An Adaptive Compressive Wideband Spectrum Sensing Algorithm Based on Least Squares Support Vector Machine

Most of the compressive wideband spectrum sensing algorithms need to recover spectrum, which require high computational complexity. Recently, a novel algorithm for without recovery (NoR) was proposed. Its complexity is several orders magnitude less than that recovery. However, enabling by structure-constrained assumption sparse NoR may fail. In order expand its scope application while reducing as much possible, we propose an adaptive (ADP) powerful hybrid no and partial (PR) algorithms. The ADP adaptively chooses or scheme depending on situation learned least squares support vector machine (LS-SVM). By simulation analysis, compared with NoR, PR another excellent (orthogonal matching pursuit), suits better practical applications.