A range-free localization algorithm using received signal strength compare is proposed to balance the accuracy of the localization result with the complex technologies for the distance estimate. The algorithm’s name is received signal strength compare (RSSC). According to the non-cooperative characteristic of primary users (PUs) in cognitive radio networks, RSSC does not need the cooperation between PUs and secondary users (SUs). RSSC localization algorithm counts the PUs’ area based on comparing the received signal strength (RSS) of SUs, and estimates the centroid of the area as the location of the PUs. RSSC performance analysis in terms of its localization error parameterized by SU density, node density and SNR presents at the end of this paper. Experiments demonstrate that the localization accuracy is enhanced a lot with RSSC as compared to other range-free localization algorithms.