How Does Distributed Control Work in Wireless Networks?


A postdoc position at Yale or a junior professorship at the Berlin Einstein Center Digital Future (ECDF)? Iranian-born Setareh Maghsudi is in no doubt: “I was always interested in doing research at the crossroads of different disciplines. That’s why I chose Berlin. I also had the chance here to be part of the development of something completely new. In my first few months at the ECDF I have already learned a lot and gained completely new perspectives.” Since August 2017 the 32-year-old Tehrani has been an ECDF junior professor at TU Berlin in the Department of Control of Convergent Access Networks. This has left her little time to pursue her personal interests such as sports, music, or languages.

After completing her bachelor’s degree in electrical engineering in Iran, she completed her master’s in digital communications at Kiel University in Germany. She then transferred to TU Berlin to do her doctorate, after which she researched as a postdoc at the University of Manitoba in Canada for one year and, as part of a two-year German Research Foundation (DFG) postdoctoral fellowship, at Yale University in the United States. She cut her stay at Yale short to take the position at the ECDF.

At the ECDF, Maghsudi researches aspects of distributed control in wireless networks. She tries to bring order to chaos in networks, drawing upon, among other things, game theory and machine learning. “Today we all use multiple mobile terminal devices that we expect to be able to use all the time and without limitation. To make these devices available around the clock, however, we need greater resources such as bandwidth and performance. If, for example, direct exchange between various mobile devices worked, there would at least be potential for greater performance and more bandwidth for everyone, meaning generally more resources. This is, initially at least, a positive development, but at the moment the concept still poses a few technical and data protection challenges. For example, it could increase the level of interference. In the context of distributed control of such networks people decide spontaneously how and when to use such resources, using unpredictable decision-making strategies. My research focuses on developing new decision-making strategies that are sensible, technically feasible, and attractive for the user. I use game theory and machine learning to do this,” explains Maghsudi.