This study compares the proposed hybrid method (wavelet robust M-estimation) to traditional ordinary least square) when there are de-noising or outlier problems for estimating multiple linear regression models using statistical criterion root mean square error (RMSE). According simulated and real data, is better than classical (Wavelet Ordinary Least Square) more accurate. The of less Wavelet S...

In this paper, we propose a new hypothesis testing method for detection of cisoids (complex sinusoids) from a single measurement of data. The testing is performed on the least square error. The least square error is shown to exhibit 1' distribution, which leads to an efficient threshold setup for the proposed method. The new method is a combined detection-estimation technique and provides impro...

A least-square infinite series approximation (L-SISA) technique is proposed for modeling amplitude modulated (AM) sinusoidal components of naturally occurring signals, such as those produced by traditional musical instruments. Each AM sinusoid is iteratively extracted using an analysis-by-synthesis technique and the problem of parameter estimation is linearised for least-square approximation th...

This paper considers an Type I Generalized Half Logistic Distribution. We discussed the scale (σ) and shape (θ) parameters estimation using the median ranks method (Benard‟s approximation). Rama Krishna(2008) 1 studied the Type I Generalized Half Logistic Distribution scale (σ) and shape (θ) parameters estimation using the Least Square Method in two step estimation method. Also we computed Aver...

Multiple input multiple output (MIMO) orthogonal frequency division multiplexing (OFDM) has received a great deal of attention of recently in achieving high data rate in wireless communication systems such as WIMAX. Channel estimation is, however, a critical issue for coherent demodulation. In this paper, a new channel estimator based on neural network with feedback for MIMO-OFDM mobile system ...