7294 research outputs found
Sort by
"A reconfigurable Gaussian/Triangular basis functions computation circuit"
Get PDFA CMOS Gaussian/Triangular Basis functions computation circuit suitable for analog neural networks is proposed. The circuit can be configured to realize any of the two functions. The circuit can approximate these functions with relative root-mean-square error less than 1%. It is shown that the center, width, and peak amplitude of the dc transfer characteristic can be independently controlled. SPICE simulation results using 0.18 μ m CMOS process model parameters of TSMC18 technology are included
Flexible digital Scrambler/De-Scrambler system
Get PDFThe structural design of a flexible Scramble/De-Scrambler that uses aprogrammable length shift register and modulo-2 adder is presented. The key feature of the proposed design is its flexibility in varying the length of the delay elements pseudo-randomly and hence the encryption code. In addition, the encryption code can be re-programmed by the manufacturer through the EPROM involved in the design and by the user through the loadable pseudo-random sequence generator
Large Signal Analysis of Wideband Nonlinear Amplifiers with Third-Order Intermodulation and Second-Harmonic Injection
Get PDFLarge signal analysis of a wideband nonlinear amplifier excited by a multisinusoidal signal is pre-sented. The special case of two equal-amplitude sinusoids and their injected second-harmonic and third-order intermodulation component is considered in detail and the results are compared, whenever possible, with pre-viously published experimental results
ON-LINE IDENTIFICATION AND CONTROL THROUGH SERIES CONVERTER VOLTAGE OF A UNIFIED POWER FLOW CONTROLLER
Get PDFThe dynamic performance of a power system can be improved by controlling the voltage magnitude and phase angle of the converter voltages in the unified power flow controller (UPFC). Self-tuning adaptive control of the voltage magnitude of the series converter for stabilization of a power system is presented in this paper. The plant parameters are identified through a regressive least square algorithm and the stabilizing control is derived through a pole-shifting technique using the adaptive plant model. The controller has been tested for ranges of operating conditions and for various disturbances. From a number of simulation studies on a single machine infinite bus power system it was observed that the adaptive algorithm converges very quickly and also provides robust damping profiles
Growth of uniform layers of nanosized zeolites on 3D ordered macroporous carbon: synthesis, characterization, and catalytic evaluation
Get PDFNovel Peak Detection Algorithms for Pileup Minimization in Gamma Ray Spectroscopy
Get PDFA fast waveform sampling facility has been recently developed and integrated into the VAX-based data acquisition system at the Center for Applied Physical Sciences (CAPS). This study uses the above facility in developing algorithms for digitally determining the basic pulse parameters and tackling the problem of pulse pile-up in Gamma-ray spectroscopy. A number of parameter estimation and digital online peak localisation algorithms are being developed, including a pulse classification technique which uses a simple peak search routine based on the smoothed first derivative method, which gave a percentage error of peak amplitude of less than 1%. The classification technique has the unique feature of cutting down the computation largely by only allowing the event of interest to be executed by a particular algorithm. The set-up was also tested with random signals from a 137Cs test source. Gamma pulses from a 3" Na(TI) scintillation detector were captured as single and double pulses for the purpose of testing the peak detection algorithms. The pulse classification technique was tested successfully on a TMS320C6000 high performance floating-point processor yielding a reduction of the execution time to 2 mse