Convex Combination of Diffusion Strategies over Networks - ANR - Agence nationale de la recherche Accéder directement au contenu
Article Dans Une Revue IEEE Transactions on Signal and Information Processing over Networks Année : 2020

Convex Combination of Diffusion Strategies over Networks

Résumé

Combining diffusion strategies with complementary properties enables enhanced performance when they can be run simultaneously. In this paper, we propose two convex combination schemes, the power-normalized one and the sign-regressor one. Without loss of generality, theoretical investigations are focused on the former. An analysis of universality shows that it cannot perform worse than any of its component strategies in terms of the excess mean-square-error (EMSE) at steady-state. A theoretical analysis of stability also reveals that it is more stable than affine combination schemes. Next, several adjustments are proposed to further improve the performance of convex combination schemes. Finally, simulation results are presented to demonstrate their effectiveness as well as the accuracy of the theoretical results.
Fichier principal
Vignette du fichier
jin2020convex.pdf (8.09 Mo) Télécharger le fichier
Origine : Fichiers produits par l'(les) auteur(s)

Dates et versions

hal-03347272 , version 1 (17-09-2021)

Identifiants

Citer

Danqi Jin, Jie Chen, Cédric Richard, Jingdong Chen, Ali Sayed. Convex Combination of Diffusion Strategies over Networks. IEEE Transactions on Signal and Information Processing over Networks, 2020, 6, pp.714-731. ⟨10.1109/TSIPN.2020.3038017⟩. ⟨hal-03347272⟩
21 Consultations
20 Téléchargements

Altmetric

Partager

Gmail Facebook X LinkedIn More