Parametric Density Estimation Using EM Algorithm for Collaborative Spectrum Sensing

Collaborative sensing of spectral occupancy can increase accuracy and relax the required sensitivity of individual sensing units. Collaborative sensing requires knowledge about the densities of collected sensing statistics to form the correct decision statistics for the optimum likelihood ratio test. In this paper, a parametric density estimation scheme using the expectation-maximization (EM) algorithm is proposed to estimate the parameters of densities that are drawn from a given family. When the log-likelihood function for the EM algorithm satisfies a certain condition, the maximization procedure is shown to require only a weighted sum of the collected sensing statistics. Numerical examples show that in various scenarios the proposed EM algorithm produces more accurate estimates than the sample average does.