2016/02/10 by Samrat Mukhopadhyay, Mukhopadhyay, Samrat, Bijit Kumar Das +3
Computer Science · Engineering · #Adaptation and Self-Organizing Systems (nlin.AO) #Advanced Adaptive Filtering Techniques #Advanced Algorithms and Applications #Blind Source Separation Techniques #FOS: Computer and information sciences #FOS: Physical sciences #Information Theory (cs.IT) #Methodology (stat.ME)
paper · pdf · doi:10.48550/arxiv.1602.03283
openalex publication_date 2016/02/10 · openalex created_date 2022/10/01 · openalex updated_date 2026/07/28
Performance analysis of l0 norm constrained Recursive least Squares (RLS)\nalgorithm is attempted in this paper. Though the performance pretty attractive\ncompared to its various alternatives, no thorough study of theoretical analysis\nhas been performed. Like the popular l0 Least Mean Squares (LMS) algorithm,\nin l0 RLS, a l0 norm penalty is added to provide zero tap attractions on\nthe instantaneous filter taps. A thorough theoretical performance analysis has\nbeen conducted in this paper with white Gaussian input data under assumptions\nsuitable for many practical scenarios. An expression for steady state MSD is\nderived and analyzed for variations of different sets of predefined variables.\nAlso a Taylor series expansion based approximate linear evolution of the\ninstantaneous MSD has been performed. Finally numerical simulations are carried\nout to corroborate the theoretical analysis and are shown to match well for a\nwide range of parameters.\n