ADAPTIVE EXPONENT SMOOTHING GRADIENT ALGORITHM

A new algorithm, called the adaptive exponent smoothing gradient algorithm (AESGA), is developed from Widrow′s LMS algorithm. It is based on the fact that LMS algorithm has properties of time delaying and low pass filtering. This paper shows that the algorithm, on the domain of {Ω 1:α∈(0,1)}×{Ω 2:β(0,∞)} , unbiasedly and asymptotically converges to the Winner solution when the signal is a stationary Gauss stochastic process. The convergent property and the performance misadjustment are analyzed in theory. And calculation method of the algorithm is also suggested. Numerical results given by computer simulations show that the algorithm is effective.

BPSK通信系统采用自适应阵列时的抗干扰性能

本文研究了采用自适应阵抗宽带干扰时ＢＰＳＫ通信系统的性能。结果证明当干扰位于静默方向性图主波以外时，通信系统能获得极 误码率性能，当干扰信号方向接近 目标信号方向时，则性能显著降低。

ALGORITHM FOR THE DETECTION AND PARAMETER ESTIMATION OF MULTICOMPONENT LFM SIGNALS

A novel algorithm based on Radon-Ambiguity Transform (RAT) and Adaptive Signal Decomposition (ASD) is presented for the detection and parameter estimation of multicompo-nent Linear Frequency Modulated (LFM) signals. The key problem lies in the chirplet estimation. Genetic algorithm is employed to search for the optimization parameter of chirplet. High estimation accuracy can be obtained even at low Signal-to-Noisc Ratio(SNR). Finally simulation results are provided to demonstrate the performance of the proposed algorithm.

BLIND ADAPTIVE MULTIUSER DETECTION BASED ON A CONSTRAINED MMSE CRITERION

The proposed blind adaptive multiuser detector utilizes the signature waveform and time information of the desired user. With each received sample vector, the proposed algorithm updates the detector and gives the symbol estimate in the current time slot. Such property facilitates it to track time-varying channels.

A Study on PC/104 Embedded Automatic Balancing Control System

In this paper,an embedded real-time control system for automatic rotor balancing was studied.Benefiting from the modular design,this system can be easily re-constituted or expanded under different working conditions.The special designed hardware resists harsh environment.As an embedded application,it was very important to save system consumptions on both hardware and software,so the algorithms for unbalance vibration identification and attenuation were deduced,meantime a unified fast algorithm structure was achieved through the geometric analysis.Based on this structure,the signal processing algorithm was tested by an open data source,while the control algorithm was simulated using a basic rotor model,and then connected to a hyper gravity machine running online auto-balancing.The result confirms that the unbalancing vibration is effectively restrained.

Space discriminative function for microphone array robust speech recognition

Based on W-disjoint orthogonality of speech mixtures, a space d,scnmlnative tunetlon was proposer1 to enumerate and localize competing speakers in the surrounding environments. Then, a Wiener-like postfiherer was developed to adaptively suppress interferences. Experimental results with a hands-free speech recognizer under various SNR and competing speakers settings show that nearly 69 % error reduction can be obtained with a two-channel small aperture microphone array against the conventional single microphone baseline system. Comparisons were made against traditional delay-and-sum and Griffiths-Jim adaptive beamforming techniques to further assess the effectiveness of this method.

PRECONDITIONED METHODS FOR SPACE-TIME ADAPTIVE PROCESSING

This paper introduces the preconditioned methods for Space-Time Adaptive Processing(STAP).Using the Block-Toeplitz-Toeplitz-Block(BTTB)structure of the clutter-plus-noise covari-ance matrix,a Block-Circulant-Circulant-Block(BCCB)preconditioner is constructed.Based on thepreconditioner,a Preconditioned Multistage Wiener Filter(PMWF)which can be implemented by thePreconditioned Conjugate Gradient(PCG)method is proposed.Simulation results show that thePMWF has faster convergence rate and lower processing rank compared with the MWF.

Hardware implementation of adaptive filter for noise cancellation using TMS320C6713

Daily, we experience the effects of audio noise, which contaminates the original information bearing signal with noise from its surrounding environment. This paper focuses on real-time hardware implementation of multi-tap adaptive noise cancellation （ANC） system by using the least mean square （LMS） algorithm on TMS320C6713 to remove undesired noise from a received signal for various audio related applications. Three different experiments are carried out by considering different audio inputs to test the efficiency of the designed ANC system. The ＇C＇ code implementation of LMS algorithm is introduced and simulated in code composer studio （CCS）, then realized on the digital signal processor （DSP） C6713. The 300 Hz, 500 Hz, 800 Hz, 1 kHz and 3 kHz of tone signals and male speech signal are used as the reference inputs to trace the noise of signal until it is eliminated. The performance of ANC system is studied in terms of convergence speed, order of the filter and signal-to-noise ratio （SNR）. The experimentam results demonstrate that the designed system shows a consider- able improvement in SNR.

Bi-conjugate gradient based computation of weight vector in space-time adaptive processing

Dear editor,Space-time adaptive processing(STAP)techniques can effectively suppress strong ground clutter and detect moving target for airborne phased array radar by combing spatial signals and the temporal pulses simultaneously[1].It is known that,when the clutter satisfies the independent and identically-distributed(i.i.d.)condition,the sample matrix inversion(SMI)-based STAP[1]requires twice the number of degree of freedom

Two-direction nonlocal model for image interpolation

How to sufficiently exploit the self-similarity of natural images for image restoration has attracted extensive interest in the field of image processing in recent years.In fact,the self-similarity implies two-direction similarity structures inherent in images,when a group of similar patches are rearranged to form a matrix,there exists similarity between both columns and rows of this matrix.In this paper,we propose a two-direction nonlocal model (TDNL) to symmetrically exploit the two-direction similarity structures in images,the model directly takes the similar patches as local adaptive dictionary to represent each patch in the image and constrain the representation coefficients by Tikhonov regularization.TDNL can achieve the best results so far and obtain significant gains over the existing methods,in terms of both peak signal to noise ratio (PSNR) measure and the visual quality when it is applied to the problem of image interpolation.

Optical notch filter design based on digital signal processing

Based on digital signal processing theory,a novel method of designing optical notch filter is proposed for Mach-Zehnder interferometer with cascaded optical fiber rings coupled structure.The method is simple and effective,and it can be used to implement the designing of the optical notch filter which has arbitrary number of notch points in one free spectrum range(FSR).A design example of notch filter based on cascaded single-fiber-rings is given.On this basis,an improved cascaded double-fiber-rings structure is presented to eliminate the effect of phase shift caused by the single-fiber-ring structure.This new structure can improve the stability and applicability of system.The change of output intensity spectrum is finally investigated for each design parameter and the tuning characteristics of the notch filter are also discussed.

NOISY OBSERVATION BASED STABILIZATION AND OPTIMIZATION FOR UNKNOWN SYSTEMS

The paper addresses optimization of a performance function which either is optimized via stabilizing and controlling the underlying unknown system or is directly optimized on the basis of its noise-corrupted observations. For the first case the unknown system is identified and then the indirect adaptive control approach is applied to optimize the performance function. For the second case the stochastic approximation method is used to optimize the objective function, and it appears that a number of problems arising from applications may be reduced to the one solvable by this approach. The paper demonstrates some basic results in the area, but with no intention to give a complete survey.