Research

Projects

System context information and weakly labeled data

Real-world datasets collected with sensor networks often contain incomplete and uncertain labels as well as artefacts arising from the system environment. To address these challenges, we propose to make use of system context information formalized in an information graph and embed it in the learning process via contrastive learning. Based on real-world data we show that this approach leads to an increased accuracy in case of weakly labeled data and leads to an increased robustness and transferability of the classifier to new sensor locations. Read more...

Natural Hazard Monitoring and On-Device Classification

In natural hazard warning systems fast decision making is vital to avoid catastrophes. Decision making at the edge of a wireless sensor network promises fast response times but is limited by the availability of energy, data transfer speed, processing and memory constraints. In this work we present a realization of a wireless sensor network for hazard monitoring based on an array of event-triggered single-channel micro-seismic sensors with advanced signal processing and characterization capabilities based on a novel co-detection technique. Read more...

Systematic Identification of Noise on Matterhorn (CH)

In this work, we present a systematic strategy to identify a multitude of external influence sources, characterize and quantify their impact and develop methods for automated identification in microseismic signals. We apply the strategy developed to a real-world, multi-sensor, multi-year microseismic monitoring experiment performed at the Matterhorn Hörnligrat (Switzerland). Read more...

Nyquist Sinc-Pulse Shaping

Nyquist sinc-pulse shaping provides spectral efficiencies close to the theoretical limit. We worked on real-time Nyquist pulse shaping implementations using field programmable gate arrays (FPGA) along with high-speed digital-to-analog and analog-to-digital converters (DAC, ADC) within optical transceivers. Read more...

Open Source

The following is a selection of open-source and open-data projects that I initiated or contributed to:

Software


foreal (Github, pypi) - A multi-modal, multi-application data processing and analysis framework designed for data exploration


logadia (Github) - A privacy-first life annotation tool


flocklab-tools (Gitlab, pypi) - Python support for using FlockLab (flocklab CLI, creating flocklab xml, visualization).


Code

Datasets

Publications

Selected Publications

Meyer, M., Weber, S., Beutel, J., & Thiele, L. (2019). Systematic identification of external influences in multi-year microseismic recordings using convolutional neural networks. Earth Surface Dynamics, 7(1), 171–190. https://doi.org/https://doi.org/10.5194/esurf-7-171-2019
Meyer, M., Farei-Campagna, T., Pasztor, A., Da Forno, R., Gsell, T., Faillettaz, J., Vieli, A., Weber, S., Beutel, J., & Thiele, L. (2018). Event-triggered Natural Hazard Monitoring with Convolutional Neural Networks on the Edge. ArXiv:1810.09409 [Cs, Stat]. http://arxiv.org/abs/1810.09409
Hillerkuss, D., Schmogrow, R., Meyer, M., Wolf, S., Jordan, M., Kleinow, P., Lindenmann, N., Schindler, P. C., Melikyan, A., & Yang, X. (2012). Single-laser 32.5 Tbit/s Nyquist WDM transmission. Journal of Optical Communications and Networking, 4(10), 715–723.
Schmogrow, R., Meyer, M., Wolf, S., Nebendahl, B., Hillerkuss, D., Baeuerle, B., Dreschmann, M., Meyer, J., Huebner, M., & Becker, J. (2012). 150 Gbit/s real-time Nyquist pulse transmission over 150 km SSMF enhanced by DSP with dynamic precision. Optical Fiber Communication Conference and Exposition (OFC/NFOEC), 2012 and the National Fiber Optic Engineers Conference, 1–3.

Full List

Meyer, M., Beutel, J., & Thiele, L. (2017, December 11). Unsupervised Feature Learning for Audio Analysis. Workshop Track - ICLR 2017. 5th International Conference on Learning Representations (ICLR), Toulon, France. http://arxiv.org/abs/1712.03835
Meyer, M., Farei-Campagna, T., Pasztor, A., Da Forno, R., Gsell, T., Weber, S., Failletaz, J., Vieli, A., Beutel, J., & Thiele, L. (2019). Event-triggered Geophone Platform for Co-Detection of Seismic Events. Geophysical Research Abstracts, 21, EGU2019-16940. https://doi.org/10.3929/ethz-b-000340844
Meyer, M., Weber, S., & Beutel, J. (2017). Data Exploration using Unsupervised Feature Extraction for Mixed Micro-Seismic Signals. Geophysical Research Abstracts, 19, EGU2017-16980. https://doi.org/10.3929/ethz-b-000161404
Meyer, M., Farei-Campagna, T., Pasztor, A., Da Forno, R., Beutel, T., & Thiele, L. (2019). Demo Abstract: How Many Climb the Matterhorn? 2019 18th ACM/IEEE International Conference on Information Processing in Sensor Networks (IPSN), 250–251.
Weber, S., Beutel, J., Forno, R. D., Geiger, A., Gruber, S., Gsell, T., Hasler, A., Keller, M., Lim, R., Limpach, P., Meyer, M., Talzi, I., Thiele, L., Tschudin, C., Vieli, A., Vonder Mühll, D., & Yücel, M. (2019). A decade of detailed observations (2008–2018) in steep bedrock permafrost at the Matterhorn Hörnligrat (Zermatt, CH). Earth System Science Data, 11(3), 1203–1237. https://doi.org/https://doi.org/10.5194/essd-11-1203-2019
Meyer, M., Weber, S., Beutel, J., & Thiele, L. (2019). Systematic identification of external influences in multi-year microseismic recordings using convolutional neural networks. Earth Surface Dynamics, 7(1), 171–190. https://doi.org/https://doi.org/10.5194/esurf-7-171-2019
Meyer, M., Farei-Campagna, T., Pasztor, A., Da Forno, R., Gsell, T., Faillettaz, J., Vieli, A., Weber, S., Beutel, J., & Thiele, L. (2018). Event-triggered Natural Hazard Monitoring with Convolutional Neural Networks on the Edge. ArXiv:1810.09409 [Cs, Stat]. http://arxiv.org/abs/1810.09409
Weber, S., Faillettaz, J., Meyer, M., Beutel, J., & Vieli, A. (2018). Acoustic and Microseismic Characterization in Steep Bedrock Permafrost on Matterhorn (CH). Journal of Geophysical Research: Earth Surface, 123(6), 1363–1385. https://doi.org/10.1029/2018JF004615
Miedl, P., He, X., Meyer, M., Bartolini, D. B., & Thiele, L. (2018). Frequency Scaling As a Security Threat on Multicore Systems. IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems, 37(11), 2497–2508. https://doi.org/10.1109/TCAD.2018.2857038
Meyer, M., Cavigelli, L., & Thiele, L. (2017). Efficient Convolutional Neural Network For Audio Event Detection. ArXiv:1709.09888 [Cs]. http://arxiv.org/abs/1709.09888
Hillerkuss, D., Schmogrow, R., Meyer, M., Wolf, S., Jordan, M., Kleinow, P., Lindenmann, N., Schindler, P. C., Melikyan, A., & Yang, X. (2012). Single-laser 32.5 Tbit/s Nyquist WDM transmission. Journal of Optical Communications and Networking, 4(10), 715–723.
Schmogrow, R., Meyer, M., Wolf, S., Nebendahl, B., Hillerkuss, D., Baeuerle, B., Dreschmann, M., Meyer, J., Huebner, M., & Becker, J. (2012). 150 Gbit/s real-time Nyquist pulse transmission over 150 km SSMF enhanced by DSP with dynamic precision. Optical Fiber Communication Conference and Exposition (OFC/NFOEC), 2012 and the National Fiber Optic Engineers Conference, 1–3.
Schmogrow, R., Winter, M., Meyer, M., Hillerkuss, D., Wolf, S., Baeuerle, B., Ludwig, A., Nebendahl, B., Ben-Ezra, S., & Meyer, J. (2012). Real-time Nyquist pulse generation beyond 100 Gbit/s and its relation to OFDM. Optics Express, 20(1), 317–337.
Schmogrow, R. M., Meyer, M., Schindler, P. C., Josten, A., Ben-Ezra, S., Koos, C., Freude, W., & Leuthold, J. (2013). 252 Gbit/s real-time Nyquist pulse generation by reducing the oversampling factor to 1.33. Optical Fiber Communication Conference, OTu2I-1.
Schmogrow, R., Bouziane, R., Meyer, M., Milder, P. A., Schindler, P. C., Bayvel, P., Killey, R. I., Freude, W., & Leuthold, J. (2012). Real-time digital Nyquist-WDM and OFDM signal generation: Spectral efficiency versus DSP complexity. Optical Communications (ECOC), 2012 38th European Conference and Exhibition On, 1–3.
Schmogrow, R., Winter, M., Meyer, M., Hillerkuss, D., Nebendahl, B., Meyer, J., Dreschmann, M., Huebner, M., Becker, J., & Koos, C. (2011). Real-time Nyquist pulse modulation transmitter generating rectangular shaped spectra of 112 Gbit/s 16QAM signals. Signal Processing in Photonic Communications, SPMA5.

Presentations

  • AMLD2020: Monitoring Climate Change at the Edge of the Cloud
  • ZHdK Environmental Listening Session: Unknown Soundscapes: Rivers, Soils, Rocks
  • IPSN2019: Event-triggered Natural Hazard Monitoring with Convolutional Neural Networks on the Edge
  • IPSN2019: How Many Climb the Matterhorn? (Demo)
  • EGU2017: Data Exploration using Unsupervised Feature Extraction for Mixed Micro-Seismic Signals