Skip to main content

Recursive Least Squares With Minimax Concave Penalty Regularization for Adaptive System Identification

Publication ,  Journal Article
Li, B; Wu, S; Tripp, EE; Pezeshki, A; Tarokh, V
Published in: IEEE Access
January 1, 2024

We develop a recursive least squares (RLS) type algorithm with a minimax concave penalty (MCP) for adaptive identification of a sparse tap-weight vector that represents a communication channel. The proposed algorithm recursively yields its estimate of the tap-vector, from noisy streaming observations of a received signal, using expectation-maximization (EM) update. We prove the convergence to a local optimum of the static least squares version of our algorithm and provide bounds for the estimation error. We study the performance of the recursive version numerically. Using simulation studies of Rayleigh fading channel, Volterra system and multivariate time series model, we demonstrate that our recursive algorithm outperforms, in the mean-squared error (MSE) sense, the standard RLS and the ℓ 1-regularized RLS.

Duke Scholars

Published In

IEEE Access

DOI

EISSN

2169-3536

Publication Date

January 1, 2024

Volume

12

Start / End Page

66993 / 67004

Related Subject Headings

  • 46 Information and computing sciences
  • 40 Engineering
  • 10 Technology
  • 09 Engineering
  • 08 Information and Computing Sciences
 

Citation

APA
Chicago
ICMJE
MLA
NLM
Li, B., Wu, S., Tripp, E. E., Pezeshki, A., & Tarokh, V. (2024). Recursive Least Squares With Minimax Concave Penalty Regularization for Adaptive System Identification. IEEE Access, 12, 66993–67004. https://doi.org/10.1109/ACCESS.2024.3398550
Li, B., S. Wu, E. E. Tripp, A. Pezeshki, and V. Tarokh. “Recursive Least Squares With Minimax Concave Penalty Regularization for Adaptive System Identification.” IEEE Access 12 (January 1, 2024): 66993–4. https://doi.org/10.1109/ACCESS.2024.3398550.
Li B, Wu S, Tripp EE, Pezeshki A, Tarokh V. Recursive Least Squares With Minimax Concave Penalty Regularization for Adaptive System Identification. IEEE Access. 2024 Jan 1;12:66993–7004.
Li, B., et al. “Recursive Least Squares With Minimax Concave Penalty Regularization for Adaptive System Identification.” IEEE Access, vol. 12, Jan. 2024, pp. 66993–7004. Scopus, doi:10.1109/ACCESS.2024.3398550.
Li B, Wu S, Tripp EE, Pezeshki A, Tarokh V. Recursive Least Squares With Minimax Concave Penalty Regularization for Adaptive System Identification. IEEE Access. 2024 Jan 1;12:66993–67004.

Published In

IEEE Access

DOI

EISSN

2169-3536

Publication Date

January 1, 2024

Volume

12

Start / End Page

66993 / 67004

Related Subject Headings

  • 46 Information and computing sciences
  • 40 Engineering
  • 10 Technology
  • 09 Engineering
  • 08 Information and Computing Sciences