BROWSE
Detail
Journal Article
Correlated and Superposed Signal Decomposition for Multipath Channel Parameterization
Cited 1 time in 
Share 
Share
- Authors
- Kyung-Won Kim, Myung-Don Kim
- Issue Date
- 2024-10
- Citation
- IEEE Antennas and Wireless Propagation Letters, v.23, no.10, pp.2865-2869
- ISSN
- 1536-1225
- Publisher
- Institute of Electrical and Electronics Engineers
- Language
- English
- Type
- Journal Article
- Abstract
- In this letter, we propose a decomposition algorithm of correlated and superposed signals for multipath component estimation. The proposed algorithm introduces an improved implementation structure to address the overfitting problem of the space-alternating generalized expectation maximization (SAGE) algorithm for accurate estimation. To obtain stable and generalizable results, the proposed algorithm aims to the sparsest representation with the minimal degrees of freedom since model complexity and generalizability tend to have an inverse relationship. The performance comparisons with the SAGE and orthogonal matching pursuit algorithms are also provided based on actual time-varying channel measurements.
- KSP Keywords
- Channel measurement, Decomposition algorithm, Degrees of freedom(DOF), Multipath channels, Orthogonal Matching pursuit, Overfitting problem, Signal decomposition, Time-varying channels, accurate estimation, expectation maximization(EM), matching pursuit algorithms