September 1995, Volume 4, Number 3
Presented paper deals with the reduction of computational requirements of gradient algorithms for the control of adaptive antenna arrays. Reduction of arithmetical complexity is reached here by the application of a block signal processing to adaptive algorithms. Block versions of the classical Least Mean Square (LMS) algorithm and the Simplified Kalman Filter (SKF) are described in this submission. Adaptation parameters of the presented algorithms are illustrated by results of computer simulations. The block SKF (BSKF) exhibits twice higher computational requirements than LMS, the same misadjustment as LMS and lower rate of convergence than LMS when transversal filters have great number of taps and when relatively high block length of BSKF is used.
This contribution deals with the concept and circumferential structure of comb filter for video signal processing. The comb filter was implemented by means of an analog shift register CCD (BCCD). The main conclusions, including the measured frequency response characteristic and time responses of the necessary control signals, are presented. This problem has been investigated as part of the GACR project.
Our aim is to present some aspects of the mathematical theory of strange behaviour of nonlinear systems, especially of systems with symmetry. Proofs are emitted, the interested reader is advised to references. Our presentation is inevitably selective. We focus on parts of the theory with possible applications to electronic circuits and systems which may display chaotic behaviour.
The architecture of various programmable logic arrays such as PAL (Programmable Array Logic), PLA (Programmable Logic Array) and MAPL (Multiple Array Programmable Logic) differ slightly in interconnection. The introduced types of devices are called PLD (Programmable Logic Devices). It is a bulk of programmable AND functions (product terms), and OR functions. The whole circuit structure is completed by input/output or dedicated output macrocells allowing to do the minimization of product term number. PLD's internal AND-array, and OR-array differs from the discrete logic AND, and OR devices whereas the functions are similar. The troubleshooting of these devices differs too.
The need for spectral analysis of the signal received by frequency modulated continuous wave (FMCW) radar sensors (FMRS) is inevitable in many applications of FMRS. It can be effectivelly performed using only digital signal processors, based on the fast Fourier transform (FFT). To make the signal analysis by FFT processor easier, it is necessary to separate the received signal (which generally varies in amplitude and in frequency ) from the broadband noisy background at the microwave mixer output by the narrowband tracking filter. Besides a brief desription of the performance and application of FMRS, the analysis of FMCW signal, with respect to moving target indication (MTI), and the function of a suitable tracking filter system is presented in this paper.
Some opinions how to understand and teach spectral representation of discrete-time (DT) signals are discussed in this contribution. There are two excessive ways often used in the educational process. The first one presents discrete - time signal as a product of sampling of continuous - time (CT) signal. The spectral properties of DT signal are then derived from the properties of CT signal and from the peculiarity of the sampling process. Second way builds the spectral description only in the area of DT signals irrespective of the CT signal spectral theory. In our paper, another approach will be presented which prefers the second way but emphasizes the connection and common properties of DT and CT signals. This method enables understanding of new subject - matter because teacher can utilize student's knowledge from the area of CT signals.