This letter presents a two-stage method to extend the degenerate unmixing estimation technique (DUET) for reverberant speech separation. First, frequency-bin-wise attenuation and delay parameters are introduced and estimated by online update rules, to handle early reflections. Next, a mask reestimation algorithm based on the precedence effect is developed to detect and fix the errors on binary masks caused by late reflections. Experimental results demonstrate that the proposed method improves separation performance significantly.

A signal model and weighted-average based estimation techniques are proposed to estimate the angle-of-arrival (AOA) parameters of multiple clusters for a low data rate ultrawide band (LR-UWB) based wireless positioning system. The optimal AOA estimation techniques for the LR-UWB wireless positioning system according to the cluster condition are introduced and it is shown that the proposed techniques are superior to the conventional technique from the standpoint of performance.

In this study, the new signal model suitable for ultrawide band (UWB) indoor environments with random angle spread is proposed to estimate the angle-of-arrivals (AOAs) of clusters in a UWB wireless communication. The subspace based estimation technique adopted for this model is investigated and the estimates of the AOA and distribution parameter on the received clusters are obtained. The proposed model and estimation technique are verified using computer simulations, and the performance of the estimation error is analyzed.