The Adaptive Step Size NLMS and FBLMS Algorithms

The normalized least mean square (NLMS) algorithm and the frequency-domain block least mean square (FBLMS) algorithm have better cancellation performance and lower computational complexity during the clutter suppression in the bistatic noise frequency modulated continuous waveform (FMCW) radar system. However, in the real applications, in order to achieve high clutter cancellation and fast convergence speed as much as possible, many attempts require to be made to set the step size, which is trivial. In addition, the step size obtained by the matrix eigendecomposition will cost a large computation amount. In this letter, the adaptive step size based NLMS algorithm (ASNLMS) and the FBLMS algorithm (ASFBLMS) are proposed, which adjust adaptively the step size by calculating the correlation coherent of the reference signal between the current iteration and the la st iteration, and have better cancellation performance. The two proposed algorithms do not require the manual adjustment and eigendecomposition. Finally, the effectiveness of the two proposed algorithms have been verified by the real measured data.