Ing. Martin Mudroch, Ph.D.

Urban catchments are characterized by diverse land cover with large ratio of impervious surfaces on which rainfall-runoff is generated extremely fast. This possess high requirements on spatio-temporal resolution of rainfall data needed for reliable runoff predictions. The contribution summarizes results from three-years experiment evaluating reliability of rainfall-runoff simulations at small urban catchment using rainfall observations from Commercial Microwave Links (CMLs). In addition, system for providing real-time CML-based rainfall product at city scale is presented. Finally, sustainability of the CML-based rainfall observation system is discussed with focus on challenges related to the fast development of communication networks.

Projekt Tel4Rain řeší vývoj aplikace poskytující online data o srážkových intenzitách se zaměřením na hydrologii městských aglomerací s prostorovým rozlišením 500x500 m2 a časovým rozlišením 1 minuta. Projekt počítá s využitím přibližně 1000 mikrovlnných spojů na území hl.m. Prahy a okolí z telekomunikační sítě společnosti T-Mobile ČR. Data o útlumu signálu jsou získávána z páteřní sítě mikrovlnných spojů společnosti v reálném čase. Surová data jsou ukládána do PostgreSQL databáze a v hydrologickém modulu jsou přepočtena na srážkové intenzity. Modul prostorové rekonstrukce zahrnuje algoritmy, které informace z liniových spojů převádějí do pravidelné souřadnicové sítě s definovaným prostorovým rozlišením.

This paper describes FSO link performance prediction based on available meteorological data using different Artificial Neural Network (ANN) approaches. Several types of ANNs were compared and their performance were evaluated. The paper introduces an ANN application utilizing real delayed data. This approach has been validated to be more precise than common feed-forward neural networks.

Free-space optical link is an alternative instrument to transmission of information in data rates up to tenths of gigabits per second. Although evolution of this technology has been proceeded for several years, many factors affecting propagation of optical beam still have not been suitable minimized. Therefore several mitigation techniques are tested nowadays. In this paper, influences of atmosphere and rain are discussed with respect to particular diversity techniques. First results from a measurement campaign are introduced.

New communication systems are emerging with growing demand on the data traffic. Fiber optics allow us to transfer data of rates higher than hundreds gigabits per second. These systems are usually used for backbone networks, although using of these systems is appearing in last mile connections. Wireless communication systems are still more and more popular owing to fast and cheaper deployment contrary to wire or fiber communication systems. On the other hand, wireless communication systems including free-space optics (FSO) are affected by weather conditions. In order to mitigate these effects, several mitigation techniques were investigated. For instance spatial diversity using more than one transmitter or receiver, auto-tracking of an optical beam, wavelength diversity utilizing radio band links as a backup etc. FSO links may be used in FSO networks to connect several building of a corporation, hospital or university campus. In case of proper network topology, route diversity can contribute to availability enhancement. In this paper, mitigation technique based on the route diversity is presented. Results are compared with hybrid FSO link.

Free-space optical (FSO) links are perspective wireless optical communication systems utilizing modulated laser radiation to transmission of information. Because of propagation of the light through an atmosphere in case of a terrestrial link, strong dependence on weather conditions is observed. Several mitigation techniques were investigated, for an instance beam tracking, spatial diversity in multiple-input multiple-output systems and a wavelength diversity technique. In this paper, performance of FSO and radio-frequency (RF) hybrid link working at the wavelength of 850 nm and 5 GHz, respectively, is analyzed in specific urban area.

This paper introduces an artificial neural network model created for weather influence evaluation of FSO link. Basic neural cluster analysis is performed and neural network model creation is discussed.

This paper describes a new progress in evaluating measured data from a unique, experimental, multiple-receiver terrestrial radio link used for remote sensing. The refractivity index height profile of the lowest troposphere layers is evaluated and the classification process is discussed. We are comparing different sizes and architectures of feed-forward artificial neural networks and their learning parameters.

Multipath propagation events are observed on the 50 km long fixed terrestrial link operating in 10 GHz band. Several receivers are located in different heights. Two year statistics of attenuation and enhancement due to multipath are obtained from measured signal fluctuations in different heights. The results are compared with cumulative distributions predicted by the two methods - the ITU-R method and the VUS method which is currently used for the planning of radio-relay links in the Czech Republic. The predicted distributions of attenuation provide slightly higher exceedance probabilities than measured.

This paper deals with the minimization of the back radiation of the Vivaldi antenna for the radar application (e.g. portable through-wall radars). The high back radiation can give rise to the reflection from the radar operator or from the area behind the radar. These reflections can influence negatively the captured useful signal. Consequently, the measured view of the guarded area can be distorted or the undesired reflections can lead to a false alarm. In the paper, several types of shielding for the back radiation minimization are presented: an antenna with planar back-plane, a boxed antenna, a skewed and boxed antenna and a skewed and boxed antenna with planar frontplane. The reflections, radiated impulses and their distortion are discussed.

This work describes the basic steps of a novel approach to weather classification by remote and local atmosphere sensing. Atmospheric data on the troposphere are gathered by 10 GHz radio links and a number of meteorological sensors. Classification is performed by artificial neural networks (ANN) and is crucial for further processing, because of the different RF propagation influences under a variety of weather conditions. Reasons for using ANN compared to other means of classification are discussed. Differences in the size and number of hidden layers of back-propagation networks used are discussed. Different learning sets of measured data and their construction are also evaluated.

This document presents first results from remote sensing of atmosphere using data gathered within an unique experiment combined with artificial neural network data processing. Extended methods of data evaluation are presented and its suitability for refractivity height profiles estimations is also shown.

In any UWB application, it is crucial to ensure the minimal distortion of the transmitted impulses. On the other hand, some impulse generators do not produce required impulse shape suitable for the antenna excitation. This paper describes the optimization process of the antenna, which distorts/forms optimally the excitation impulse to the required and radiated impulse shapes. The particle swarm optimization method is used.

This paper describes an improved approach in an optimization of different UWB dipole antennas. The new procedure of tuning different optimization algorithms is presented. Fine-tuning of optimizing algorithm which is not elementary problem is discussed.

This paper introduces an extensive innovative experiment for both local and remote combined sensing of the troposphere. The experiment is based on a unique set of mutually linked, time-space dependent, synchronized radio-electrical and radio-meteorological measurements. The results for the first two months of operation are also presented and discussed.

The principal aim of the paper is to introduce results of measurements methods performed in the frequency domain as well as in the time domain. The aforementioned methods are based on the measured data processing that is carried out by means of the measured transmission coefficient between two antennas. Moreover, the additional aim of the paper is to present the differences between various and several approaches (in both domains).

This paper describes the special EMC issue which emerges during shielding effectiveness measurement of relatively (compare to the wavelength) small enclosures. The main aim of this paper is to show difference of the shielding effectiveness values in dependence on different types of used measuring linearly and circularly polarized antennas.