Past projects

The aim of this project is to design and build a machine learning system that is networked across different levels, from sensors to the cloud, and optimised as a whole. The advantages of such a system can be optimally demonstrated by using the latest radar sensor technology. For this purpose, novel ML signal processing algorithms for person recognition are developed in order to realise high-resolution environment detection for autonomous transport vehicles. The focus for the system should be on modularity, reusability, flexibility and scalability, as well as the closest possible interlocking of the subcomponents.

Passive radar technology represents a promising addition to conventionalradar systems. With increasing demands from economy and politics to completelyuse the limited spectrum of the frequency bands limited for telecommunicationsand location, the interest in this technology is increasing.

The aim of this research project is to establish the technology of locationusing passive radar technology in civil air traffic control in Germany and to opennew areas of application.

To improve the detection performance, various options for setting up afrequency-selective analog receiver for the FM band are being developed andintegrated into an existing passive radar system. For the highest possiblesensitivity, filtering in different stages of the receiver is essential.However, this must be evaluated together with the frequency-converting stagesin the overall system context in order not to degrade the signal quality,including through possible imperfections in the analog implementation. Furthermore,attention must also be paid to an optimal balance between circuit complexity,costs and compactness of the receiver. For this purpose, the receiverarchitectures are first examined in system simulations and evaluated regardingthe requirements from the application. This is followed by a prototypeconstruction of the most promising concepts with metrological verification ofthe individual components and evaluation of the entire system in a field test.

The fundamental goal of the ANDANTE project is to leverage innovative hardware platforms to build strong hardware / software platforms for artificial neural networks (ANN) and spiking neural networks (SNN) as a basis for future products in the Edge IoT domain, combining extreme power efficiency with robust neuromorphic computing capabilities and demonstrate them in key application areas.

The main objective of ANDANTE is to build and expand the European eco-system around the definition, development, production and application of neuromorphic hardware through an efficient cross-fertilization between major European foundries, chip design, system houses, application companies and research partners, as presented by the European Leader Group (ELG). 

In people with chronic diseases, sensors are increasingly being used in or on the body to monitor the patient's state of health and detect deterioration at an early stage. Which data are collected in this context is initially a medical-technical question. However, with the increasing prevalence of mobile data collection in everyday life and the usability of this data by different interest groups, it is clear that this is also an ethical question about data sovereignty. Therefore, it will be investigated how guidelines for the design of implantable or wearable medical sensor systems can be developed so that diagnostic, technical and also ethical requirements are met. The question has two objectives: on the one hand, to derive concrete instructions for medical engineers; on the other hand, the methodological question of the extent to which the different requirements can be weighed up against each other at all: Since incommensurability of norms can already occur within a closed ethical approach (e.g., Principlism), this is to be expected even more strongly for the requirements from different disciplines.

The ever-increasing number of agile Internet users and the concomitant growth in data volumes, driven in particular by the use of mobile Internet, video and cloud streaming services ("streaming on demand"), are already causing bandwidth bottlenecks in existing data and mobile communication systems. The MassiveData6G project aims to address the emerging bandwidth constraints in infrastructure to provide at least 100 Gbps per mobile user in the future. The required low-power and low-cost 140 GHz transceiver uses a MIMO architecture with at least 5 GHz signal bandwidth and high spectral efficiency (512/1024 QAM signal modulation). In addition, to address the mass market, this project uses a low cost, low power 22-nanometre FDSOI (Fully Depleted Silicon On Insulator) CMOS technology (22FDX), which not only allows a high performance implementation of the digital signal processing components, but is also perfectly suited for the 140 GHz RF components.

Alternating-Contactthin-film transistors (ACTFTs) provide new degrees of freedom for deviceoptimization and deployment. This project specifically aims at providingcost-effective implementation of flexible RF circuits through the use of shortchannel ACTFTs with self-aligned contacts. With the Chair of Electron Devicesand the Institute of Electronics Engineering of the FAU Erlangen-Nuremberg, tworenowned institutes of semiconductor electronics and RF circuit technology workhand in hand on the integrated development of RF circuits and systems. Based onmetal oxide ACTFTs, key components of receivers and transmitters (e.g. lownoise amplifiers, oscillators, or mixers) are implemented on flexiblesubstrates. New perspectives for thin, flexible applications in industrial,consumer and textile / wearable electronics are presented.

Millimeter-wave radars are insensitive to the environment and, because of that, essential in automatic imaging like gesture recognition. Unlike the usual frequency-modulated continuous-wave (FMCW) waveform, phase-modulated continuous-wave (PMCW) radars use binary phase-shift keying (BPSK) modulated signals that are digitally processed in the receiver. However, as their range resolution depends on the bandwidth, higher frequency bands must be used for the desired application. In the REGGAE project, the LTE will develop an integrated PMCW radar transmitter operating in the D-band at a center frequency of 140 GHz and a bandwidth of 25 GHz. The circuits will be realized in an advanced 22 nm fully-depleted silicon on insulator (FDSOI) technology that exhibits state-of-the-art millimeter-wave performance combined with competitive digital cells. In cooperation with our project partners from KIT and TUD, we will realize a complete demonstrator featuring four transmit, and eight receive channels capable of recognizing hand gestures.

Am 1. Mai 2018 startete das Forschungs- und Innovationsprojekt PRYSTINE, unter gemeinsamer Finanzierung der Europäischen Union durch ECSEL und den nationalen Regierungen der ECSEL-Mitgliedstaaten. Der Lehrstuhl für Technische Elektronik repräsentiert im Konsortium von über 50 europäischen Partnern die FAU.
Unter den tatsächlichen Trends, die die Gesellschaft in den kommenden Jahren beeinflussen werden, zeichnet sich das autonome Fahren insbesondere durch das Potenzial aus, die Automobilindustrie, wie wir sie heute kennen, zu verändern. In der Folge wird dies auch die Halbleiterindustrie stark beeinflussen und neue Marktchancen eröffnen, da Halbleiter als „Enabler“ für autonome Fahrzeuge eine unverzichtbare Rolle spielen. Autonomes Fahren wurde als eine der wichtigsten Voraussetzungen für die Bewältigung der gesellschaftlichen Herausforderungen einer sicheren, sauberen und effizienten Mobilität identifiziert. Dazu ist ein ausfallsicheres Verhalten unerlässlich, um sicherheitskritische Situationen aus eigener Kraft zu bewältigen. Dies wird mit heutigen Ansätzen auch aufgrund fehlender zuverlässiger Umgebungswahrnehmung und unzureichender Sensorfusion nicht erreicht.
Im Projekt mit dem Titel „Programmable Systems for Intelligence in Automobiles“ (PRYSTINE) geht es im Allgemeinen darum, eine robuste und ausfallsichere rundum Wahrnehmung der Umgebung von Fahrzeugen zu realisieren. Mittels robuster Sensordatenfusion von Radar-, LiDAR- und Kameradaten, sowie ausfallsicheren Steuerungsfunktionen, soll möglichst sicheres autonomes Fahren in städtischer und ländlicher Umgebung ermöglicht werden.
Am Lehrstuhl für Technische Elektronik soll im Rahmen von PRYSTINE eine robuste Umwelterfassung und Bildgebung mittels MIMO Radarsensoren erfolgen. Hierbei sollen auch unterschiedliche Einflüsse und Szenarien, wie zum Beispiel Funkinterferenzen oder die Detektion im Nahfeld für Automobilradare betrachtet werden. Des Weiteren sollen Teile der traditionellen Radarsignalverarbeitungskette, von der Interferenzreduktion, bis hin zu Detektion, Klassifikation und Tracking von Verkehrsteilnehmern, schrittweise durch maschinelles Lernen ersetzt werden.
Vollständige Informationen über dieses Projekt finden Sie auf der offiziellen Website: www.prystine.eu

Motivation

In the care of seriously ill people, the recording of breathing and heartbeat is an important tool for crisis detection. The recording via electrodes and cables that has been necessary up to now is prone to interference and restricts the self-determination and quality of life of those in need of care. The GUARDIAN project aims to enable the contactless and continuous recording of vital parameters.

Goal and strategy

In GUARDIAN, the contactless recording of vital parameters from a distance of several meters using a multimodal high-frequency sensor is being developed. For this purpose, a weak electromagnetic high-frequency signal is emitted and its change is analyzed. Due to the high distance resolution, movements causing respiration and heartbeat can be extracted from the measurement signal and analyzed. In the process, superimposed motion artifacts must be compensated for. GUARDIAN will thus make it possible to detect complaints such as pain and shortness of breath as well as health crises such as cardiac arrhythmias and cardiovascular arrest immediately and automatically. At the same time, the ethical, legal and social issues of the procedure as well as its effects on palliative and intensive care, people in need of care, care professionals and relatives will be intensively investigated.

Innovation and perspective

By using six-port interferometry as a new concept, all body movements can be recorded contactlessly from a distance of up to several meters with a previously unattainable distance resolution in the micrometer range, and respiration and heartbeat can be extracted. The consortium partners see great potential in the technology to be developed for monitoring the health and complaints of people in need of care in hospitals, but also in the outpatient sector in nursing homes and at home.

This project Mobile-BAT will investigate methods for automated tracking of migration routes of bats based on miniaturized sensing module for low power passive cellular network detection. This module will be mounted to the dorsum of the migrant Noctule Bats and will track the route over one total season with an accuracy which will enable to draw conclusions to migrant and rout selection strategies of the animal. For limiting as far as possible the mobility restrictions of the bat and disturbance of the natural behavior by the localization module, the sensor node has to exhibit a weight below 2 gram including battery, circuitry as well as antenna system and feature a suitable form factor. Due to the required operation time for covering one entire migration period of up to six months concepts have to be found for ensuring the localization from the limited energy resources. The data logger will be manually retrieved and the stored data evaluated after the return of the animal to the initial habitat. The locating of the respective individuals will be supported by automated direction finding and triangulation of a specific low power VHF telemetry signal transmitted by the sensor node after detection of the return to the initial habitat. From the logged cellular base station parameters, the trajectories of the routes chosen have to be extracted. For this goal topographic information will be added to propagation models for mobile phone signals enabling an automated calculation of an estimated signal constellation for any arbitrary coordinate. The stored sensor node data will be mapped to this database and will result in a highly accurate migration route trajectory. The benefit of the proposed methodology is that due to the passive system approach the sensor node is not required to loggin to a certain provider and is therefore able to logg all receivable base station signals in any frequency bands. This will lead to a large amount of analyzed base stations and therefore to calculated position with high accuracy.The project Mobile-BAT will enable fundamental insight into migration strategies of bats. Besides this concrete application it is expeted that the cellular based self-localization will lead to essential findings in the context of wireless sensor networks and internet of things applicable in nearly any country of the world.

Reliability and radiation damage issues have a long and important history in the domain of satellites and space missions. Qualification standards were established and expertise was built up in space agencies (ESA), supporting institutes and organizations (CNES, DLR, etc.) as well as universities and specialized companies. During recent years, radiation concerns are gaining attention also in aviation, automotive, medical and other industrial sectors due to the growing ubiquit…

In state of the art thin-film-transistors (TFTs), both source and drain electrodes are placed at the same side or interface of the semiconductor layer. Positioning the two contacts on opposite interfaces of the semiconductor in an Alternating Contact TFT (ACTFT) enables new degrees of freedom for device design, optimization, and operation. The ability to enable short channel lengths is explored for application in radio frequency (RF) circuitry in this project.Two research groups of FAU Erlangen Nuremberg being experts in device technology (Chair of Electron Devices) and RF circuits engineering (Institute of Electronics Engineering) join forces to cover the integrated development of ACTFTs towards basic RF building blocks and systems based on flexible metal oxide TFTs. Studies on device physics, RF behavior, and novel circuit concepts will open perspectives for the use of large area, thin, and bendable TFT technologies in future industrial, consumer, and wearable electronics.

Zur Erforschung des Verhaltens von Fledermäusen soll im Projekt ein Sensorsystem entworfen werden. Diese Sensoren müssen auf der Fledermaus angebracht werden, um die Fledermaus im Flug zu orten. Damit sie unbeeinträchtigt ist, muss der Sensorknoten leicht und sehr kompakt sein. In dem hier vorgestellten Teilprojekt soll die Modul-Integration der miniaturisierten drahtlosen Sensorknoten mit Ortungsfunktionalität erfolgen. Für den avisierten Einsatz auf einer fliegenden Fledermaus sind dabei die wichtigsten Randbedingungen ein minimales Gesamtgewicht (max. 2 Gramm inklusive Batterie, Schaltungsträger und Antenne) und ein Formfaktor, der die Fledermaus in ihren natürlichen Bewegungen nicht einschränkt. Für dieses Teilprojekt stellen diese beiden Vorgaben eine große Herausforderung an den Entwurf einer Multiband-Antennenlösung dar, die sowohl in ihrer Geometrie stark verkürzt als auch dreidimensional an den Körper der Fledermaus anzupassen ist. Auch die Aerodynamik muss hierbei berücksichtigt werden. Neben einer Ortungsfunktionalität, die durch Integration des in TP 8 entworfenen Ortungs-ICs realisiert wird, soll auch eine Kommunikation zwischen verschiedenen Sensorknoten möglich sein. Um die Lebensdauer der eingesetzten Batterie zu maximieren und das zu entwerfende Energiemanagement des Moduls zu entlasten sollen energieeffiziente Übertragungsprotokolle untersucht werden. Durch die Staffelung der Arbeitspakete wird nach einer Realisierung der Grundfunktionalität im ersten Schritt die Komplexität des mobilen Sensorknotens durch Hinzunahme weiterer Funktionen nach und nach erhöht und gipfelt zum Projektende in einem leichten und miniaturisierten drahtlosen Sensorknoten mit Lokalisierungs- und Kommunikationsschnittstelle für den Einsatz auf einer fliegenden Fledermaus.

Teilprojekt 2 beschäftigt sich mit der multiphysikalischen Modellentwicklung und Optimierung mikroakustischer MEMS‐Komponenten. Dabei werden die Schwerpunkte auf die Charakterisierung und Simulation verschiedener BAW‐Komponenten gelegt. Je nach Einsatz der jeweiligen Komponenten in dem im Rahmen der gesamten Forschergruppe zu entwerfenden mikroelektromechanischen Frontend wird der Fokus vor allem auf die Leistungsverträglichkeit, das Temperaturverhalten und die Analyse von Nichtlinearitäten gerichtet, da starke Temperatureinflüsse und hohe Leistungen zu unerwünschten Frequenzverschiebungen, Schädigungen und Alterung der Bauelemente führen. Parallel dazu werden Schnittstellen mit den anderen Teilprojekten der Forschergruppe MUSIK identifiziert und entwickelt, um elektrische und thermische Wechselwirkungen zwischen den Bauelementgruppen berücksichtigen sowie die komplementären Modellansätze zu einer ganzheitlichen und durchgängigen Modellierung der resultierenden HF‐MEMS‐Schaltung zusammenführen zu können. Aus der Bauteilanalyse gewonnene Daten führen zu Modellen für die Beschreibung des temperaturabhängigen übertragungsverhaltens, welche anschließend bei der Optimierung des Entwurfs eingesetzt werden. In einem weiteren Schritt wird die thermische Interaktion zwischen wichtigen Komponenten des MEMS‐Funksystems wie Oszillatoren, Leistungsverstärkern und passiven Komponenten erforscht.

 

In the "Smart Sensors B" research project, work is being done within the framework of the Medical Valley Leading Edge Cluster on a high-frequency-based sensor node for non-invasive measurement of blood parameters. The electrical properties undergo a characteristic change depending on the concentration. These changes can be detected by non-invasive means using integrated high-frequency circuits, whereby costs are getting lower and lower all the time. In future, this could enable portable, automatable long-term measurement of various blood parameters.