Forschung / Research

Ergebnisse des Projekts Edge-Power als Video-Präsentation verfügbar
Results of the Edge-Power project are available as video presentation

Auf der Fachkonferenz „Edge-Computing 2025: Von der Forschung zur Anwendung“ präsentierte das Edge-Power Konsortium seine gemeinsam erzielten Projektergebnisse. Diese Video-Präsentation ist auch öffentlich verfügbar bei Youtube.

Der Beitrag des Lehrstuhls für Messtechnik waren interpretierbare und energieeffiziente ML-Algorithmen, wobei unser FESC/R (Feature Extraction, Selection and Classification/Regression) Ansatz als tiefe neuronale Netzwerke repräsentiert wurde, um besonders effizient auf AI-Beschleunigern ausgeführt zu werden (im Video dargestellt ab Minute 10:00). Wir danken allen Partnern für die hervorragende Zusammenarbeit.

At the conference “Edge Computing 2025: From Research to Application,” the Edge-Power Consortium presented its joint project results. This video presentation is also publicly available on YouTube.

The Chair of Measurement Technology contributed interpretable and energy-efficient machine learning algorithms, with our FESC/R (Feature Extraction, Selection and Classification/Regression) approach represented as deep neural networks for particularly efficient execution on AI accelerators (presented in the video starting at 10:00). We thank all partners for the excellent collaboration over the course of this project.

Weitere Veröffentlichungen aus dem Edge-Power Projekt / further publications from the Edge-Power project:

• J. Schauer, P. Goodarzi, J. Morsch, A. Schütze: A Performance Study of Deep Neural Network Representations of Interpretable ML on Edge Devices with AI Accelerators, Sensors 2025, 25(18), 5681, doi: 10.3390/s25185681
• J. Schauer, P. Goodarzi, A. Schütze, T. Schneider: Efficient hardware implementation of interpretable machine learning based on deep neural network representations for sensor data processing, J. Sens. Sens. Syst., 14, 169–185, doi: 10.5194/jsss-14-169-2025

Das Projekt Edge-Power wurde gefördert vom Bundesministerium für Forschung, Technologie und Raumfahrt (BMFTR) in der Fördermaßnahme „Elektroniksysteme für vertrauenswürdige und energieeffiziente dezentrale Datenverarbeitung im Edge-Computing (OCTOPUS)“.

The Edge-Power project was funded by the Federal Ministry for Research, Technology and Space (BMFTR) in the funding measure “Electronic systems for trustworthy and energy-efficient decentralized data processing in edge computing (OCTOPUS)”.

From September 14 to 19, the IABR Breath Summit will be held in Innsbruck, Austria.

LMT will present two contributions at the Breath Summit:

• C. Bur, W. Reimringer, T. Maus, F. Maurer, S. Kreuer:
  Simultaneous Analysis of Breath Condensate and Exhaled Breath in an Isolated Porcine Lung Model
  Oral presentation, session "inorganics and non-volatiles", Wednesday, September 17, 10.05

• C. Bur, M. Stopp, K. Lorenz, N. Nourkami-Tutdibi, M. Zemlin, S. Goedicke-Fritz:
  Monitoring the Atmosphere of Neonatal Incubators by Metal Oxide Semiconductor Gas Sensors
  Poster presentation

From September 7 to 10 EUROSENSORS XXXVII, will be held in Wroclaw, Poland.

LMT will present two contributions at EUROSENSORS 2025:

• Sebastian Pültz, Christian Bur, Andreas Schütze:
  Bayesian Inference for Reliable Gas Sensing with Metal-Oxide Sensors
  Poster presentation MP2, poster session I, Monday, September 8

• Hamza Ali Imran, Oliver Brieger, Christian Bur, Andreas Schütze:
  Optimizing AC Excitation Frequency for Linear Complex Impedance Response in Metal Oxide Semiconductor Gas Sensing
  Poster presentation TP21, poster session II, Tuesday, September 9
  (Results from the MSCA project SERENADE)

PhD position open in EU project SERENADE on sensors for food quality

Food production and distribution is a huge sector, with significant environmental impact. The EU is prioritising green and circular solutions across the entire food supply chain. Funded by the Marie Skłodowska-Curie Actions programme, the SERENADE project focuses on the end of the food supply chain, developing solutions that target food waste at households, supermarkets and retailers.

Project work spans three pillars, namely food, sensors and materials technologies. It will produce two innovations: a smart, sustainable, sensor-based food container to monitor food freshness; and a handheld food spoilage AI-based analyser to assess freshness of unpackaged products.

The position of DC4 within the project will work on the latter task: a handheld sensor system for assessing food spoilage comprised of a miniaturized GC with a MOS sensor as highly sensitive detector. Applications are requested until May 31, the official recruitment offer can be found here: https://euraxess.ec.europa.eu/jobs/342508

The successful PhD candidate is expected to start as early as possible and will be co-supervised by USAAR-LMT and JLM Innovation GmbH, Tübingen, as industrial partner.

For any questions concerning the position, please contact Prof. Andreas Schütze or Dr. Christian Bur.