scientific article; zbMATH DE number 1160277
From MaRDI portal
Publication:4392489
The adaptive exponent smoothing gradient algorithm (AESGA), based on the LMS algorithm, is given by the authors as \[ \begin{aligned} L(j) &=L(j-1)+\alpha(\gamma(j)X(j)-L(j))\\ W(j)&=W(j-1)+\beta L(j)\\ \alpha\in &(0,1)\\ \beta\in & (0,\infty).\end{aligned}\tag{1} \] The authors claim that the algorithm (1) converges asymptotically to the Winner solution when the signal is a stationary Gaussian process. They have provided two examples.
Recommendations
This page was built for publication:
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q4392489)