Title :
Partial update NLMS algorithm for sparse system identification with switching between coefficient-based and input-based selection
Author :
Wu, Jinhong ; Doroslovacki, Milos
Author_Institution :
ECE Dept., George Washington Univ., Washington, DC
Abstract :
Long impulse response system identification presents two challenges for standard normalized least mean square (NLMS) filtering method: heavy computational load and slow convergence. When the response is sparse, partial update algorithms can reduce the computational complexity, but most often at the expense of performance. This paper discusses the tap selection rule for partial update NLMS algorithm in the case of white Gaussian input. We consider output mean square error (MSE) minimization based on gradient analysis and propose an algorithm that switches tap selection criterion between the one based on filter coefficient magnitudes and the one based on input magnitudes. We show that for identifying sparse systems, the new algorithm can outperform standard NLMS significantly with a reduced computational load.
Keywords :
communication complexity; filtering theory; gradient methods; mean square error methods; telecommunication switching; coefficient-based selection; computational complexity; filter coefficient magnitude; gradient analysis; input-based selection; long impulse response system identification; mean square error minimization; normalized least mean square; sparse system identification; switching; tap selection rule; white Gaussian; Algorithm design and analysis; Computational complexity; Convergence; Filtering algorithms; Finite impulse response filter; Mean square error methods; Minimization methods; Switches; System identification; Wireless communication;
Conference_Titel :
Information Sciences and Systems, 2008. CISS 2008. 42nd Annual Conference on
Conference_Location :
Princeton, NJ
Print_ISBN :
978-1-4244-2246-3
Electronic_ISBN :
978-1-4244-2247-0
DOI :
10.1109/CISS.2008.4558528