Catálogo de publicaciones - libros

In this paper, an adaptive array beamforming by an unstructured neural network based on the mathematics of holographic storage is presented. This work is inspired by similarities between brain waves and the wave propagation and subsequent interference patterns seen in holograms. Then the mathematics to produce a general mathematical description of the holographic process is analyzed. From this analysis it is shown that how the holographic process can be used as an associative memory network. Additionally, the process may also be used a regular feed-forward network. The most striking aspect of these network is that, using the holographic process, the apriori knowledge of the system may be better utilized to tailor the neural network for an adaptive beamforming problem. This aspect, makes this neural network formation process particularly useful for the beamforming.

Kohonen neural network is capable of self-organizing and recognizingclustering center, which is used in many artificial intelligence (AI) fields. One electronic support measures (ESM) system must sort the received radar pulses to cells with same features by pulse parameters, such as radio frequency (RF), angle of arrival (AOA), pulse width (PW), Pulse Repetition Interval(PRI), etc. Kohonen SOFM algorithm is one valid method for clustering, which can be used to accomplish such radar pulses sorting. Considering the variety character of pulses parameters which is the character of modern radar system, a new definition of “distance” in the SOFM neural net is proposed in this paper, which decreases the effect of large variety range of special parameter among them. This paper employs the “distance” to improve the clustering capability in such special environments. The computer simulation shows the validity of these improvements.

In this paper, the application of minimal resource allocation network (MRAN) trained with Unscented Kalman Filter (UKF) to the nonlinear channel equalization problems was discussed. Using novel criterion and prune strategy, the algorithm uses online learning, and has the ability to grow and prune the hidden neurons to realize a minimal network structure. Simulation results show that the equalizer is well suited for nonlinear channel equalization problems and the proposed equalizer required short training data to attain good performance.

A Jumping Genes Paradigm that combines with fuzzy rules is applied for optimizing the digital IIR filters. The criteria that govern the quality of the optimization procedure are based on two basic measures. A newly formulated performance metric for the digital IIR filter is formed for checking its performance while its system order which usually reflects upon the required computational power is also adopted as another objective function for the optimization. The proposed scheme in this paper was able to obtain frequency-selective filters for lowpass, highpass, bandpass and bandstop with better performance than those previously obtained and the filter system order was also optimized with lower possible number.

Magnetoencephalography (MEG) is an important noninvasive, non-hazardous technology for functional brain mapping, measuring the magnetic fields due to the intracellular neuronal current flow in the brain. However, the inherent level of noise in the data collection process is large enough to obscure the signal(s) of interest most often. In this paper, a practical denoising technique based on the wavelet transform and the multiresolution signal decomposition technique is presented. The proposed technique is substantiated by the application results using three different mother wavelets on the recorded MEG signal.

Single ion-channel signal of cell membrane is a stochastic ionic current in the order of picoampere (pA). Because of the weakness of the signal, the background noise always dominates in the patch-clamp recordings. The threshold detector is traditionally used to denoise and restore the ionic single channel currents. However, this method cannot work satisfactorily when signal-to-noise ratio is lower. A new approach based on hidden Markov model (HMM) is presented to restore ionic single-channel currents and estimate model parameters under white background noise. In the study, a global optimization method of HMM parameters based on stochastic relaxation (SR) algorithm is used to estimate the kinetic parameters of channel. Then, the ideal channel currents are reconstructed applying Viterbi algorithm from the patch-clamp recordings contaminated by noise. The theory and experiments have shown that the method performs effectively under the low signal-to-noise ratio (SNR<5.0) and has fast parameter convergence, high restoration precision and strong noise robusticity.

As radar signal environments become denser and radar signals become more complex, the task of an ESM operator becomes more difficult. This paper presented a de-interleaving/recognition system of radar pulses based on the combination of SVC and K-means clustering. Compared the conventional de-interleaving system, it can produce more complex and compact clustering boundaries according to the distribution characteristics of data set and has good generalization performance. The simulation experiment result shows that the system can sort efficiently radar signals in the high density and complex pulses environment.

Reconstruction of a three-dimensional scene using images taken from two views is possible if the relative pose of the cameras is known. A traditional approach to estimating the pose of the cameras uses eight pairs of corresponding points and involves the solution of a set of homogeneous equations. We propose a multi-layered feedforward network solution. Empirical results demonstrate the feasibility of using the network to recover the relative pose of the cameras in the three-dimensional world.

The adaptive disparity filter is refined, along with the monocular feature filter designed based on neuron dynamics, and a new stereopsis model is set up in this study. The disparities of the matched binocular features in stereo image pair are detected by the adaptive disparity filter, and removed by the monocular feature filter, only leaving those unmatched monocular features to be added to all depth planes determined by the disparities of the matched binocular features. Finally, visible surface perception is generated by the closed boundaries in the corresponding depth plane during the filling-in processing. By the above mechanism, the depth perception of surfaces in the occluded scenic images is realized, also, the figure-ground segmentation.

The camera model could be approximated by a set of linear models defined on a set of local receptive fields regions. Camera calibration could then be a learning procedure to evolve the size and shape of every receptive field as well as parameters of the associated linear model. For a multi-camera system, its unified model is obtained from a fusion procedure integrated with all linear models weighted by their corresponding approximation measurements. The 3-D measurements of the multi-camera vision system are produced from a weighted regression fusion on all receptive fields of cameras. The resultant calibration model of a multi-camera system is expected to have higher accuracy than either of them. Simulation and experiment results illustrate effectiveness and properties of the proposed method. Comparisons with the Tsai’s method are also provided to exhibit advantages of the method.