Distributed Recursive Least Squares Strategies for Adaptive Reconstruction of Graph Signals

None, None; None, None; None, None; None, None; None, None

Distributed Recursive Least Squares Strategies for Adaptive Reconstruction of Graph Signals

Conference Paper (2017)

Author(s)

Paolo Di Lorenzo (Università degli Studi di Perugia)

Elvin Isufi (TU Delft - Electrical Engineering, Mathematics and Computer Science)

Paolo Banelli (Università degli Studi di Perugia)

Sergio Barbarossa (Sapienza University of Rome)

Geert Leus (TU Delft - Electrical Engineering, Mathematics and Computer Science)

Research Group

Signal Processing Systems

Sampling Graph signal processing Adaptive networks Recursive least squares estimation

DOI related publication

https://doi.org/10.23919/EUSIPCO.2017.8081618 Final published version

To reference this document use

https://resolver.tudelft.nl/uuid:5c4ad87a-36fa-4cd9-9f6b-6ba3600c3e50

More Info

expand_more

Publication Year

2017

Language

English

Research Group

Signal Processing Systems

Pages (from-to)

2289-2293

ISBN (electronic)

978-0-9928626-7-1

Event

EUSIPCO 2017 (2017-08-28 - 2017-09-02), Kos Island, Greece

Downloads counter

158

Abstract

This work proposes distributed recursive least squares (RLS) strategies for adaptive reconstruction and learning of signals defined over graphs. First, we introduce a centralized RLS estimation strategy with probabilistic sampling, and we propose a sparse sensing method that selects the sampling probability at each node in the graph in order to guarantee adaptive signal reconstruction and a target steady-state performance. Then, a distributed RLS strategy is derived and is shown to be convergent to its centralized counterpart. The performed numerical tests show the performance of the proposed adaptive method for distributed learning of graph signals.