@inproceedings{GiskeBankwitzEngeletal.2024, author = {Giske, Karl and Bankwitz, Madeleine and Engel, Christian and Bendyk, Ricky and Hirsch, Soeren}, title = {Advanced Sensor Systems for Sustainable Building Modernization: A Technological Approach to Enhance CO2 Savings}, series = {EUROSENSORS XXXVI, Debrecen, Hungary, 2024}, booktitle = {EUROSENSORS XXXVI, Debrecen, Hungary, 2024}, isbn = {978-3-910600-03-4}, doi = {10.5162/EUROSENSORSXXXVI/OT10.184}, pages = {216 -- 217}, year = {2024}, abstract = {This work presents an advanced sensor system design developed to improve diagnostic methods for sustainable building renovation, with focus on the optimization of energy efficiency and the achievement of significant CO2 reduction in the construction sector. Introduction Considering the challenges posed by climate change, it is imperative to adopt innovative strategies in all sectors, with particular focus on construction and real estate, which are well-known contributors to energy consumption and CO2 emissions. Sustainable renovation represents a pivotal approach for enhancing energy efficiency , reducing energy demand and improving environmental footprints through precise resource management and energy-efficient construction practices [1].}, language = {en} } @inproceedings{DaupayevReberBendyketal.2025, author = {Daupayev, Nursultan and Reber, Paul and Bendyk, Ricky and Engel, Christian and Hirsch, S{\"o}ren}, title = {Data Reduction for Energy-Constrained Sensors via Event-Aware Sampling}, series = {Konferenz: 2025 IEEE SENSORS, Vancouver, BC, Canada, 2025}, booktitle = {Konferenz: 2025 IEEE SENSORS, Vancouver, BC, Canada, 2025}, publisher = {IEEE}, isbn = {979-8-3315-4467-6}, issn = {2168-9229}, doi = {10.1109/SENSORS59705.2025.11330656}, pages = {1 -- 4}, year = {2025}, abstract = {Environmental monitoring plays a crucial role in analyzing environmental parameters and detecting anomalies. However, sensor systems work continuously, which results in constant energy consumption and data redundancy, especially for sensors with limited computing power and memory. In addition, installing and maintaining sensors in remote places creates additional challenges. An adaptive environmental sensing approach is developed to reduce data redundancy and energy consumption. A custom-designed sensor based on PIC16LF19156 microcontroller measures CO2, humidity and temperature simultaneously. The sensor is connected to a Raspberry Pi, where a signal processing algorithm is executed, aimed at reducing data redundancy, thereby increasing the energy efficiency of the system. The algorithm includes the discrete wavelet transform (DWT) to extract spectral features from the signals. A machine learning model trained on previous data estimates the daily variability of the signal and saves data labeled as SAVE (deviations) or SKIP (consistency). As a result, only the relevant intervals showing significant fluctuations are retained. The effectiveness of the proposed approach was evaluated by reconstructing the compressed signals and comparing them with the original data based on RMSE and MAE metrics, which confirms the insignificant loss of information.}, language = {en} }