Yu, Chung-KaiSayed, Ali H.2017-12-192017-12-192017-12-19201310.1109/ICASSP.2013.6638527https://infoscience.epfl.ch/handle/20.500.14299/143338This work proposes a strategy to adjust the combination weights of an adaptive network in order to attain both faster convergence during the transient phase and lower mean-square-error during the steady-state phase. Optimal combination weights are designed for both phases, and a procedure for detecting the transition from one phase to the other is also described. Simulation results illustrate the operation of the proposed strategy.text::conference output::conference proceedings::conference paper