Abstract: A new algorithm is proposed to achieve parameter estimation which is intensively studied and has been widely applied to many areas in signal processing.The idea of Compressed Sensing（CS） theory is introduced into parameter estimation.It integrates the processes of data sampling and compression in tranditional way by means of few non-adaptation random projections using CS algorithm.Compared with Matching Pursuit（MP） algorithm，CS algorithm has the obvious advantage even under sub-Nyquist sample rate.Theoretic analysis and experimental results illustrate the validity of the proposed algorithm.