Home » Multi-Kernel Adaptive Filtering
Multi-Kernel Adaptive Filtering
Over the decades, kernel adaptive filtering (kernel online learning) has received significant attention for online estimation tasks of nonlinear functions. This approach realizes considerable improvements compared to the conventional linear adaptive filtering, provided that a reasonable model is available with an appropriately designed kernel function. However, such a reasonable model is not always available particularly for real-time implementation. In nonstationary environments, an appropriate kernel, and therefore a reasonable model, may change over time. In these cases, the improvements are hardly attained. In this talk, a practical solution, which is "multi-kernel adaptive filtering", is presented. This approach employs multiple kernels which make the algorithm more robust against the choice of kernel functions. It does not explicitly design weights to the kernels unlike the existing multiple kernel learning (MKL) techniques. Indeed, those weights are embedded in the filter coefficients which are updated recursively by means of the metric projection onto a hyperplane (or a hyperslab) defined in a parameter space. The proposed algorithms are thus fully adaptive because the weights to the kernels are implicitly determined by the filter coefficients and have no need to be designed separately. Another remarkable feature of the multi-kernel approach is its higher degree of freedom than the MKL techniques. The efficacy of the approach is demonstrated by simulation in applications to nonlinear channel equalization and time-series data predictions.
Masahiro Yukawa received the B.E., M.E., and Ph.D. degrees from Tokyo Institute of Technology in 2002, 2004, and 2006, respectively. From October 2006 to March 2007, he was a Visiting Researcher at the Department of Electronics, the University of York, U.K. From April 2007 to March 2008, he was with the Next Generation Mobile Communications Laboratory at RIKEN, Saitama, Japan, and, from April 2008 to March 2010, he was with the Brain Science Institute at RIKEN. From August to November 2008, he was a Guest Researcher at the Associate Institute for Signal Processing, the Technical University of Munich, Germany. He is currently an Associate Professor at Niigata University, Japan. His research interests include mathematical adaptive signal processing, constrained optimization, and their applications to acoustic/communication systems. From April 2005 to March 2007, he was a recipient of the Research Fellowship of the Japan Society for the Promotion of Science (JSPS) . He received the Excellent Paper Award and the Young Researcher Award from the IEICE in 2006 and in 2010, respectively, the Yasujiro Niwa Outstanding Paper Award from Tokyo Denki University in 2007, and the Ericsson Young Scientist Award from Nippon Ericsson in 2009. He is a member of the Institute of Electrical and Electronics Engineers (IEEE) and the Institute of Electrical, Information and Communication Engineers (IEICE) of Japan.
1. Masahiro Yukawa, "Multi-kernel adaptive filtering", IEEE Trans. Signal Processing, accepted.
2. Lei Wang, Rodrigo C. de Lamare, and Masahiro Yukawa, ``Adaptive reduced-rank constrained constant modulus algorithms based on joint iterative optimization of filters for beamforming,'' IEEE Trans. Signal Processing, vol. 58, no. 6, pp. 2983--2997, June 2010.
3. Masahiro Yukawa, Konstantinos Slavakis, and Isao Yamada, ``Multi-domain adaptive learning based on feasibility splitting and adaptive projected subgradient method,'' IEICE Trans. Fundamentals, vol. E93-A, no. 2, pp. 456--466, February 2010.
4. Masahiro Yukawa and Isao Yamada, ``A unified view of adaptive variable-metric projection algorithms,'' EURASIP Journal on Advances in Signal Processing, vol. 2009, Article ID 589260, pp. 1--13, 2009.
5. Masahiro Yukawa, Rodrigo C. de Lamare, and Isao Yamada, ``Robust reduced rank adaptive algorithm based on parallel subgradient projection and Krylov subspace,'' IEEE Trans. Signal Processing, vol. 57, no. 12, pp. 4660--4674, December 2009.
6. Masahiro Yukawa, ``Krylov-proportionate adaptive filtering techniques not limited to sparse systems,'' IEEE Trans. Signal Processing, vol. 57, no. 3, pp. 927--943, March 2009.
7. Masahiro Yukawa, Rodrigo C. de Lamare, and Raimundo Sampaio-Neto, ``Efficient acoustic echo cancellation with reduced-rank adaptive filtering based on selective decimation and adaptive interpolation,'' IEEE Trans. Audio, Speech and Language Processing, vol. 16, no. 4, pp. 696--710, May 2008.
8. Masahiro Yukawa, Konstantinos Slavakis, and Isao Yamada, "Adaptive parallel quadratic-metric projection algorithms," IEEE Trans. Audio, Speech and Language Processing, vol. 15, no. 5, pp. 1665--1680, July 2007.
9. Masahiro Yukawa and Isao Yamada, "Pairwise optimal weight realization ---Acceleration technique for set-theoretic adaptive parallel subgradient projection algorithm," IEEE Trans. Signal Processing, vol. 54, no. 12, pp. 4557--4571, Dec. 2006.