Guest lecture: Statistical Modeling and Inference for Massive Physical-Binary Sensor

Postdoc Manuel S. Stein, Technische Universität München (TUM), Germany.

2018.11.01 | Jens Kargaard Madsen

Date Fri 09 Nov
Time 14:15 15:00
Location Room 424, building 5125, Finlandsgade 22, 8200 Aarhus N

Abstract
Systems Fall of the Analog-to-Digital Information Bottleneck? While the technological capabilities regarding digital data transmission, storage, and computation have exponentially increased during the last decades, the advances associated with analog sensor circuits have been rather moderate. Therefore, today hardware cost and power consumption of sensor front-ends form fundamental obstacles for designing advanced measurement systems featuring either ultra-low complexity or ultra-high performance. However, in the advent of the Internet of Things (IoT), where cheap and small devices are supposed to perform challenging sensing tasks, and with the increasing demand for performance and reliability in medical, critical infrastructure, and big science applications, it becomes inevitable to push system architectures further towards these extremes. In this context, I will give an introduction to massive binary sensing, an emerging concept trading digitization complexity for other system design aspects which provide significant additional sensing performance. Nevertheless, the thorough understanding of the opportunities and challenges surrounding binary sensor systems requires an interdisciplinary approach uniting expertise from physics, mathematics, computer science, and engineering. To foster the joint discourse, I will outline the perspective of a theoretic engineer on the physically consistent modeling of binary sensor data streams and the performance analysis concerning statistical inference tasks. Potential practical implications will be illustrated in the talk via applications which require precise time-of-arrival (ToA) measurements of wireless signals. In particular, I will point to the ToA sensing gain when starting to exploit the temporal and spatial design dimensions becoming massively available at low cost in binary sensor systems.

Manuel S. Stein received the Dipl.-Ing. and the Dr.-Ing. degree in electrical engineering and information technology from the Technische Universität München (TUM), Germany, in 2010 and 2016, respectively. From 2011 to 2016, he has been a research associate at the former Institute of Circuit Theory and Signal Processing, TUM, where he worked on his doctoral thesis concerning signal parameter estimation with low-complexity analog-to-digital conversion. Since then he has been holding postdoctoral positions in the Signal and System Theory Group, Universität Paderborn, Germany, at the Mathematics Department, Vrije Universiteit Brussel, Belgium, and at the Chair for Stochastics, Universität Bayreuth, Germany. He is recipient of the P.R.I.M.E. Marie Skłodowska-Curie Fellowship granted by the German Academic Exchange Service (DAAD) and was awarded the [PEGASUS]² Marie Skłodowska-Curie Fellowship by the Research Foundation Flanders (FWO). His research focuses on the foundations of statistical signal processing for hardware-aware systems and applications in satellite-based radio systems, array processing, wireless communication, and radar.

Lecture / talk