Résumé: In recent years, distributed adaptive processing has received much attention from both theoretical and practical aspects. One of the efficient cooperation structures in distributed adaptive processing is the diffusion strategy, which provides a platform for the cooperation of nodes that run an adaptive algorithm, such as the least mean-squares (LMS) algorithm. Despite the studies that have been done on the diffusion-based LMS algorithm, the effect of deficient length on such structures has been overlooked. Accordingly, in this paper, we study the steady-state performance of the deficient length diffusion LMS algorithm. The results of this study show, in particular, that setting the tap length below its actual value leads to drastic degradation of the steady-state excess mean-square error (EMSE) and mean-square deviation (MSD) in diffusion adaptive networks. Furthermore, unlike the full-length case, where the steady-state MSD and EMSE decrease significantly with the step size reduction, this study shows that in the deficient-length scenario, there are no significant improvements in the steady-state performance by reducing the step size. Therefore, according to this study, the tap length plays a key role in diffusion adaptive networks since the performance deterioration due to deficient selection of tap length could not be compensated by an adjustment in the step size. Experiments exhibit a very good match between simulations and theory.

Langue: Anglais