A fast simulation annealing(SA)algorithm for the design of diffractive optical elements(DOE) for uniform illumination is presented. The Tsallis statistic and corresponding utility function are introduced into a hybrid algorithm in which the self-iterative and SA algorithms are combined to enhance the efficiency. Compared with traditional SA algorithm,simulated results show that it saves 99% of time to converge the incident energy into a desired region with the same mean square error(MSE).

Refractive index can be exactly measured between 124 nm and 1 700 nm, and mean square error(MSE) is far less than 1. Such result suggested that the measurement is very precise.

The Mean Square Error (MSE) of channel estimation and Bit Error Rate (BER) performance are given and simulation results demonstrate that the iterative algorithm using method B has good performance approaching the ideal condition.

Compared with the normal scheme without STBC, 3dB Mean Square Error (MSE) performance gains and fewer restrictions on the length of channel impulse response are demonstrated.

We first compute the minimum mean square error (MSE) of the LS channel estimation, and then derive the optimal criteria of the training sequence with respect to the minimum MSE.

Compared with WDM, WVF can reduce the mean square error (MSE) by 50% when the signal to noise ration (SNR) is in the range of -15dB to -11dB.

A discount factor which minimizes the total Mean Square Error (MSE) of the process output is obtained under the assumption that the process disturbance is either a white noise series or an IMA(1,1) series.

The traditional adaptive algorithm(PG algorithm) has a low computational complexity and is suitable for realtime implementation. However, its convergence of adaptation is not fast and the MSE of the filter coefficient is high, especially in the low input SNR. This paper presented a new variable stepsize adaptive algorithm for a second order adaptive IIR notch filter. It uses the estimate of the instantaneous output signal autocorrelation to control the step size update. This technique can reduce the effect...

The traditional adaptive algorithm(PG algorithm) has a low computational complexity and is suitable for realtime implementation. However, its convergence of adaptation is not fast and the MSE of the filter coefficient is high, especially in the low input SNR. This paper presented a new variable stepsize adaptive algorithm for a second order adaptive IIR notch filter. It uses the estimate of the instantaneous output signal autocorrelation to control the step size update. This technique can reduce the effect of the uncorrelated noise in a low input SNR situation. It provides fast convergence and reduces the MSE of the filter coefficient. In the satellite communication system, the VSLMSL algorithm is applied to NarrowBand interference cancellation. The simulation result shows that under the condition of input SNR=-32.77dB, the output SNR=-6.93dB is got, which demonstrates the robustness and advance of our algorithm.

A new image denoising method based on wavelet transform is proposed. The coefficients including low frequency and high frequency of wavelet transform are magnified properly before using image denoising method based on wavelet transform. Then the noise in image is eliminated by using local threshold wavelet method. In the last, the coefficients of wavelet transform are reduced aptly and restructured. According to the result of experiment, the given algorithm is much better than the conditional median value...

A new image denoising method based on wavelet transform is proposed. The coefficients including low frequency and high frequency of wavelet transform are magnified properly before using image denoising method based on wavelet transform. Then the noise in image is eliminated by using local threshold wavelet method. In the last, the coefficients of wavelet transform are reduced aptly and restructured. According to the result of experiment, the given algorithm is much better than the conditional median value filtering method and Wiener filtering method. Especially, the MSE of image has been decreased a lot while the PSNR being improved markedly.

Orthogonal Frequency Division Multiplexing (OFDM) with a suitably chosen guard interval is considered as an effective means for eliminating intersymbol interference of high-rate transmission over dispersive channels. Time variance of the channel, however, leads to a loss of subchannel orthogonality, resulting in an interchannel interference (ICI). In this paper, a multilayer perceptron scheme trained with MSE criterion was proposed for the nonlinear channel equalization. The proposed method was used...

Orthogonal Frequency Division Multiplexing (OFDM) with a suitably chosen guard interval is considered as an effective means for eliminating intersymbol interference of high-rate transmission over dispersive channels. Time variance of the channel, however, leads to a loss of subchannel orthogonality, resulting in an interchannel interference (ICI). In this paper, a multilayer perceptron scheme trained with MSE criterion was proposed for the nonlinear channel equalization. The proposed method was used in a realistic low voltage power lines with wideband coupling amplifier, which formed a nonlinear channel for the experiments. The performances of traditionally used linear equalizer and the nonlinear equalizers trained with MSE criterion proposed in this paper are compared. The effectiveness of the proposed approach is demonstrated via experiments by applying it to a PLC OFDM system with time-varying multipath fading channel.