Przemyslaw Korpas

This paper discusses the use of IoT sensor networks and spatial data mining methods to support the design process in the revitalization of the university campus of the Warsaw University of Technology (WUT) in the spirit of universal design. The aim of the research was to develop a methodology for the use of IoT and edge computing for the acquisition of spatial knowledge based on spatial big data, as well as for the development of an open (geo)information society that shares the responsibility for the process of shaping the spaces of smart cities. The purpose of the article is to verify the hypothesis on whether it is possible to obtain spatial–temporal quantitative data that are useful in the process of designing the space of a university campus using low-cost Internet of Things sensors, i.e., already existing networks of CCTV cameras supported by simple installed beam-crossing sensors. The methodological approach proposed in the article combines two main areas—the use of IT technologies (IoT, big data, spatial data mining) and data-driven design based on analysis of urban space actors’ behavior for participatory revitalization of a university campus. The research method applied involves placing a network of locally communicating heterogeneous IoT sensors in the space of a campus. These sensors collect data on the behavior of urban space actors: people and vehicles. The data collected and the knowledge gained from its analysis are used to discuss the shape of the campus space. The testbed of the developed methodology was the central campus of the WUT (Warsaw University of Technology), which made it possible to analyze the time-varying use of the selected campus spaces and to identify the premises for the revitalization project in accordance with contemporary trends in the design of the space of HEIs (higher education institutions), as well as the needs of the academic community and the residents of the capital. The results are used not only to optimize the process of redesigning the WUT campus, but also to support the process of discussion and activation of the community in the development of deliberative democracy and participatory shaping of space in general.

The ability of microwaves to heat various kinds of substances has been known for decades [1] . Today, this phenomenon is used in many technological and scientific fields, such as food processing (both in home-use microwave ovens and large industrial installations) [2] , wood drying [3] , medicine (microwave ablation treatment) [4] , and waste recycling (pyrolysis) [5] , just to name a few. Opportunities for new applications are emerging with technological advances allowing for increasingly precise manipulation of microwave signals.

This paper presents a model-based PID tuning method for a reactor used in microwave-assisted chemistry. The reactor is equipped with a solid-state source of microwaves and a PID controller capable of increasing or decreasing the delivered microwave power to maintain the reacting substances at the desired temperature. The model has the form of an algorithm applied in numerical simulations of two simultaneous processes: heating a substance by absorbing microwave radiation and cooling it by dissipating heat to the surroundings. It has proven its suitability for tuning the PID controller in a time-efficient manner. Despite some noticeable inaccuracies, the presented approach easily finds PID coefficients that result in stable and repeatable controller operation. In this way, significant time savings can be achieved, as well as minimizing the risks associated with, for example, boiling liquid spills. The article demonstrates that a carefully designed, but still relatively simple, model can yield significant benefits in tuning PID controllers.

This paper presents a universal hardware platform (UHP) based on software defined radio (SDR) technology and developed to accelerate the prototyping and testing new hardware and algorithmic solutions for modern radio systems. There are two integrated transceivers and a baseband processor on the board. Thus, four complete RF transmitting and receiving channels, with observation paths for transmitted signals monitoring could be supported. Our platform can be operated in both time-division (TDD) and frequency-division (FDD) duplex schemes over a 300 MHz to 6 GHz frequency range with simultaneous synthesis and analysis in the bandwidth of up to 100 MHz. The radio system is formed by attaching an appropriate RF front-end to the UHP and uploading standard-specific software. As examples of possible UHP applications, two LTE-Advanced micro base stations for private networks with standard-defined RF front-ends operating in the LTE band B43 (3.6 GHz) with time duplex and the band B31 (450 MHz) in FDD mode are presented.

This paper is related to the project of an ultrawideband radio receiver operating between 1 and 6 GHz with instantaneous bandwidth of 500 MHz. A single frequency conversion (0-IF) architecture is based on analog I/Q demodulator (ADL5380) and two channel 14-bit analog-to-digital converter (AD9680). Preliminary measurements of the receiver have been performed and presented in the paper. Key results include: noise figure NF < 7.5 dB, dynamic range over 62 dB (enhanced by frontend and baseband leveling circuitry), high linearity: IIP3 > 21 dBm.

In this paper the performance of two wideband PLL frequency synthesizers realized with ADF5355 and HMC833 chips respectively has been compared. Both of them were investigated as a local oscillator for 0IF receiver operating from 1 GHz to 6 GHz. In order to provide required harmonics level below -40 dBc the final solution is based on the ADF5355 followed by a switched bank of four low pass filters. The phase noise level is better than -90 dBc @ 1 kHz and spurious are lower than -70 dBc.

This paper presents a new type of an electromagnetic sensor for nondestructive testing of carbon-fiber-reinforced polymer composites. The sensor utilizes coupled planar spiral inductors operating typically in the range of 10-500 MHz. The method proposed here shows some similarity to the eddy current technique, but as will be shown, the principles of operation are different as the sensitivity to defects is mostly due to the magnetic field components tangential to the surface of a material. It is shown that the method is applicable to 3-D inspection of carbon-fiber-reinforced composites widely employed in the aerospace industry.