The participants conduct a simulation project over the whole semester in groups of 2 to 4. The participants may submit project proposals or work on one of the following projects:1) Starlink Internet Access Link:
SpaceX Starlink, a Low Earth orbit (LEO) satellite megaconstellation, is interesting for users who do not have access to terrestrial Broadband Internet access. Compared to geostationary satellites, the latency of LEO satellites as significantly lower (round trip times: GEO ~600ms, LEO ~50ms), which is beneficial for many Internet applications and protocols.
Measurements with a Starlink terminal have shown that both data rate and latency is very unstable given the currently deployed system. Both data rate and latency show significant fluctuations. For testing applications and protocols in a network simulation environment, it is important to have a model of such link characteristics.
In this project, a model of an Starlink Internet access link shall be created. This should include data rate and latency, as shown above. Optionally, packet loss and latency under load could be measured and modeled (c.f., Bufferbloat).
2) HTTP web traffic model:
Modern web pages are very diverse: Some have only a few objects and are optimized for fast loading, while others are very complex. See following two figures for illustration. It is not possible to define a typical website. Therefore, simplified models are desirable.
The basis for this project topic are the publications [1] and [2]. They measure page size, number of objects, and object sizes for a large number of web pages and provide corresponding models.
As websites have significantly changed since 2012, the results from [1] are outdated. A more recent model of website parameters is given in [2] but lacks some aspects which should be considered when analyzing modern websites: HTTP version, number of contacted servers, mobile vs. desktop websites, impact of cookie banners, etc. Also, the data set used for evaluation should be made available publicly.
3) Hydrogen energy storage for communal E-Vehicle charging:
The transition to a CO2 neutral energy system is one of the most challenging endeavours of this decade. Even though there are several potential technical solutions, there are still many open questions. Especially regarding energy storage and management in a decentralized system.
A promissing solution for storing energy in the usage of hydrogen (H2). Most commonly used are electrolysers to convert water to H2, which can be burned to retrieve the stored energy using turbines.
The goal of this project is to evaluate the useage of H2-Energy storage for a communal charging service for 20 to 50 electric vehicles. The town wants to use wind power plants with a height of 100 to 140 meters. The charging service shall mainly be used as a park and ride solution. Suitable models for energy production, storage, retrieval and user behavior and useage patterns have to be developed.
Optional questions revolve around the interface to the general energy grid and the energy market. For example: How much energy has to be bought in addition? How much can be sold? How could a price dependent control strategy look like and what is the impact on costs and quality of service?
4) Thermal energy storage for residential buildings:
The transition to a CO2 neutral energy system is one of the most challenging endeavours of this decade. Even though there are several potential technical solutions, there are still many open questions. Especially regarding energy storage and management in a decentralized system.
In residental buildings around 80% of the total energy is used for heating. A common technology for heating is the use of thermal energy storage, i.e. heating water in an thermally insulated tank. This heat can also be used to retrieve the energy and generate electricity.
The goal of this project is to evaluate the useage of thermal energy storage for a residential building. The theoretical building has an area of 80 to 160m² and a roof surface of 50m² that is usable for photovoltaic solar panels or thermal solar panels. In addition the stored and generated energy shall be sufficient to charge one electric car over night.
Optional questions revolve around the interface to the general energy grid and the energy market. For example: How much energy has to be bought in addition? How much can be sold? How could a price dependent control strategy look like and what is the impact on costs and quality of service?