Title :
A Generalised Mixed Norm Stochastic Gradient Algorithm
Author :
Boukis, C. ; Mandic, D.P. ; Constantinides, A.G.
Author_Institution :
Athens Inf. Technol., Athens
Abstract :
A novel stochastic gradient algorithm for finite impulse response (FIR) adaptive filters, termed the least sum of exponentials (LSE), is introduced. In order to provide a generalisation of the class of weighted mixed norm algorithms and at the same time avoid problems associated with a large number of free paramaters of such algorithms, LSE is derived by minimising a sum of error exponentials. A rigourous mathematical analysis is provided, resulting in closed form expressions for the optimal weights and the upper bound of the learning rate. The analysis is supported by simulations in a system identification setting.
Keywords :
FIR filters; adaptive filters; gradient methods; stochastic processes; adaptive filters; closed form expressions; error exponentials; finite impulse response filters; generalised mixed norm algorithm; least sum of exponentials; stochastic gradient algorithm; system identification setting; Adaptive filters; Cost function; Finite impulse response filter; Grid computing; Information technology; Mathematical analysis; Noise robustness; Stochastic processes; System identification; Upper bound; Adaptive Filtering; FIR filters; Gradient Descent Algorithms;
Conference_Titel :
Digital Signal Processing, 2007 15th International Conference on
Conference_Location :
Cardiff
Print_ISBN :
1-4244-0882-2
Electronic_ISBN :
1-4244-0882-2
DOI :
10.1109/ICDSP.2007.4288510
