The least mean square (LMS) algorithm is investigated for inputs from nonstationary random processes having periodic (period P) statistics. By defining P different coefficient vectors, each vector can be shown to ‘converge in the mean’ to a biased solution which is dependent upon the algorithm step size μ.
No takes yet. Share an insight, caveat, or question.
Des McLernon (1991) studied this question.
Synapse has enriched 5 related papers. Consider them for comparative context: