Stability analysis of covariance matrices in recursive least squares equalization for underwater acoustic communication: results from a sea trial
Abstract
Abstract Multi-channel equalization based on the recursive least squares (RLS) algorithm has been studied and applied for several decades in underwater acoustic communications. Recently, a forced-diagonalization approach has been introduced to the equalization process to reduce computational complexity. Although improved performance in at-sea experiments was statistically demonstrated in a previous study, the underlying reasons have not been discussed. This study investigates how forced diagonalization enhances performance through a stability analysis of covariance matrices obtained from an actual sea trial.
Fields of Study & Subjects
Underwater Vehicles and Communication Systems
Advanced Adaptive Filtering Techniques
Underwater Acoustics Research
Publication details
- Journal / source
- Japanese Journal of Applied Physics
- Volume
- 65
- Issue
- 4
- Pages
- 048004–048004
- Publication date
- 10 February 2026
- Language
- EN