Kriging-Based Interference Power Constraint for Spectrum Sharing Based on Radio Environment Map

This paper proposes a probabilistic interference constraint method with radio environment map (REM) for spectrum sharing. Kriging interpolation is a well-known technique that can optimize the REM accuracy. However, although many researchers have maintained continuous interests in the accuracy of the REM, little study has been done to actually explore the interference constraint considering the estimation error. The proposed method uses ordinary kriging interpolation for the spectrum cartography. According to the predicted distribution of the estimation error, the allowed interference power to the primary user (PU) is approximately formulated. Numerical results show that the proposed method can achieve the probabilistic interference constraint asymptotically. Additionally, we compare the performance of the proposed technique with two methods: limit of the path loss-based method and perfect estimation. From the comparison results, it is shown that the kriging-based method has higher spectrum sharing opportunity than the path loss-based method, even if there are few measurement data.