On-line adaptive estimation of symbol period

The application of the least-mean-square (LMS) and recursive-least-square (RLS) algorithms to the estimation of symbol period is discussed. The algorithms are based on the measurement of the time between two consecutive detected transitions in noisy waveforms. Two versions of the algorithm are developed for white and colored measurement noise models. Conditions are derived that guarantee proper behavior, i.e. the convergence, of the LMS and RLS algorithms. Simulation results show the convergence of algorithms and compare the algorithms with respect to convergence speed.