This PDF file contains the front matter associated with SPIE Proceedings Volume 7502, including the Title Page, Copyright information, Table of Contents, Introduction (if any), and the Conference Committee listing

The paper is a digest of work presented during a cyclic Ph.D. student symposium on Photonics and Web Engineering WILGA 2009. Symposium is organized by ISE PW in cooperation with professional organizations IEEE, SPIE, PSP and KEiT PAN. There are presented mainly Ph.D. and M.Sc. theses as well as achievements of young researchers. These papers, presented in such a big number, more than 250 in some years, are in certain sense a good digest of the condition of academic research capabilities in this branch of science and technology. The undertaken research subjects for Ph.D. theses in electronics is determined by the interest and research capacity (financial, laboratory and intellectual) of the young researchers and their tutors. Basically, the condition of academic electronics research depends on financing coming from applications areas. During Wilga 2009 there were organized, and thus the paper debates, the following topical sessions concerning applications of advanced electronics and photonics systems: merging of electronic systems and photonics, Internet engineering, distributed measurement systems, security in information technology, astronomy and cosmic technology, HEP experiments, environment protection, image processing and biometry. The paper contains also more general remarks concerning the workshops organized by and for the Ph.D. students in advanced photonics and electronics systems.

We present a novel technique of image data extraction for calibration and modeling purposes. It employs structured light approach using sequence of Gray stripes patterns for points localization and indexation. Code images for the indexation are build based on simple classification of pixel color. Points co-ordinates are extracted from intersection of code image edges without using time consuming filtering with moving window operator. UNION-FIND algorithm is used for removing spurious edge points appearing due to noise presence. Technique results in high accuracy and reliability of extracted points.

Many modern image analysis systems such as digital face image based authentication, 3D reconstruction, entertainment systems and many others, require fast and accurate face detection system. More advanced techniques, achieving better results, usually require high amount of computations and therefore use to much of the system resources. Using parallel GPU architecture in the crucial parts of the algorithms can significantly reduce the time needed for the detection. This paper prooves, that it is possible to use advantages of the GPU over CPU in the face detection on the example of the Discrete Gabor Jets face and facial features detection algorithm.

In this paper we present an approach to teaching image processing and pattern recognition with the use of the OpenCV library. Image processing, pattern recognition and computer vision are important branches of science and apply to tasks ranging from critical, involving medical diagnostics, to everyday tasks including art and entertainment purposes. It is therefore crucial to provide students of image processing and pattern recognition with the most up-to-date solutions available. In the Institute of Electronics at the Technical University of Lodz we facilitate the teaching process in this subject with the OpenCV library, which is an open-source set of classes, functions and procedures that can be used in programming efficient and innovative algorithms for various purposes. The topics of student projects completed with the help of the OpenCV library range from automatic correction of image quality parameters or creation of panoramic images from video to pedestrian tracking in surveillance camera video sequences or head-movement-based mouse cursor control for the motorically impaired.

Individualized driving assistance system approach is not explored extensively. We are developing an algorithm that could aid to driver's assistance on a curved country road. Steering angle predictions were made using neural network and look-up table approaches. The predictions were made using four parameters based on analysis of the image sequence obtained from the camera installed in a car. Two drivers' steering angle prediction results were compared and evaluated by calculating the mean squared error between the true and predicted signals.

H.264/MPEG4-AVC is the newest video coding standard recommended by International Telecommunication Union - Telecommunication Standardization Section (ITU-T) and the ISO/IEC Moving Picture Expert Group (MPEG). The H.264/MPEG4-AVC has recently become leading standard for generic audiovisual services, since deployment for digital television. Nowadays is commonly used in wide range of video application ranging like mobile services, videoconferencing, IPTV, HDTV, video storage and many more. In this article, author briefly describes the technology applied in the H.264/MPEG4-AVC video coding standard, the way of real-time implementation and the way of future development.

The objective of this paper is designing, modeling, simulation and synthesis of four Image Enhancement techniques on FPGA. Image Enhancement Algorithms can be classified as point processing Techniques, in which operation is done on pixel level and Spatial Filtering Technique, in which operation is performed within neighborhood of a pixel. Algorithms of all the techniques are studied and hardware circuits are realized for them. Then hardware logic is modeled in Matlab Simulink using Xilinx System Generator Block set and synthesized onto Virtex4 xc4vsx35-10ff668 FPGA chip. Using hardware co-simulation feature of FPGA kit, the algorithms developed are validated.

In the paper using of High Dynamic Ranging (HDR) method for SAR images enhancement is presented. A strong scatterers are presented on SAR images as wide, circle points occupied a lot of pixels. It means, that some displayed scatterers are out of image dynamic range. In results of this the details of strong scatterer can be unrecognized and some important information can be lost. The proposed method is based on known, from computer graphics, photography, etc. High Dynamic Ranging technique. Applying of this method in SAR imaging processing allows to obtain more detailed image and also can be used as a start point to another method improving quality of final image (ex. CLEAN method).

In a distributed video coding (DVC) system, the total bit-rate depends on bit-rate of the key frames (Intra frames) and the Wyner-Ziv (WZ) frames. The key frames bit-rate is relatively easy to control since they are encoded with an Intra coding scheme (e.g. H.264/AVC Intra), and there are many proposed solutions in literature which address this issue^{1, 2}. On the other hand, rate control (RC) of the WZ frames at the encoder is more difficult since the bit-rate of WZ frames is difficult to predict and control due to the absence of the side information at the decoder side. In this work, an RC algorithm developed within the VISNET II, the European Network of Excellence, is presented as an efficient solution to achieve and maintain the target bit-rate for the overall Intra frames and WZ bitstream, mainly by changing the degree of compression of the Intra frames which is controlled by quantization parameter (QP). In order to maintain a similar quality for the Intra and WZ frames, the WZ quantization index (QIndex) follows the QP changes. A statistical model is used to describe the relationship between QIndex and the WZ frames bit-rate. Additionally, an analysis of influence of the key frames residuum complexity on WZ frames bitrate was conducted. The proposed algorithm adapted to the VISNET2 WZ video codec³ confirms its efficiency in terms of achieving and maintaining the target bit-rate.

This paper presents research on importance of color information in face verification system. Four most popular color spaces where used: RGB, YIQ, YC_bC_r, luminance and compared using four types of discriminant classifiers. Experiments conducted on facial databases with complex background, different poses and light condition show that color information can improve the verification accuracy compared to the traditionally used luminance information. To achieve the best performance we recommend to use multi frames verification encoded to YIQ color space.

The paper presents algorithms for hands segmentation in digital images. Two approaches has be analyzed - segmentation from images with full colour information and images with reduced colour information. The testing data used for analysis of algorithms using colour information were videos recorded on Technical University of Warsaw with HD camera. The videos with reduced colour information were acquired for the purpose of Visnet II project, using camera with a filter which reduced colour information.

The "Pi of the Sky" robotic telescope has been designed for monitoring of a significant fraction of the sky with good time resolution and range. The main goal of the "Pi of the Sky" detector is to look for short timescale optical transients arising from various astrophysical phenomena, mainly for optical counterparts of Gamma Ray Burst (GRB). System design, observations methodology as well as developed algorithms also make this detector a sophisticated instrument looking for novae and supernovae stars and monitoring of blasars and AGNs activity. Final detector consists of two sets of 16 cameras, one camera covering field of view of 20°x20°. It follows that the full system will cover field of view of SWIFT BAT satellite. The final system is under construction now. Required hardware and software tests were performed with a prototype located in Las Campanas Observatory in Chile since June 2004. The most important result by the "Pi of the Sky" prototype so far was an independent detection and observation of the prompt optical emission from the naked-eye GRB080319B.

The ultimate goal of the "Pi of the Sky" apparatus is observation of optical flashes of astronomical origin and other light sources variable on short timescales, down to tens of seconds. We search mainly for optical emissions of Gamma Ray Bursts, but also variable stars, novae, blazars, etc. This task requires a precise photometry - accurate measurement of the source's brightness (and it's variability). "Pi of the Sky" single cameras' field of view is about 20° x 20°. This causes a significant deformation of a point spread function (PSF), reducing quality of brightness measurement with standard photometric algorithms. To improve photometry, an attempt to investigate PSF based on real star images was made. However, results turned out to be inconclusive due to miscellaneous sky-observing effects. Therefore we decided to perform laboratory measurements, using a CCD camera and an artificial light source as a star simulator. This work shows preliminary results of this study - a set of high resolution PSF shapes, pixel response and pixel sensitivity functions. Finally, an idea how to simulate a real star image in the "Pi of the Sky" system is presented.

Real-time analysis of data from optical telescopes becomes very important topic in the modern astronomy. Particularly interesting and important is fast automatic identification of short optical transients. Many violent and interesting processes occur on very short timescales. Such kind of processes may be related to gamma-ray bursts, but there are also other interesting processes in the short time regime. One of the biggest problems with their investigation is fast detection and distribution, which would allow deeper studies in the most interesting phase of the event. Thus development of real-time pipelines for identification of optical transients is a very hot topic. Another issue which becomes important is that data streams produced by modern astronomical experiments becomes larger and larger, which requires real-time analysis and pre-selection of only most interesting data. We present implementation of such a pipeline dedicated to identification of optical transients in the "Pi of the Sky" experiment.

Gamma-Ray Bursts (GRB) are short impulses of gamma radiation (time duration ranges from 0.1 to 100 s) distributed almost isotropically across the sky. Most probably they arise during explosions or collisions of stars at cosmological distances. In this paper we discuss their origin, properties and current models. The methods of their observations will be also described. At the end short summary of "Pi of the Sky" GRBs observations will be presented, mainly focused on the naked-eye burst GRB080319B.

We present the first few steps of creation the second edition of the variable stars catalogue, based on the "Pi of the Sky" data, collected during two years 2006-2007. We have chosen - 3000 variable star candidates from about 1.5 million objects.

This paper describes integrated monitoring and control system being prepared for the "Pi of the Sky" experiment. The "Pi of the Sky" experiment looking for rapidly changing optical objects such as Gamma Ray Bursts (GRB) will consist of two sites, each one built of 12 CCD cameras installed on 3 parallactic mounts, equipped in many dedicated electronic and software systems and operated by a cluster of PC computers. All system components, both hardware and software, will operate in remote location as a robot and will have to be equipped in a monitoring and control system that allows to trace all parameters of the system and enable to find and recover from all problems quickly either by automatic procedures or by remote operator. Architecture and implementation of such system based on established hardware and software tools is presented in this paper.

The "Pi of the Sky" system was initially created for searching of optical partners of gamma ray bursts [5,7]. The system is located in Chile and consists of two CCD cameras on a common mounting base [1]. Currently, it performs various astronomical observations [1,5,7,8]. Among others, system also records passages of satellites and fragments of Earth's artificial satellites, so called "space debris". Since now, this kind of data was identify as a disturbing transient signal and team members usually focus on identify and eliminating it. Handle of this problem is especially important, because software algorithm which search for optical companion of gamma ray bursts focus on transient phenomena events. On the other hand, comparison of the "Pi of the Sky" system parameters with different facilities which are used for space debris searching and monitoring shows, that its properties are similar to some of those systems. It means, that it should be possible, to obtain valuable data of artificial satellites motion analyzing "Pi of the Sky" data. Moreover, expected in the near future system upgraded to 6 small telescopes equipped totally with 24 CCD detectors system [1] will cover a major part of the sky (field of view of each detector equals 20 x 20 degrees) which gives possibility of visual measurements objects up to 16 magnitude. It will be an unusual advantage in the field of system categories with that size. In the paper, we would like to shortly describe a space debris community activities, present properties of selected systems which are using for space debris surveys and finally examples of observational data will be presented.

The possibility for using geostationary satellite signals to determine hydrometeors parameters has been examined. The method description and experimental results from the analysis of main forms of hydrometeors such us clouds, rainand snow-falls are presented. Instantaneous and average spectra are shown. The conclusions about the environmental safety and possibility of extensive use of the method have been made. This method is applicable to warning of different hazardous events.

The first Hungarian satellite, Masat-1 is designed and built at the Budapest University of Technology and Economics. Instead of working from pre-built components, the members develop the whole satellite, including design, implementation and testing. This paper introduces the key issues about designing a small satellite, with a detailed description of the Communication subsystem. In the paper, special attention is being paid for the digital signal processing part of the development, as it is the key to fulfill the easy detection constraint. The possible solutions are examined and there is a recommendation of a method which could easily fulfill the proposed requirements.

The paper presents a satellite platform developed in Institute of Radioelectronics, Warsaw University of Technology, for present and future student pico-satellite missions. The platform consists of all systems necessary to proper operation of payload equipment that could be placed on the satellite: mechanical structure, power supply unit, communications and control system.

Presented paper describes an inexpensive computer for CubeSat missions based on widely available commercial of-the-shelf components. Main requirements for the discussed project were: ability to handle satellite onboard data, platform flexibility and portability. Hardware design of the presented project is based on an ARM microcontroller that is able to run FreeRTOS real time kernel used as a base for software design. Presented project is a part of the PW-Sat satellite. It is scheduled to be launched during Vega maiden flight.

In this paper a processing algorithm for sliding spotlight SAR raw data with consideration of orbit geometry is presented. The processing is based on the chirp scaling algorithm with a pre-processing step for the sliding spotlight mode. The chirp scaling processing algorithm for SAR raw data assumes a flat Earth geometry. This approximation is not valid for a high resolution satellite Proposed algorithm considers a circular orbit and corrects the processing algorithm accordingly through a range dependent velocity ratio. The processing algorithm is evaluated on a simulated raw data of a scene consisting of ideal point targets.

In modern High Energy Physics Experiments the electronics modules for controlling and/or data acquisition are very often exposed to high neutron radiation. It's well known that it can cause many failures during work of such systems, even total break downs. In such case, the studies of radiation sensitivity of electronic devices should take place to avoid problems during the lifetime of the experiments. This article describes the Single Event Effects (SEE) tests of Xilinx Spartan3 XC3S400 FPGA populated on Altium Live Design Evaluation Board. The results of the tests are presented and discussed. Additionaly gamma-ray spectrometry analysis of the board has been performed.

Passive Bistatic Radar (PBR) research is at its zenith with several notable PBR systems currently operational, or available for deployment. Such PBRs include the Manastash Ridge Radar (MRR) developed for and by academia; Silent Sentry developed as a commercial concern by Lockheed Martin; and Homeland Alerter (HA100) also a commercial system developed by Thales. However at present, despite the existence of numerous PBR prototypes, take up of commercial passive radar technology remains slow. This is due in part to technology immaturity, in part to politics, and particularly due to the fact that monostatic radars perform so well. If PBRs are to enjoy longevity as a viable technology then it is imperative that they address certain niche application areas, with the aforementioned MRR being one prime example of this. The focus of this paper will be an analysis of a PBR system that utilised FM radio signals of opportunity to detect aircraft targets with an RCS generally not lower than 20 m². The paper will demonstrate the theoretical detection coverage of an FM based PBR operating in a severe interference environment.

The development of new sensor systems able to detect, identify and find position of the targets equipped with STEALTH technology began early in 1990s. Some of the sensor systems utilise acoustic, magnetic, seismic and/or other physical effects of target activity. A reason motivating the development of new sensor systems based on other than radar or optical principal of operation in detecting targets is that the systems usually emit no electromagnetic energy during operation. Thanks to their passive principle they provide the users with the advantages of hidden positioning and difficult discovery with reconnaissance tools. Therefore, some of the new UGS systems also allow detecting low-flying targets, such as helicopters, propeller or jet aircraft, etc., in the detection range of up to several kilometres. The information of flight direction is usually estimated and deduced from spatiotemporal sequence detections by multiple interlinked UGS systems. The submitted paper analyses low-flying target position finding principle on Time Direction Of Arrival (TDOA) basis. It presents the qualities of found UGS arrangement topologies and the characteristics of the unambiguous position determination of low-flying targets. It also contains mathematical description of signal digital processing intended to find low-flying target's position. The processed results are presented in table and diagram forms created in Matlab mathematical environment. All the presented detection and identification results were obtained from real recorded signals.

Passive detection systems guarding battle space, often called Unattended Ground Sensors (UGS) in publications, have been utilised in military applications for several years experiencing a number of changes and updates in design recently. Their primary task is to detect and identify the targets of interest that may be defined as personnel, wheel or track vehicles or other type. As standard, seismic, magnetic, passive infrared, acoustic or other sensors are used for detection and identification, if applicable. UGS systems then digitally process and evaluated the signals generated by the sensors. The submitted paper explains signal digital analysis and the options for detecting low-flying targets through surface seismic wave utilisation. It contains mathematical description of and the results from processing the surface seismic waves excited by some types of aircraft flying at low ground altitudes. It also mentions the options for low-flying target identification and suitable type descriptor selection. The processed results are presented in table and diagram forms created in Matlab mathematical environment. All the presented detection and identification results were obtained from real recorded signals.

The purpose of this paper is to briefly introduce the structure and features of the developed virtual passive FM radar implemented in Matlab system of numerical computations and to present many alternative ways of its performance. An idea of the proposed solution is based on analytic representation of transmitted direct signals and reflected echo signals. As a spatial filtering subsystem a beamforming network of ULA and UCA dipole configuration dedicated to bistatic radar concept is considered and computationally efficient procedures are presented in details. Finally, exemplary results of the computer simulations of the elaborated virtual simulator are provided and discussed.

A Beamforming Network (BN) concept of Uniform Linear Array (ULA) and Uniform Circular Array (UCA) dipole configuration designed to multistatic passive radar is considered in details. In the case of UCA configuration, computationally efficient procedure of beamspace transformation from UCA to virtual ULA configuration with omnidirectional coverage is utilized. If effect, the idea of the proposed solution is equivalent to the techniques of antenna array factor shaping dedicated to ULA structure. Finally, exemplary results from the computer software simulations of elaborated spatial filtering solutions to reference and surveillance channels are provided and discussed.

In the paper the analysis of the WiMAX signals for passive radar application is presented. Passive radar, or Passive Coherent Location (PCL) uses non-cooperative sources of radiation in order to detect targets. The detection is carried out by comparing the echo signal with the reference signal, obtained with a separate antenna or reconstructed from the received signal. The telecommunication signals which are often used by PCL are not designed for target detection, therefore, are not ideal for passive radar applications. The purpose of this paper is to analyze the usefulness of the WiMAX signal for the passive radar.

The paper describes the first results of noise radar experiments carried out at Warsaw University of Technology. The radar system was built with Commercial Off-The-Shelf (COTS) components: log-periodic antennas, an arbitrary waveform generator and a two-channel spectrum analyzer. The radar operated in the continuous-wave mode, and the aim was to detect moving targets in the received signal. The paper shows the system setup as well as the numerical results obtained from the recorded signals.

The paper describes stepped frequency continuous wave (SFCW) ground penetrating radar (GPR), where signal's frequency is discretely increased in N linear steps, each separated by a fixed ▵f increment from the previous one. SFCW radar determines distance from phase shift in a reflected signal, by constructing synthetic range profile in spatial time domain using the IFFT. Each quadrature sample is termed a range bin, as it represents the signal from a range window of length cτ/2, where τ is duration of single frequency segment. IFFT of those data samples resolves the range bin in into fine range bins of c/2N▵f width, thus creating the synthetic range profile in a GPR - a time domain approximation of the frequency response of a combination of the medium through which electromagnetic waves propagates (soil) and any targets or dielectric interfaces (water, air, other types of soil) present in the beam width of the radar. In the paper, certain practical measurements done by a monostatic SFCW GPR were presented. Due to complex nature of signal source, E5062A VNA made by Agilent was used as a signal generator, allowing number of frequency steps N to go as high as 1601, with generated frequency ranging from 300kHz to 3 GHz.

This paper is focusing on the influence of the geometry of a multilateration surveillance system onto the accuracy of measuring emitter (target) coordinates regardless of calculation algorithm. Cramer-Rao lower bound concept is used. The way to find an optimal sensors configuration is investigated.

The paper presents the concept of the algorithm to detect and estimate motion parameters of ground slow moving targets. Detection algorithm based on Sigma-Delta antenna with two spatial channels STAP processing. Motion parameters were estimated using novel Non Iterative Map-Drift (NIMD) technique, typically dedicated as an autofocus technique for earth imaging in SAR technique. The main goal of this paper was to extend and test NIMD technique as an algorithm for ground moving targets detected with two channels (sigma/delta) antenna.

The paper describes application of the subsurface GPR sensing in the area of "Kirilov" Church (Ukraine, Kiev) to determine the moisture and the structure of the underground soil layers under the church. Sensing was performed with the use of Stepped Frequency Continuous Wave (SFCW) signals with frequency band from 130 up to 270 MHz. The sensing results showing soil moisturizing level received using the signals' phase structure are shown in the paper. Based on these results conclusions about the reasons of church building deformations can be used when developing engineering activities to prevent these deformations.

This paper studies Doppler spectra of the reflections from rain at different polarizations. Particularly, it presents results of Doppler-polarimetric radar signal modelling and real data processing that show correlation between Differential Doppler Velocity (DDV), which is the difference between mean Doppler velocities at horizontal and vertical polarizations, and parameters of rain including rain microstructure and turbulence in rain.

A fuzzy detector and classifier of dangerous weather phenomena based on polarimetric radar measurements are described in this paper. Five polarimetric radar measurands, namely, horizontal reflectivity factor, differential reflectivity factor, linear depolarization ratio, specific differential phase, cross-correlation coefficient and altitude of resolution volume serve as inputs of the fuzzy detector and classifier. The output of the fuzzy detector and classifier is one of 8 possible classes: 0) No dangerous weather phenomenon is detected; 1) Lightning; 2) Aircraft icing; 3) Hail; 4) Hail+rain; 5) Heavy rain; 6) Wet snow; 7) Dense snow. A neural network backpropagation algorithm is also considered for training the fuzzy detector and classifier in case of having verified data.

This paper is focusing on simulation of the process of windshear radar detection over the surface to estimate the accuracy and reliability of operational algorithms used to make a decision at the detection of dangerous weather phenomena. The range of detection in infrared and microwave frequency bands is estimated.

This document presents the application of the Software Defined Radio-based platform in the multistatic radar. This platform consists of four-sensor linear antenna, Universal Software Radio Peripheral (USRP) hardware (radio frequency frontend) and GNU-Radio PC software. The paper provides information about architecture of digital signal processing performed by USRP's FPGA (digital down converting blocks) and PC host (implementation of the multichannel digital beamforming). The preliminary results of the signal recording performed by our experimental platform are presented.

The work is dedicated to analysis of influence of the phase errors in Noise Waveform SAR on the precision of the measurements made with the help of the differential SAR interferometry (d-InSAR) method. The motivation of the research is analysis of applications of d-InSAR in monitoring of engineer constructions state with the help of ground based SAR systems. The aim of the current work is to provide analysis of the phase errors in differential SAR interferometry caused by antenna distortions. And to develop a model for estimation of their influence on the interferograms obtained. We give analytical model of the processes taking part when antenna path is distorted, analyze mechanisms of phase errors occurring and give some examples obtained using modeling of NW SAR operation. The results included in the work are used in SAR Tx/Rx unit and scanning antenna design and for analyzing obtained interferograms.

The article presents a prototype wearable device that corrects inaccurate GPS readouts during pedestrian travel. The electronic circuit consists of a microcontroller, an accelerometer and a digital compass. The accelerometer readouts are filtered to detect the steps of the pedestrian and are also used to estimate the stride length. The digital compass provides the direction of motion. When the GPS parameters warn of a high dilution of precision, the location of the pedestrian is corrected by data provided by the accelerometer and the digital compass.

In this paper we present the concept of multiple sensors data acquisition from onboard of an Unmanned Air Vehicule (UAV). Because of flight instabilities caused by atmosferic movements (winds, thermals etc..) it is necessary to apply active stabilization in order to obtain reliable readings from observation sensors. The most stabilization-demanding sensor is Synthetic Aperture Radar (SAR) and in this paper two methods of stabilization are presented: hybrid (electromechanical) and electronic.

The paper presents the idea of digital navigation system for miniature quadrocopter unmmanded aerial vehilce (UAV). The navigation system is composed of two subsystems: the attitude and heading stabilization and high-level satelite navigation subsystem. Overall quadrocopter concept is presented as well as the solutions used for both navigation subsystems.

A new adaptive-type algorithm for target tracking in presence of fluctuating signals has been developed. The algorithm was designed with using the non-linear filtering theory and the Gaussian approximation method for the Swerling radar cross section (RCS) fluctuations model. The simulations results in different environment have revealed its high efficiency.

The paper presents a brief survey of the main approaches for solving the problem of radar waveform selection for optimal tracking. The traditional approach to the system design has been to treat the detection and tracking subsystems as completely separate entities. In last years some new types of radars equipped with highly agile software driven waveform generators have appeared. This makes it possible to change the transmitted waveform at each time step for obtaining minimal tracking errors. Thus the problem of the waveform selection can be formulated as the optimisation problem of the following form: for given class of signals and target movement model to find the tracking filter structure and waveform parameter sequence which result in a minimum variance estimate of the target state vector. Two main approaches to waveform optimising have been considered in the paper: the control theoretic approach and information theoretic approach. The relationships between sensor characteristics and tracking algorithm are discussed.

The paper presents an algorithm for calculating the 3D target position and the velocity vector from bistatic passive radar measurements corresponding to multiple transmitter-receiver pairs. The procedure uses closedform equations, which results in a very short computation time. The algorithm can be extended to use an arbitrary number of transmitter-receiver pairs.

Compression of speech signals is very important in signal processing, data transmission and data storage. Therefore in this paper a new method for compression speech signals that have a significant number of zeros and ones after encoding has been proposed and the proposed name for this method is Zeros-Ones Position Method (ZOPM). Also a new statistical evaluation method is used to evaluate the reconstructed signal and compare it with the original one. Results obtained from the simulation program show that in speech signals containing 60% of silent intervals can give 2.3 compression ratio (CR) due to ZOPM. In this paper the proposed ZOPM method is compared with Huffman encoding for signals with enough large numbers of zeros.

This paper deals with robust adaptive detection of a useful target in the presence of interfering signals. The background environment is assumed homogeneous and Gaussian with unknown covariance matrix. At the design stage, we devise a robust receiver with angular rejection capabilities, accounting for covariance and steering uncertainties. We prove that the maximization of the concentrated likelihood function shares a hidden convexity property. Specifically, exploiting some recent results concerning trigonometric polynomials, we formulate the apparently nonconvex optimization over the phase as a Semidefinite Programming convex optimization problem. At the analysis stage, we assess the performance of the new receiver in comparison with classic detectors available in open literature.

Problems concerning the calculation of continuous wavelet transform are considered for the case when adapted mother wavelet is specified only by its discrete values. The new technique of wavelet coefficients' computation for arbitrary scales is proposed. Experimental results of identifying of the epileptiform complexes in the test human electroencephalogram have proven the usefulness of developed technique.

An algorithm for parametric modelling of a specific type of time series, namely a series of time intervals is proposed and discussed in the paper. The necessary preprocessing steps are presented. They include timebase computation, interpolation, re-sampling and, depending on the application, detrending. The proposed approach of parametric modelling is based on autoregressive moving average (ARMA) model. The methods for ARMA model derivation assume that the model orders (the lengths of autoregressive and moving average filters) are known a priori. Since the result is highly sensitive to the order choice, it is necessary to establish rules for proper order selection. The solution to this, often underestimated, problem is also addressed in the article.

The window method is a fast and efficient approach to fractional delay filter design. However, gain of designed filter, differs from the ideal value which should be equal one. This effect, called gain deficit, negatively affects filter quality. The essence of the gain deficit problem as well as method and results of gain deficit correction are stated in this article. The influence of gain deficit correction on frequency response of the fractional delay filter and approximation error is discussed. An algorithm for finding optimum gain deficit correction factor has been proposed and relations between correction factor and filter and window parameters are presented in the paper.

Linear time invariant system with fuzzy parameters is considered. The concept of fuzzy transfer function is discussed. The problem is closely related to the question how to apply fuzzy techniques to such systems. In system theory Fourier transform, convolution operation and correlation function are frequently used. Generally, filters are described in two ways: in time domain, by impulse or step response, and in frequency domain, by amplitude and phase frequency response. In the paper, linear filters with fuzzy parameters using fuzzy convolution operation and fuzzy Fourier transform are analyzed. The approach is compared with conventional one. Herein, some examples of fuzzy filters for image applications are presented.

The Digital Signal Processing (DSP) requires high computing power. Von Neuman architecture reaches it's limits. To meet the high requirement of DSP algorithms the parallel processing techniques are being used. The paper presents an example of DSP application on NVIDIA graphic card using CUDA environment. The presented result shows the high effectiveness of multi-core graphical card in Digital Signal Processing. The processing time of the example spectrogram is reduced 10 times compared to Central Processing Unit (CPU) of the standard PC

The article presents wavelet-based spectral smoothing and Principal Component Analysis as a pre-processing step for Head Related Transfer Function (HRTF) filter design. Spectral smoothing of HRTF by means of Stationary Wavelet Transform discards low-value peaks and notches in HRTF which can be treated as noise. The Principal Component Analysis applied to the HRTF data extracts most important frequency components responsible for sound source localization in azimuth and elevation. These methods used in combination are used to reduce the order of filters that are applied in 3D audio rendering system.

Work is devoted to important topic of acoustic signals processing in a pilot's cabin of aircraft in which the high noise level is observed. We have investigated heuristic approach of acoustic signal nonlinear filtration. First of all the kernel estimate of the cumulative distribution function was done. The signal was transformed using the estimate of the cumulative distribution function as a functional transform. Then measurements of acoustic signals' parameters and an estimation of their spectral density were done. The estimation was measured by means of fast Fourier transform procedure with use of window functions. At the second stage the new procedure of the adaptive filtration based on the Wiener frequency approach has been offered. The estimations of spectra received at the first stage have been thus used. Results are confirmed by experimental processing of spoken language signals.

Walsh-Hadamard transform is a transformation that lets to express discrete, real-valued functions as linear combination of discrete Walsh functions. Particularly this transform can be applied to real-valued counterparts of binary Boolean functions and by that we can examine linear properties of Boolean functions. More specifically speaking, it lets us to test probabilities of linear equations for any Boolean function.

The presented paper describes a novel approach to detection of speech corrupted by noise. The proposed procedure is based on fractal dimension, which is being evaluated directly from speech signal samples using two different methods: box-counting and the approach proposed by Katz. The recordings, taken from TIMIT database, were corrupted by five different types of noise (white, pink, hf-channel, babble and factory) with four noise amplitudes (5,10,15,20 dB). The resulting noisy speech was the subject of the analysis. The Otsu's method was used to determine a threshold value for differentiating between noise-only and noisy-speech segments. It has been shown that fractal dimension-based approach provides good basis for detecting speech under a presence of noise.

In this paper, a project of cluster of DSP is presented. This project is realized as a extended card for PC computers. A block diagram of a board was described. A DSP processor properties for cluster computation was described. A block diagram of the board was discussed. Next, a details of hardware realization of this card was described. The PCB project was discussed. A details of multi-layers board was described.

The paper includes description of a novel method of the structural optimization of least squares support vector classifier. The virtual leave-one-out residuals are applied as the criterion for selection of the most influential data. The analytic form of the solution enables to obtain a high gain of the computational cost. The presented method eliminates the drawback of the LS-SVM classifiers - lack of sparseness in the solution. The quality of the method was tested on the artificial data sets - two moons problem and Ripley data set.

Minimum Shift Keying (MSK) is considered to be a spectrally efficient modulation scheme, in comparison with other families of constant envelope modulation. However, its main sidelobe can be of some worry in digital data transmission, particularly over nonlinear channels. Therefore, it is practically of interest to search for efficient pulse shaping schemes, largely proposed either through baseband pulse shaping and/or filtering processing. In this paper a new pulse shaping method based on Walsh functions is proposed to be used in MSK-type signaling. A methodology is proposed for designing the optimum pulse shape; and this attains improvement in spectral efficiency (i.e., bps/Hz) over straight MSK, for the same channel bandwidth. The fractional out-of-band power is used to compare the behavior of the system under the new proposed technique. As well, error rate performance in such system is used in the evaluation of this work. Results reveal that good improvement can be attained in spectral efficiency at the expense of error rate performance when these functions are used.

Hostile wireless environment requires use of sophisticated signal processing methods. The paper concerns on Ultra Wideband (UWB) transmission over Personal Area Networks (PAN) including MB-OFDM specification of physical layer. In presented work the transmission system with OFDM modulation was connected with LDPC encoder/decoder. Additionally the frame and bit error rate (FER and BER) of the system was decreased using results from the LDPC decoder in a kind of turbo equalization algorithm for better channel estimation. Computational block using evolutionary strategy, from genetic algorithms family, was also used in presented system. It was placed after SPA (Sum-Product Algorithm) decoder and is conditionally turned on in the decoding process. The result is increased effectiveness of the whole system, especially lower FER. The system was tested with two types of LDPC codes, depending on type of parity check matrices: randomly generated and constructed deterministically, optimized for practical decoder architecture implemented in the FPGA device.

In this paper the computer implementation of the OFDM modulator and demodulator has been described. The main goal was to create the educational model of OFDM receiver working in acoustic band. The OFDM modulator and demodulator general structure has been discussed and original design has been proposed. The most important parts of the design are symbol synchronization algorithm and QMI / QDD modules adapting raw OFDM signal to the bandlimited channel.

This paper addresses the problem of carrier frequency offset estimation in OFDM systems. Perfect time synchronization and burst transmission are assumed. A preamble consisting of a series of identical blocks is available for the frequency offset estimation. The proposed estimator is based on the correlation of two blocks, which is a commonly used approach. In contrast to existing papers, the author shows here how to adjust the correlation parameters in order to exploit the periodic structure of the preamble and to obtain best estimation results. The degree of accuracy of the estimates obtained using the proposed method is compared with the Cramér-Rao lower bound.

This paper compares different hybrid blind equalization algorithms used for QAM signals. In hybrid equalizers, a penalty term, with zeros at constellation points coordinates, called constellation matching error (CME) is added to the criterion of one of the standard algorithms, such as the constant modulus algorithm (CMA), the multimodulus algorithm (MMA) or the recently proposed extended constant modulus algorithm (ECMA). Among the CMEs, to be considered, we have recently introduced a new one which is the product of l1-norm of the deviations of equalizer output from the constellation points. The hybrid algorithms, obtained by combining different CMEs with the CMA or the ECMA are compared through simulations on 16-QAM, 64-QAM and 256-QAM signals transmitted over different channels.

Considering the growing threat of terrorist or accidental explosions in underground stations, underground highway and railway sections improvement of system for protecting people from explosions appears urgent. Current automatic protective devices with blast identification module and blast damping absorbers of various designs as their basic elements cannot be considered effective. Analysis revealed that low reliability of blast detection and delayed generation of start signal for the activation of an absorber are the major disadvantages of protective devices. Besides the transmission of trigger signal to an energy absorber through cable communication reduces the reliability of the operation of protective device due to a possible damage of electric wiring under blast or mechanical attack. This paper presents the outcomes of the studies conducted to select accurate criteria for blast identification and to design wireless system of activation of defensive device. The results of testing of blast detection methods (seismic, EMP, optical, on overpressure) showed that the proposed method, which implies constant monitoring of overpressure in terms of its reliability and response speed, best meets the requirements. Proposed wireless system for explosions identification and activation of protective device consists of transmitter and receiver modules. Transmitter module contains sensor and microprocessor equipped with blast identification software. Receiver module produces activation signal for operation of absorber. Tests were performed in the underground experimental base of Mining Institute. The time between the moment of receiving signal by the sensor and activation of absorber - 640 microsecond; distance between transmitter and receiver in direct tunnel - at least 150m; in tunnel with 900 bending - 50m. This research is sponsored by NATO's Public Diplomacy Division in the framework of "Science for Peace".

The paper presents the first, to the best knowledge of the author, successful transmission of the video signal beyond of the baseband of the multimode fibre at the 850 nm wavelength using the subcarrier multiplexing technology (SCM). The transmission system has been built on the basis of generally available, inexpensive optical and electronic components.

In many cases one can faced with situation of unknown and unstable statistical models of signals and interferences. It is known that markovian processes give good chance for simulation of random processes with non-Gaussian distribution. In this paper the basic approaches to synthesis of signal detection algorithms are considered under the conditions of parametrical aprioristic uncertainty of both signal and interference situations.

In this paper the approach to estimate reflected signal depolarization due to turbulence with a single unipolarized antenna is considered. The approach allows to make estimate by current amplitude fluctuation in the receiving antenna due to changing the depolarization angle of reflected wave. The frequency characteristics of the sounding wave reflectors under the wind influence are analyzed and taken into account.

Web applications became most popular medium in the Internet. Popularity, easiness of web application frameworks together with careless development results in high number of vulnerabilities and attacks. There are several types of attacks possible because of improper input validation. SQL injection is ability to execute arbitrary SQL queries in a database through an existing application. Cross-site scripting is the vulnerability which allows malicious web users to inject code into the web pages viewed by other users. Cross-Site Request Forgery (CSRF) is an attack that tricks the victim into loading a page that contains malicious request. Web spam in blogs. In order to secure web applications intrusion detection (IDS) and intrusion prevention systems (IPS) are being used. Intrusion detection systems are divided in two groups: misuse detection (traditional IDS) and anomaly detection. Misuse detection systems are signature based, have high accuracy in detecting many kinds of known attacks but cannot detect unknown and emerging attacks. This can be complemented with anomaly based intrusion detection and prevention systems. This paper presents anomaly driven proxy as an IPS and data mining based algorithm which was used to detecting anomalies. The principle of this method is the comparison of the incoming HTTP traffic with a previously built profile that contains a representation of the "normal" or expected web application usage sequence patterns. The frequent sequence patterns are found with GSP algorithm. Some basic tests show that the software catches malicious requests.

The development of accelerator technology in Poland is strictly combined with the cooperation with specialist accelerator centers of global character, where the relevant knowledge is generated, allowing to build big and modern machines. These are relatively costly undertakings of interdisciplinary character. Most of them are financed from the local resources. Only the biggest machines are financed commonly by many nations like: LHC in CERN, ILC in Fermi Lab, E-XFEL in DESY. A similar financing solution has to be implemented in Poland, where a scientific and political campaign is underway on behalf of building two big machines, a Polish Synchrotron in Kraków and a Polish FEL in Swierk. Around these two projects, there are realized a dozen or so smaller ones.

There are described coordinating actions of the accelerator science in Europe in 2003-2009. The actions embrace basic science, as well as development and applications. The accelerator research was not coordinated in Europe at a global scale but was rather concentrated in a few centers owning large infrastructure. These centers include: CERN, DESY, GSI, INFN, LAL, PSI etc. Such coordinating actions enable a lot of positive processes including new possibilities for research centers in this country. It is much easier for them to extend, deepen or even start from the beginning their activities in the field of the accelerator technology. This field includes also free electron lasers. There are described two European framework projects CARE and EuCARD on accelerator technology, their extent and the participation of ISE WUT in them.

This article describes design of a new multichannel acquisition system with photomultiplier tubes (PMTs). The system allows mean current measurement and photon counting for 2 PMT coincident operation. In order to obtain very high dynamic range, signal from the PMT is sampled by 14 bit resolution ADC and analyzed by high speed window discriminators and constant fraction discriminators (CFD). Output data is processed in the FPGA and transmitted to the embedded computer. The system has been developed for astrophysical and high energy physics research.

In this work the energy transfer between Nd³⁺ and Yb³⁺ ions in oxyfluoride silicate glass at room temperature was analyzed. In order to match energy transfer (Nd³⁺ -> Yb³⁺), absorption cross-section of glass doped with various Yb³⁺ ions concentration and emission cross-section of glass doped with Nd3+ were calculated. The percentage efficiency of energy transfer calculated according to the Dextera-Forster model for the glass doped with 0.15Nd³⁺ : 0.75Yb³⁺ amounts to 60%. The emission spectrum with 100 nm bandwidth of the Nd³⁺/Yb³⁺ co-doped glass samples under excitation at 808 nm was measured.

The results of Raman and SEM studies of Pd-carbonaceous films obtained by two methods are presented in this paper. The first method is PVD (Physical Vapour Deposition) with two separated sources containing precursors for films (1 - fullerenes C₆₀, 2- palladium acetate). The second method consists of two steps, the first step being PVD followed by CVD (Chemical Vapour Deposition) - the second step. In the first step a nanocrystalline Pd-C film is prepared. Later this film is used to obtain a nanoporous Pd - carbonaceous film in the CVD process. Prepared films contain 8 - 34 %wt. Pd. Both kinds of films were studied using Raman and SEM methods. For the films obtained in the two-step method SEM images show porous structure for all samples. Raman spectra for the film with 8 % wt. Pd exhibit C60 characteristic bands, while Raman spectra for films with higher content of Pd show D and G graphite characteristic bands. Films obtained by PVD are structurally different from films obtained by 2-steps method which is reflected in SEM images where no porous structure has been found.

This work presents a comparison of numerical simulation and experimental data for nanoindentation studies of nanostructural Me-carbonaceous films (where Me = Pd or Ni). Films containing palladium or nickel nanograins embedded in a carbonaceous matrix were synthesized by a PVD process. The topography of films was characterized by AFM and the mechanical properties were analyzed by nanoindentation. Nanoindentation experiments were performed with a tip made of diamond with a shape of a 90° cube corner three-sided pyramid in a Hysitron Triboscope. The finite element method (FEM) was used to simulate the indentation experiment in such heterogeneous material on nanoscale.

In the paper the phase - locked emitters in multicore optical fibers for high power fiber lasers are presented. The simulation of coupling coefficient between cores influences on phase - locking and in the consequence on the far field pattern of the fiber laser based on the analyzed mutlicore optical fiber were performed. In the analyzed fiber laser, while exchanging power of the generated radiation between two adjacent cores on the fiber length (L=5m) within the range of 4% - 10%, the radiation becomes phased. Having satisfied this condition, in the far -field low-divergence, high-power laser beam (supermode) is attained.

The paper presents an idea and directives on construction of a measurement system for estimation of ions' concentration in water. System presented in paper has been fully designed and manufactured in Warsaw University of Technology in Institute of Electronic Systems. The measurement system works with cheap ion-selective potentiometric sensors. System allows for potentiometric, transient response and voltamperometric measurements. Data fusion method has been implemented in the system to increase the estimation's accuracy. Presented solution contains of many modern electronic elements like 32bit ARM microcontroller, precise operational amplifiers and some hydraulics subsystems essential for chemical measurements.

Mathematical models of chemo-electrical characteristics of ion-sensitive chemical sensors is essential for development of measuring algorithms to be implemented in firmware of water probes. A crucial point of algorithms like these is a data fusion task, where a matrix of sensors sensitive to a matrix of ions give a matrix of electrical responses. Hence unknown ion activities may be calculated if a parameterized mathematical model of sensor is known. During three years of activity in FP6 WARMER project a diversity of physical and behavioral approaches to sensors modeling has been considered. Since a model selection influence accuracy of measurements, then a plenty of different modeling approaches was discussed and practically verified on the base of laboratory measurements of sensors. Some typical measurements have been used and verification of accuracy of existing has been performed. This contribution contains a discussion of possibilities of modeling of potentiometric sensors. Several physical and behavioral models have been taken into account. It has been shown that in data fusion application essential is small number of parameters and small calculation complexity since accuracy of model has not to be better than accuracy of sensors which is rather low. Hence we must focus on a trade-off between necessity of accurate modeling and availability of accurate sensors.

Universal Measurement System with Web Interface (UMSWI) is presented in this paper. This is an autonomous, universal acquisition system which required development of flexible and easily extensible control. The choice of technologies, their flexibility, respect of limited resources and simplicity were the main consideration during design and development. Example implementation of oscilloscope and spectrum analyzer Web Graphic User Interface (WGUI) and Measurement Interface (ME) are described in details.

The analysis of double frequency method for measurement of particles dimensions of solid aerosols (sand, dust) is presented. Also results of numerical simulation of sounding process are represented based on rigorous solution of diffraction problem. Limits of the method applicability and influence of medium polydispersity are estimated.

In this paper we present the State of The Art in Cryptographic Random Number Generators (RNG). We provide analysis of every of the most popular types of RNGs such as linear generators (i.e. congruential, multiple recursive), non-linear generators (i.e. Quadratic, Blum-Blum-Shub) and cryptographic algorithms based (i.e. RSA generator, SHA-1 generator). Finally we choose solutions which are suitable to Distributed Measurement Systems (DMS) specific requirements according to cryptographic security, computational efficiency (throughput) and complexity of implementation (VHDL targeted at FPGA and ASIC devices). Strong asymmetry of computing power and memory capacity is taken into account in both software and hardware solutions.

A noise sounding signal for an acoustic radar (sodar) is offered. The experimental prototype of a noise sodar is developed and made. Algorithms of a signal detection, range and velocity measurements are suggested and developed. The experimental measurements are done. The suggested sodar can be used in meteorology for obtaining information about direction and velocity of the wind.

The paper introduces a novel model for visualization and management of complex hardware architectures. This model settles component oriented environment for both hardware modules and software application. The purpose of the paper is to present such arranged hardware in FPGA chips and its management by object oriented software in JAVA. Requirements and achievements for the model are described.

This paper describes a concept of distributed measurement system, hardware that accomplishes the concept, firmware based on usage of HDL structures from existing libraries or possibilities of using so called "open cores" and user software capable of controlling such system.

Distributed systems became most popular systems in big companies. Nowadays many telecommunications companies want to hold large volumes of data about all customers. Obviously, those data cannot be stored in single database because of many technical difficulties, such as data access efficiency, security reasons, etc. On the other hand there is no need to hold all data in one place, because companies already have dedicated systems to perform specific tasks. In the distributed systems there is a redundancy of data and each system holds only interesting data in appropriate form. Data updated in one system should be also updated in the rest of systems, which hold that data. There are technical problems to update those data in all systems in transactional way. This article is about data anomalies in distributed systems. Avail data anomalies detection methods are shown. Furthermore, a new initial concept of new data anomalies detection methods is described on the last section.

This paper presents new techniques of landmine detection and localization using thermal methods. Described methods use both dynamical and static analysis. The work is based on datasets obtained from the Humanitarian Demining Laboratory of Università La Sapienza di Roma, Italy.

The paper presents a new approach to automatic constructing and training Radial Basis Function (RBF) neural networks based on Differential Evolution (DE) algorithm. The method, called Differential Evolution-Radial Basis Function Network (DE-RBFN) is tested on approximation tasks of exemplary one- and two- dimensional Gaussian functions. Experiments are performed in Matlab environment. The results show that application of DE-RBFN enables to obtain optimal sparse network architecture by tuning the position and width of each basis function. The performance of the method is better than other related procedures applied to RBF networks.

Age-related Macular Degeneration (AMD), according to epidemiological data, is a main reason of social blindness among elderly people in developed countries. There are two forms of AMD: dry and wet. The first one is of good prognosis with low possibility of serious visual deterioration, while the second one usually leads to quick and severe visual impairment. The aim of our investigations is to analyse results of so called real-life treatment of wet AMD. We analysed outcomes of our patients treated with intravitreal injections of anti-VEGF drugs: Lucentis (61 patients) and Avastin (78 patients). We analysed changes in visual acuity (functional effect) and central retinal thickness (anatomic effect). Both drugs occurred to be efficient in treatment of wet form of AMD, however results were more satisfying in patients with better baseline visual acuity. In our approach we used R environment - an integrated suite of software facilities for data analysis and graphics.

In the paper some results of research on discomfort glare measurement are described. The idea of practical realization of this measurement in interiors on the basis of the analysis of the investigated room image is presented. To obtain the necessary geometric quantities the coordinates of proper pixels in the image and some parameters of the imaging system should be known. The luminance values are extracted after the analysis of the brightness distribution of calibrated image, which is registered by the pinhole digital camera. The most important properties of the camera, such as spectral sensitivity, linearity, noise and the influence of exposition parameters on the pixel brightness were investigated. After correcting the spectral sensitivity experiments in the laboratory and in the didactic room were conducted. In the laboratory test the error of the UGR measurement was lower - it reached 6,9%. Some possibilities of improving the accuracy of measurement are indicated.

The aim of this article is to present an application of Bayesian networks for discovery of affinity relationships based on genetic data. The presented solution uses a search and score algorithm to discover the Bayesian network structure which best fits the data i.e. the alleles of single nucleotide polymorphisms detected by DNA microarrays. The algorithm perceives structure learning as a combinatorial optimization problem. It is a randomized local search algorithm, which uses a Bayesian-Dirichlet scoring function. The algorithm's testing procedure encompasses tests on synthetic data, generated from given Bayesian networks by a forward sampling procedure as well as tests on real-world genetic data. The comparison of Bayesian networks generated by the application and the genetic evidence data confirms the usability of the presented methods.

The purpose of this paper is to figure out a new, better than so-far-known, evaluation method of diagnostic tests in Wilson disease. In order to find the most interesting classification models various data mining techniques were applied to real, suffering from Wilson disease, set of patients. It occurred that a combination of two classification algorithms with its implementations in Weka environment may significantly increase classification ability.

The paper describes an innovative approach to discover new knowledge in non-alcoholic fatty liver disease (NAFLD). In order to determine the factors that may cause the disease a number of classification and attribute selection algorithms have been applied. Only those with the best classification results were chosen. Several interesting facts associated with this unclear disease have been discovered. All data mining computations were made in Weka environment.

Detection and quantitative parameterization of brain blood vessels in magnetic resonance images (MRI) are an important aid to diagnosing neoplasmic diseases, planning surgical operations or detecting the atrophy of blood vessels. Fast and effective computer programs are needed to extract quantitative information from MRI data - to increase objectivity, accuracy and repeatability of the diagnosis. To develop such programs we must use algorithms for 3D images segmentation, necessary to build geometrical models of the blood vessels. These models are then used for vessel tree visualization and quantitative description.

In this article there is presented a proposal for the measurement system that provides a direct determination of the value of the absorption coefficient of radiation by a very thin sample - cell cultures. The absorption coefficient can be obtained by applying the measured thickness of biological sample for resolved Bouguer's law. A fundamental understanding of how lasers interact with tissue and knowledge of its optical properties will enable to choose the most appropriate laser for a clinical treatment.

The slowing pace of performance improvement in the commonly available processors is a cause of concern amongst many computational scientists. This combined with the ever increasing need for computational power has caused us to turn to alternative architectures in search of performance gains. Two main candidates were the Compute Unified Device Architecture (CUDA) and the Cell Broadband Engine (CELL BE) architecture. This paper focuses on the latter, outlining the architecture and basic programming paradigms, and also contains performance comparison of algorithms currently developed by our team.

Exploration of database of infants with symptomatic epilepsy is presented. The R environment is used for 67 attributes analysis in three groups of infants with epilepsy, different diagnostic results, and different therapeutic procedures . The estimation of the birth weight, time of labor, and Apgar score allows to eliminate these perinatal factors as the reasons for epilepsy. It is also presented, by using R environment, that postinflamatory viral brain damage and microcephaly is strongly connected with severe epilepsy, and, therefore, polytherapy with antiepileptic drugs is necessary.

Electric bioimpedance is one of methods to assess the hydrate status in hemodialyzed patients. It is also being used for assessing the hydration level among peritoneal dialysed patients, diagnosed with neoplastic diseases, patients after organ transplantations and the ones infected with HIV virus. During measurements sets were obtained from two groups, which were named a control (healthy volunteers) and test group (hemodialyzed patients). Zscored, discretized data and data retrieval results were computed in R language environment in order to find a simple rule for recognizing health problems. The executed experiments affirm possibilities of creating good classifiers for detecting a proper patient with the help of medical data sets, but only with previous training.

Blood pressure in childhood and adolescents is important indicator of good health and strong predictor of BP in adulthood. Genetic susceptibility, environmental and socioeconomic factors are related both with life style, obesity and cardiovascular risk including elevated BP. Increased body mass index is strictly correlated with BP, and obesity and overweight is main intermediate phenotype of childhood hypertension. However, despite current obesity epidemic available data do not fully support the hypothesis that it has resulted in increase of BP in children. We analysed data obtained from 7591 children participating in nation-wide health survey using data mining methodology. Results reveal relationships of obesity and high blood pressure with school environment characteristics.

Nowadays computers successfully analyze medical data giving results used for futher treatment. Every year we develop new technology which gives us better and more precise diagnose. We chose esophageal manometry (EFT) which has been considered as a "gold standard" test for the evaluation of esophageal motility. EFT allows physicians to get informations about esophageal peristalsis, amplitude and duration of the esophageal contraction and liquid/viscous bolus transit time from mouth through stomach. We examined 80 patients during 2008 year. Everybody got EFT, endoscopy and X-Ray examination. It was important to ask about symptoms which we correlate and connect with data from EFT. We tried to find a good algorithm for this job in order to do a simple and helpful tool for physician to make right diagnose. Connection between data and symptoms seems to be right and clear, but finding a good algorithm for given data is the main problem.

This paper presents the application of differential evolution, an evolutionary algorithm of solving a single objective optimization problem - tuning the hiperparameters of least-square support vector machine classifier. The goal was to improve the classification of patients with sustained ventricular tachycardia after myocardial infarction based on a signal-averaged electrocardiography dataset received from the Medical University of Warsaw. The applied method attained a classification rate of 96% of the SVT+ group.