Complex Kernel Risk-Sensitive Loss: Application to Robust Adaptive Filtering in Complex Domain

Recently, the maximum complex correntropy criterion (MCCC) algorithm has shown its superiority for adaptive filter in complex domain. Compared with the traditional mean square error (MSE) criterion-based algorithms, MCCC uses complex correntropy as similarity measure, which is robust even in the presence of impulse noise. However, the error performance surface of complex correntropic loss (CC-loss) is not optimal, which is steep nearby the optimal solution and flat far from the optimal solution. In this paper, we propose a new similarity measure in complex domain, namely, complex kernel risk-sensitive loss (CKRSL). Based on CKRSL, we derive a new adaptive filter algorithm in the complex domain by using the gradient-based method, i.e., the CKRSL algorithm. Compared with the MCCC algorithm, the CKRSL algorithm has faster convergence rate and higher filtering accuracy. Meanwhile, it is robust against outlier. In addition, we provide the stability analysis and the steady-state excess MSE for the CKRSL algorithm. Simulation results confirm the correctness of the theoretical results and the superiority of the CKRSL algorithm.

[1]  Shiyuan Wang,et al.  Generalized Complex Correntropy: Application to Adaptive Filtering of Complex Data , 2018, IEEE Access.

[2]  B. Widrow,et al.  Stationary and nonstationary learning characteristics of the LMS adaptive filter , 1976, Proceedings of the IEEE.

[3]  Xian-Da Zhang,et al.  Matrix Analysis and Applications , 2017 .

[4]  Liu Xi,et al.  Maximum Correntropy Kalman Filter With State Constraints , 2017, IEEE Access.

[5]  Xi Liu,et al.  > Replace This Line with Your Paper Identification Number (double-click Here to Edit) < , 2022 .

[6]  Paulo S. R. Diniz,et al.  Adaptive Filtering: Algorithms and Practical Implementation , 1997 .

[7]  Shukai Duan,et al.  Quantized kernel maximum correntropy and its mean square convergence analysis , 2017, Digit. Signal Process..

[8]  Azam Khalili,et al.  Steady-State Tracking Analysis of Adaptive Filter With Maximum Correntropy Criterion , 2017, Circuits Syst. Signal Process..

[9]  Weifeng Liu,et al.  Kernel Adaptive Filtering: A Comprehensive Introduction , 2010 .

[10]  Xin Wang,et al.  Mixture correntropy for robust learning , 2018, Pattern Recognit..

[11]  Simon Haykin,et al.  Adaptive filter theory (2nd ed.) , 1991 .

[12]  Liang Wang,et al.  Robust Recognition via Information Theoretic Learning , 2014, SpringerBriefs in Computer Science.

[13]  Hua Qu,et al.  Convergence performance analysis of an adaptive kernel width MCC algorithm , 2017 .

[14]  Fei Wang,et al.  Maximum total correntropy adaptive filtering against heavy-tailed noises , 2017, Signal Process..

[15]  Chi K. Tse,et al.  Kernel Adaptive Filters With Feedback Based on Maximum Correntropy , 2018, IEEE Access.

[16]  Nanning Zheng,et al.  Steady-State Mean-Square Error Analysis for Adaptive Filtering under the Maximum Correntropy Criterion , 2014, IEEE Signal Processing Letters.

[17]  Nanning Zheng,et al.  Generalized Correntropy for Robust Adaptive Filtering , 2015, IEEE Transactions on Signal Processing.

[18]  Badong Chen,et al.  Kernel adaptive filtering with maximum correntropy criterion , 2011, The 2011 International Joint Conference on Neural Networks.

[19]  José Carlos Príncipe,et al.  Complex Correntropy: Probabilistic Interpretation and Application to Complex-Valued Data , 2017, IEEE Signal Processing Letters.

[20]  Panos M. Pardalos,et al.  On the optimization properties of the correntropic loss function in data analysis , 2014, Optim. Lett..

[21]  Weifeng Liu,et al.  Correntropy: A Localized Similarity Measure , 2006, The 2006 IEEE International Joint Conference on Neural Network Proceedings.

[22]  D. Mandic,et al.  Complex Valued Nonlinear Adaptive Filters: Noncircularity, Widely Linear and Neural Models , 2009 .

[23]  Fei Wang,et al.  Kernel adaptive filtering under generalized Maximum Correntropy Criterion , 2016, 2016 International Joint Conference on Neural Networks (IJCNN).

[24]  Badong Chen,et al.  Risk-sensitive loss in kernel space for robust adaptive filtering , 2015, 2015 IEEE International Conference on Digital Signal Processing (DSP).

[25]  Nanning Zheng,et al.  Convergence of a Fixed-Point Algorithm under Maximum Correntropy Criterion , 2015, IEEE Signal Processing Letters.

[26]  João P. F. Guimarães,et al.  Complex Correntropy Function: properties, and application to a channel equalization problem , 2017, Expert Syst. Appl..

[27]  Weifeng Liu,et al.  Correntropy: Properties and Applications in Non-Gaussian Signal Processing , 2007, IEEE Transactions on Signal Processing.

[28]  José Carlos Príncipe,et al.  Using Correntropy as a cost function in linear adaptive filters , 2009, 2009 International Joint Conference on Neural Networks.

[29]  Bernard C. Picinbono,et al.  On circularity , 1994, IEEE Trans. Signal Process..

[30]  Hongbin Zhang,et al.  Kernel Recursive Generalized Maximum Correntropy , 2017, IEEE Signal Processing Letters.

[31]  Liangjun Chen,et al.  Generalized Correntropy based deep learning in presence of non-Gaussian noises , 2018, Neurocomputing.

[32]  Jacob Benesty,et al.  A fast exact least mean square adaptive algorithm , 1990, International Conference on Acoustics, Speech, and Signal Processing.

[33]  José Carlos Príncipe,et al.  A loss function for classification based on a robust similarity metric , 2010, The 2010 International Joint Conference on Neural Networks (IJCNN).

[34]  Bo Du,et al.  Robust Graph-Based Semisupervised Learning for Noisy Labeled Data via Maximum Correntropy Criterion , 2019, IEEE Transactions on Cybernetics.

[35]  Nanning Zheng,et al.  Kernel Risk-Sensitive Loss: Definition, Properties and Application to Robust Adaptive Filtering , 2016, IEEE Transactions on Signal Processing.