Collaborative Adaptive Filters for Online Knowledge Extraction and Information Fusion

We present a method for extracting information (or knowledge) about the nature of a signal. This is achieved by employing recent developments in signal characterisation for online analysis of the changes in signal modality. We show that it is possible to use the fusion of the outputs of adaptive filters to produce a single collaborative hybrid filter and that by tracking the dynamics of the mixing parameter of this filter rather than the actual filter performance, a clear indication as to the nature of the signal is given. Implementations of the proposed hybrid filter in both the real R and the complex C domains are analysed and the potential of such a scheme for tracking signal nonlinearity in both domains is highlighted. Simulations on linear and nonlinear signals in a prediction configuration support the analysis; real world applications of the approach have been illustrated on electroencephalogram (EEG), radar and wind data.

[1]  M. Hulle,et al.  The Delay Vector Variance Method for Detecting Determinism and Nonlinearity in Time Series , 2004 .

[2]  Mo Chen,et al.  An Online Method for Detecting Nonlinearity Within a Signal , 2006, KES.

[3]  Danilo P. Mandic,et al.  Signal nonlinearity in fMRI: a comparison between BOLD and MION , 2003, IEEE Transactions on Medical Imaging.

[4]  Andrew C. Singer,et al.  Multi-stage adaptive signal processing algorithms , 2000, Proceedings of the 2000 IEEE Sensor Array and Multichannel Signal Processing Workshop. SAM 2000 (Cat. No.00EX410).

[5]  T. Schreiber.  Interdisciplinary application of nonlinear time series methods , 1998, chao-dyn/9807001.

[6]  S. Thomas Alexander,et al.  Adaptive Signal Processing , 1986, Texts and Monographs in Computer Science.

[7]  Danilo P. Mandic,et al.  Collaborative Adaptive Learning using Hybrid Filters , 2007, 2007 IEEE International Conference on Acoustics, Speech and Signal Processing - ICASSP '07.

[8]  S. Haykin,et al.  Making sense of a complex world [chaotic events modeling] , 1998, IEEE Signal Process. Mag..

[9]  Richard D. Braatz,et al.  On the "Identification and control of dynamical systems using neural networks" , 1997, IEEE Trans. Neural Networks.

[10]  D. P. Mandic,et al.  NNGD algorithm for neural adaptive filters , 2000 .

[11]  P. Dokopoulos,et al.  Short-term forecasting of wind speed and related electrical power , 1998 .

[12]  Kazuo Yana,et al.  Adaptive estimation of the degree of system nonlinearity , 2000, Proceedings of the IEEE 2000 Adaptive Systems for Signal Processing, Communications, and Control Symposium (Cat. No.00EX373).

[13]  D.P Mandic,et al.  On the characterization of the deterministic/stochastic and linear/nonlinear nature of time series , 2008, Proceedings of the Royal Society A: Mathematical, Physical and Engineering Sciences.

[14]  B. Widrow,et al.  The complex LMS algorithm , 1975, Proceedings of the IEEE.

[15]  T. Schreiber,et al.  Surrogate time series , 1999, chao-dyn/9909037.

[16]  Tülay Adali,et al.  Approximation by Fully Complex Multilayer Perceptrons , 2003, Neural Computation.

[17]  Ali H. Sayed,et al.  Steady state performance of convex combinations of adaptive filters , 2005, Proceedings. (ICASSP '05). IEEE International Conference on Acoustics, Speech, and Signal Processing, 2005..

[18]  Danilo P. Mandic,et al.  A generalized normalized gradient descent algorithm , 2004, IEEE Signal Processing Letters.

[19]  T. Schreiber,et al.  Discrimination power of measures for nonlinearity in a time series , 1997, chao-dyn/9909043.

[20]  Danilo P. Mandic,et al.  A Complex-Valued RTRL Algorithm for Recurrent Neural Networks , 2004, Neural Computation.