Training and optimization of an artificial neural network controlling a hybrid power filter

A hybrid power compensator (HPC) consisting of a static VAr compensator and a dynamic compensator needs to be optimally controlled during the compensation of nonlinear loads. The HPC must be controlled to meet minimum requirements in terms of power factor and harmonic distortion, while at the same time minimizing its total cost. An artificial neural network (ANN) is used to control the HPC amidst a very dynamic power system environment. The performance of a reference ANN is evaluated while controlling an HPC connected to a typical nonlinear industrial load. The training and performance of the ANN is then optimized in terms of training set size, training set packing and ANN topology and the performance compared to the reference ANN. This paper highlights the importance of optimising the mentioned ANN parameters to achieve optimum ANN training and modeling accuracy. The results obtained reveals that the application of an ANN in controlling an HPC is feasible given that the ANN parameters are chosen appropriately.     

A VQ-based blind image restoration algorithm

Learning-based algorithms for image restoration and blind image restoration are proposed. Such algorithms deviate from the traditional approaches in this area, by utilizing priors that are learned from similar images. Original images and their degraded versions by the known degradation operator (restoration problem) are utilized for designing the VQ codebooks. The codevectors are designed using the blurred images. For each such vector, the high frequency information obtained from the original images is also available. During restoration, the high frequency information of a given degraded image is estimated from its low frequency information based on the codebooks. For the blind restoration problem, a number of codebooks are designed corresponding to various versions of the blurring function. Given a noisy and blurred image, one of the codebooks is chosen based on a similarity measure, therefore providing the identification of the blur. To make the restoration process computationally efficient, the principal component analysis (PCA) and VQ-nearest neighbor approaches are utilized. Simulation results are presented to demonstrate the effectiveness of the proposed algorithms.     

High-performance MIM capacitor using ALD high-k HfO/sub 2/-Al/sub 2/O/sub 3/ laminate dielectrics

For the first time, we successfully fabricated and demonstrated high performance metal-insulator-metal (MIM) capacitors with HfO/sub 2/-Al/sub 2/O/sub 3/ laminate dielectric using atomic layer deposition (ALD) technique. Our data indicates that the laminate MIM capacitor can provide high capacitance density of 12.8 fF//spl mu/m/sup 2/ from 10 kHz up to 20 GHz, very low leakage current of 3.2 /spl times/ 10/sup -8/ A/cm/sup 2/ at 3.3 V, small linear voltage coefficient of capacitance of 240 ppm/V together with quadratic one of 1830 ppm/V/sup 2/, temperature coefficient of capacitance of 182 ppm//spl deg/C, and high breakdown field of /spl sim/6 MV/cm as well as promising reliability. As a result, the HfO/sub 2/-Al/sub 2/O/sub 3/ laminate is a very promising candidate for next generation MIM capacitor for radio frequency and mixed signal integrated circuit applications.     

Hydrogen annealing effect on DC and low-frequency noise characteristics in CMOS FinFETs

The hydrogen annealing process has been used to improve surface roughness of the Si-fin in CMOS FinFETs for the first time. Hydrogen annealing was performed after Si-fin etch and before gate oxidation. As a result, increased saturation current with a lowered threshold voltage and a decreased low-frequency noise level over the entire range of drain current have been attained. The low-frequency noise characteristics indicate that the oxide trap density is reduced by a factor of 3 due to annealing. These results suggest that hydrogen annealing is very effective for improving device performance and for attaining a high-quality surface of the etched Si-fin.     

Serially concatenated space-time codes with iterative decoding and performance limits of block-fading channels

This work considers space-time channel coding for systems with multiple-transmit and a single-receive antenna, over space uncorrelated block-fading (quasi-static) channels. Analysis of the outage probability over such channels reveals the existence of a threshold phenomenon. The outage probability can be made arbitrary small by increasing the number of transmit antennas, only if the E/sub b//N/sub 0/ is above a threshold which depends on the coding rate. Furthermore, it is shown that when the number of transmit antennas is increased, the /spl epsi/-capacity of a block-fading Rayleigh channel tends to the Shannon capacity of an additive white Gaussian noise channel. This paper also presents space-time codes constructed as a serial concatenation of component convolutional codes separated by an interleaver. These schemes provide full transmit diversity and are suitable for iterative decoding. The rate of these schemes is less than 1 bit/s/Hz, but can be made arbitrary close to 1 bit/s/Hz by the use of Wyner-Ash codes as outer components. Comparison of these schemes with structures from literature shows that performance gains can be obtained at the expense of a small decrease in rate. Computer simulation results over block-fading Rayleigh channels show that the frame-error rate of several of these schemes is within 2-3 dB from the theoretical outage probability.     

Analysis of feedback for higher order polarization-mode dispersion mitigation of NRZ 40-Gb/s optical transmission

First-order polarization-mode dispersion (PMD) compensation by means of a polarization controller and a differential delay line is not sufficient to guarantee error-free transmission for 40-Gb/s channels when higher order effects severely increase signal distortion. Higher order mitigation is possible by cascading more than one first-order block. However, only two-stage or three-stage devices remain simple enough to be actually controlled. The performance of such higher order PMD compensators is evaluated by means of numerical simulations. Two different feedback signals have been used, demonstrating that first-order and higher order PMD distortion of nonreturn-to-zero (NRZ) pulses at 40 Gb/s can be strongly mitigated for instantaneous values of the differential group delay (DGD) up to the bit slot, when the compensator is properly controlled.     

High-resolution photonics-based interference suppression filter with wide passband

A new topology for a photonic signal processor, which overcomes the basic recursive frequency response problem that limits the passband range, is presented. The structure is based on a new multiple-wavelength offset-cavity structure that is cascaded with a series of unbalanced delay line structures. This not only can synthesize a very narrow notch response with good shape factor but also permits a multifold extension of the free spectral range (FSR) and passband width. Results on the interference mitigation filter demonstrate a stopband of 1% of center frequency and a fourfold increase in the FSR and passband width, while also having a very small shape factor, in excellent agreement with predictions.     

Pseudorandom active UWB reflectors for accurate ranging

In this letter, the concept of pseudorandom active reflector, based on the ultra-wideband (UWB) technology, is introduced. It consists of a simple device that repeats a slightly delayed version of the received UWB signal only in certain time intervals according to a suitable pseudorandom time-hopping sequence. An example of application of this device for accurate ranging in precise location systems is given. The advantages of this solution are in the hardware simplicity (only the analog section is present), in the low power consumption of the reflector and in the low timing constraint regarding the relative transmitter and reflector clock rates.     

Geometric derivation of expectation-maximization and generalized successive interference cancellation algorithms with applications to CDMA channel estimation

The expectation-maximization (EM) algorithm is well established as a computationally efficient method for separable signal parameter estimation. Here, a new geometric derivation and interpretation of the EM algorithm is given that facilitates the understanding of EM convergence properties. Geometric considerations lead to an alternative separable signal parameter estimator based on successive cancellation. The new generalized successive interference cancellation (GSIC) algorithm is then applied to multiuser delay and channel estimation for code-division multiple access (CDMA) radiolocation and communication. The radiolocation application uses a handshaking protocol in which multiple reference nodes transmit acknowledge (ACK) CDMA packets to a master node, which then computes round-trip travel times (RTTs). Simulation results comparing EM and GSIC are presented for the CDMA channel-estimation problem.