The radio spectrum has gradually become a deficient resource in wireless communications. Based on dynamic spectrum allocation, the cognitive radio (CR) has been proposed to overcome the spectrum deficiency. When the spectrum band is not occupied by primary user (PU) who is the licensed user of the spectrum band, the CR networks permits secondary users (SU) to operate an own data transmission. Since SUs can make interference to PU’s communication, CR network is required a target detection performance. In this thesis, we propose a radio resource allocation (RRA) scheme with satisfying the target detection performance in overlay CR network. Also we derive the near-optimal algorithm by adapting the transformation of original problem under interference-limited environment. By so saving the energy and time spent in sensing, the proposed RRA scheme provides the improvement of spectral efficiency. Through computer simulations, it is shown that the proposed RRA scheme exhibits better spectral efficiency performance over the conventional RRA scheme in terms of total throughput of SUs.