TY - CONF A1 - Neumann, Patrick P. A1 - Bartholmai, Matthias A1 - Lazik, D. T1 - Near real-time reconstruction of 2D soil gas distribution from a regular network of linear gas sensors T2 - Proceedings of the IEEE Sensors 2015 N2 - A monitoring method is introduced that creates, in near real-time, two-dimensional (2D) maps of the soil gas distribution. The method combines linear gas sensing technology for in-situ monitoring of gases in soil with the mapping capabilities of Computed Tomography (CT) to reconstruct spatial and temporal resolved gas distribution maps. A weighted iterative algebraic reconstruction method based on Maximum Likelihood with Expectation Maximization (MLEM) in combination with a source-by-source reconstruction approach is introduced that works with a sparse setup of orthogonally-aligned linear gas sensors. The reconstruction method successfully reduces artifact production, especially when multiple gas sources are present, allowing the discrimination between true and non-existing so-called ghost source locations. A first experimental test indicates the high potential of the proposed method for, e.g., rapid gas leak localization. PB - IEEE T2 - IEEE Sensors 2015 CY - Busan, South Korea DA - 01.11.2015 KW - Distributed linear sensor KW - Membrane-based gas sensing KW - Subsurface monitoring KW - Gas storage areas KW - Computed tomography PY - 2015 UR - https://opus4.kobv.de/opus4-bam/frontdoor/index/index/docId/34849 AN - OPUS4-34849 SN - 978-1-4799-8202-8 SN - 1930-0395 SP - 1550 EP - 1553 AD - Bundesanstalt fuer Materialforschung und -pruefung (BAM), Berlin, Germany