Journal
IEEE TRANSACTIONS ON SIGNAL PROCESSING
Volume 69, Issue -, Pages 2188-2202Publisher
IEEE-INST ELECTRICAL ELECTRONICS ENGINEERS INC
DOI: 10.1109/TSP.2021.3067735
Keywords
Convergence; Adaptive filters; Signal processing algorithms; Filtering; Adaptation models; Kernel; Filtering algorithms; Complex domain; errors-in-variables; adaptive filtering; fixed point
Categories
Funding
- National Natural Science Foundation of China [62071391, 61701419]
- Natural Science Foundation of Chongqing [cstc2020jcyj-msxmX0234]
- Fundamental Research Funds for the Central Universities [2020jd001]
Ask authors/readers for more resources
The study develops a fixed point maximum total complex correntropy (FP-MTCC) adaptive filtering algorithm to improve the performance of MTCC. The convergence analysis of the FP-MTCC is provided in the paper. Additionally, two recursive FP-MTCC (RFP-MTCC) algorithms are developed for online adaptive filtering with transient analysis.
Adaptive filtering for complex-valued data plays a key role in the field of signal processing. So far, there has been very little research for the adaptive filtering in complex-valued errors-in-variables (EIV) model. Compared with the complex correntropy, the total complex correntropy has shown superior performance in the EIV model. However, the current gradient based maximum total complex correntropy (MTCC) adaptive filtering algorithm has suffered from the tradeoff between fast convergence rate and low weight error power. In order to improve the performance of MTCC, we develop a fixed point maximum total complex correntropy (FP-MTCC) adaptive filtering algorithm in this study. The convergence analysis of the FP-MTCC is also provided in the paper. Furthermore, we develop two recursive FP-MTCC (RFP-MTCC) algorithms for the online adaptive filtering and provide the transient analysis of RFP-MTCC. Finally, the validity of the convergence and the superiority of the proposed algorithms are verified by simulations.
Authors
I am an author on this paper
Click your name to claim this paper and add it to your profile.
Reviews
Recommended
No Data Available