Implement an ML pipeline for plaque detection/classification and quantify robustness under controlled interference (sampling, timing, noise)

Primary Goals:

• Check and understand existing CFD simulation data, understand data augmentation and conversion into impulse response
• Set up ML pipeline with data splitting and different methods: Define target (classification vs prediction), investigate and implement ML algorithms, define evaluation metrics. Run a few test cases for each algorithm and check results.
• Add different challenges/disturbances to the data: Spatial sampling, temporal sampling, additional noise models
• Evaluate the different ML algorithms systematically based on the disturbances: Identify when the models fail depending on the disturbance

• Extend an existing molecular communication testbed setup to support multiple transmitters (and receivers)
• Implement at least three different multiple access schemes (TMDA, MDMA, NOMA)
• Evaluate the bit-error-rate performance of all schemes under different conditions
• Potential extensions: extend beyond 2 TXs/RXs, implement more schemes

Molecular communication (MC) is an alternative to classical electromagnetic wave-based communication, where molecules are used for information exchange. MC is expected to enable in-body networks for future medical applications in the Internet of Bio-Nano Things, a vision for 6G and beyond.

We seek a working student to help us operate and maintain a molecular communication networks testbed at the chair. The testbed is based on ink molecules transmitted through a water-filled tube system with a background flow. Spectral absorption measurements are used to detect the information molecules.

What you will do:

• Set up the tubes, pumps, sensors, and other components
• Program microcontrollers to control microfluidic pumps and spectral sensors
• Debugging and fault detection in both software and hardware
• Implement data collection and visualization methods
• Support with experiments conducted on the testbed

Prerequisites

The Wireless Sensor Networks lab offers the opportunity to develop software solutions for the wireless sensor networking system, targeting innovative applications. For the next semester, a position is available to assist the participants in learning the programming environment and during the project development phase. The lab is planned to be held on-site every Tuesday 15:00 to 17:00.

Prerequisites