Theses

We have some open positions to write theses and also publish ongoing and finished ones.

How to apply for a thesis

If you think about writing a thesis with WISE, please fill out our application form and we'll try to find a tailor-made topic for you.

No results match your search criteria.

  2023 Open

FPGA development and experimental analysis of beamforming and beam-tracking in 6G networks

Supervisor: Arash Asadi Waqar Ahmed

Millimeter-wave frequencies (30-300 GHz) will be dominating 6G communications, providing users with tens of Gbps data rates. However, communication at such high frequencies requires using highly directional beams to compensate for the propagation loss. In our group, we have access to unique software-defined radios capable of communication at 70 GHz with 4GHz of bandwidth. If you are interested in performing experimental studies in this area and contributing to the research in the next generation of mobile networks, this could be your topic. Research objective: Test and development of agile beamforming/tracking for 6G systems Expected gain of knowledge: Wireless communication, FPGA programming

  2023 Open

Unsupervised learning from video segmentation to person/object tracking in wireless networks

Supervisor: Arash Asadi

There is a large body of work on using commercial wireless devices to detect, identify and localize people as well as their motion, gestures, and even vital signs. The underlying techniques span from machine learning techniques to signal processing and Radar. To some extent, the impact of a person’s body/motion on the wireless signals can resemble an image/video. While there has been extensive use of advanced Machine learning techniques for people/object tracking in videos, there is very little work on using these techniques in the wireless domain. For example, applying the works presented here (https://www.youtube.com/watch?v=tSBWZ6nYld0) to wireless sensing. If you find this interesting, send me an email to discuss further details.

  2023 Open

Preserving Privacy against WiFi Sensing

Supervisor: Arash Asadi

Your WiFi router is constantly monitoring the surrounding. You can analyze the channel state information to detect the location and even trajectory of people in their homes. There are many other applications, including detecting heartbeat, breathing rate, reading lips, etc. If you are interested in implementing one of these systems using real hardware and finding solutions to fight against it, send me an email. Note that these are rather challenging topics as they require good knowledge of communication as well as signal processing. Research objective WiFi Sensing countermeasures Expected gain of knowledge Wireless communication, Signal processing

  2023 Open

mmWave full duplex joint sensing and communication design

Supervisor: Lu Wang Arash Asadi

Joint Sensing and Communication technology is one of the key 6G technologies. It makes your phone/vehicle/device, etc., smarter with the function of sensing and communication simultaneously [1]. Configuring this technology in mmWave band, better performance such as higher date rate can be realized. If you are interested in the joint sensing and communication system design, feel free to contact us. Research objective: 1. Sensing parameters estimation 2. full duplex joint sensing and communication system design Expected gain of knowledge: Wireless communication [1] Y. Cui, F. Liu, X. Jing and J. Mu, “Integrating Sensing and Communications for Ubiquitous IoT: Applications, Trends, and Challenges,” in IEEE Network, vol. 35, no. 5, pp. 158-167, September/October 2021.

  2023 Open

Explainable machine learning to explain blackbox human activity sensing

Supervisor: Arash Asadi

Your WiFi router is constantly monitoring the surrounding. You can analyze the channel state information to detect the location and even trajectory of people in their homes. The majority of these works leverage black box machine learning, which questions their reliability. In this thesis, you will be working on cutting-edge explanation methods [1] for Deep learning models. Research objective: Explanation methods for WiFi Sensing Expected gain of knowledge: Wireless communication, Explainable machine learning [1] https://cloud.google.com/explainable-ai

  2022 Open

Beam management in mmWave full duplex joint sensing and communication system

Supervisor: Lu Wang Arash Asadi

Joint Sensing and Communication technology is one of the key 6G technologies. It makes your phone/vehicle/device smarter with the function of sensing and communication simultaneously [1]. By beaming the transmitted data in a direct way, both sensing and communication performance can be improved. If you are interested in the beam-related design in the joint sensing and communication system, feel free to contact us. Research objective: 1. Sensing parameters estimation 2. Beam management(searching/tracking) and beamforming design 3. full duplex joint sensing and communication system design Expected gain of knowledge: Wireless communication [1] Y. Cui, F. Liu, X. Jing and J. Mu, “Integrating Sensing and Communications for Ubiquitous IoT: Applications, Trends, and Challenges,” in IEEE Network, vol. 35, no. 5, pp. 158-167, September/October 2021.

  2022 Open

Going against the tide: Using interpretable machine learning instead of black box DNN for wireless sensing

Supervisor: Arash Asadi

Your WiFi router is constantly monitoring the surrounding. You can analyze the channel state information to detect the location and even trajectory of people in their homes. The majority of these works leverage black box machine learning, which questions their reliability. While many believe that the black box models provide higher performance and are less complex, new studies suggest otherwise [1]. If you are interested in going against the tide and proving interpretable learning can perform similar to black box models in wireless sensing, send me an email. Research objective: Explanation methods for WiFi Sensing Expected gain of knowledge: Wireless communication, Interpretable machine learning [1] C. Rudin, “Stop explaining black box machine learning models for high stakes decisions and use interpretable models instead,” Nat Mach Intell, vol. 1, no. 5, pp. 206–215, 2019, doi: 10.1038/s42256-019-0048-x.

  2021 Completed

Privacy-Preserving radiometric fingerprinting (Datenlotsen Awardee)

Supervisor: Arash Asadi

  2021 Completed

Device-Free Indoor Localization: A User-Privacy Perspective

Supervisor: Arash Asadi

  2021 Completed

Protecting Heartbeat and Respiration Information in WiFi Sensing Applications

Supervisor: Arash Asadi

  2018 Completed

Experimental Evaluation on Inband Device-to-Device Communication in LTE

Supervisor: Arash Asadi

  2017 Completed

Evaluation of Latency Reduction Techniques for 5th Generation Mobile Network

Supervisor: Arash Asadi