Sparsity regularised recursive least squares adaptive filtering

E. M. Eksioglu*

*Corresponding author for this work

Research output: Contribution to journalArticlepeer-review

54 Citations (Scopus)


The authors propose a new approach for the adaptive identification of sparse systems. This approach improves on the recursive least squares (RLS) algorithm by adding a sparsity inducing weighted l1 norm penalty to the RLS cost function. Subgradient analysis is utilised to develop the recursive update equations for the calculation of the optimum system estimate, which minimises the regularised cost function. Two new algorithms are introduced by considering two different weighting scenarios for the l1 norm penalty. These new l1 relaxation-based RLS algorithms emphasise sparsity during the adaptive filtering process, and they allow for faster convergence than standard RLS when the system under consideration is sparse. The authors test the performance of the novel algorithms and compare it with standard RLS and other adaptive algorithms for sparse system identification.

Original languageEnglish
Pages (from-to)480-487
Number of pages8
JournalIET Signal Processing
Issue number5
Publication statusPublished - Aug 2011


Dive into the research topics of 'Sparsity regularised recursive least squares adaptive filtering'. Together they form a unique fingerprint.

Cite this