For the upcoming academic year I propose the following topics:
Implementation and Validation of a Reinforcement Learning-Based Rate Selection Algorithm for IEEE 802.11 Networks – The goal is to select a rate selection algorithm proposed in the literature for IEEE 802.11 (Wi-Fi) networks which uses reinforcement learning, implement it using the Reinforced-lib Python library and validate its performance in ns-3. A description of this library and an implementation of exemplary rate selection algorithms is given in Zastosowanie algorytmów wielorękich bandytów do wyboru szybkości transmisji w sieciach IEEE 802.
For the upcoming academic year I would like to propose the following topics/areas for obtaining a master’s degree:
Optimizing IEEE 802.11 Networks with Machine Learning – There are at least two tools (ns3-gym, ns3-ai) which allow applying machine learning algorithms to ns-3. You can also look at a previous work I supervised (paper, GitHub repo), in which reinforcement learning was used to dynamically set 802.11’s contention window value. The exact topic in this area is still to be agreed upon.
For the upcoming academic year I would like to propose the following general topics (to be specified later with selected candidates):
Evaluating Machine Learning Extensions for Wi-Fi Performance Analysis in ns-3 – The goal is to demonstrate how the performance of Wi-Fi can be improved using two machine learning frameworks available for ns-3: ns3-gym and ns3-ai. A Python-based Simulator of NR-U FBE Channel Access in Unlicensed Bands – The goal is to implement and validate a simulator written in Python that implements the channel access mechanism of NR-U networks operating in shared channels under the frame-based equipment (FBE) variant of listen-before talk (LBT) defined in ETSI EN 301 893.
For the upcoming academic year I would like to propose the following topics/areas for obtaining a master’s degree:
Optimizing IEEE 802.11 Networks with Machine Learning – There are at least two tools (ns3-gym, ns3-ai) which allow applying machine learning algorithms to ns-3. You can also look at a previous work I supervised (paper, GitHub repo), in which reinforcement learning was used to dynamically set 802.11’s contention window value. The exact topic in this area is still to be agreed upon.
For the upcoming 2021-2022 academic year I would like to propose the following general topics (to be specified later with selected candidates):
A Python-based Simulator of 4G/5G Channel Access in Unlicensed Bands – The goal is to implement and validate a simulator written in Python that implements the channel access mechanism of 4G/5G networks (LAA/NR-U) operating in shared channels. I believe this topic can be realized using SimPy, although I’m open to alternative approaches.
For the upcoming academic year I would like to propose the following topics for obtaining a master’s degree:
Performance Analysis of IEEE 802.11ax OFDMA in ns-3 – The new multi-user channel access method provided by 11ax is being implemented in ns-3. You will get access to the development version of this code and your goal will be to assess the performance of OFDMA in comparison with single-user access. Optimizing IEEE 802.
For the upcoming academic year I would like to propose the following topic: A Python-based Simulator of Channel Access in IEEE 802.11 Networks. I put forward this topic for the consideration of 3rd year students of
Elektronika i Telekomunikacja, Electronics and Telecommunications, Teleinformatyka, who plan to complete their engineer’s degrees by defending their diploma theses in January 2021.
Here’s a brief description of the topic:
The goal of the work is to implement and validate a simulator written in Python that implements the competitive channel access mechanism of IEEE 802.