposed channel allocation only requires the channel state of neighboring users while channel allocation minimizing the total interference power requires the channel state of all users. Simulation results reveal that the proposed channel allocation reduces both the required information exchange and the number of channel updates at a slight loss of network throughput.