This is an opportunity to join a group of leading specialists in digital communications group specifically to work on 5G systems and beyond. You will be in collaboration with the EU Horizon 2020 project “Bring Reinforcement-learning Into Radio Light Network for Massive Connections”(6G BRAINS) partners towards 6G networks
This PhD project is to work on 6G systems. You will be involved in new waveform design for the THz channel and AI enabled resource scheme for GF NOMA with CF massive MIMO over the high dynamic D2D clusters. This project is to develop a resource allocation scheme includes the dynamic precoding, accurate beamforming, intelligent user pairing, channel estimation, beam scheduler and high accurate positioning using multi-agent deep reinforcement learning scheme over Cell-free massive D2D clusters. Furthermore, this project is to design a radio waveform dedicated for THz. THz is faced with more difficult challenges, such as higher phase noise (PN), extreme propagation loss, high atmospheric absorption in certain frequencies, lower power amplifier (PA) efficiency, and strict transmitted power spectral density regulatory requirements, when compared to lower frequency bands.
Applications are invited for an EU Horizon 2020 6G Project “Bring Reinforcement-learning Into Radio Light Network for Massive Connections”(6G BRAINS) PhD studentship in the area of AI driven resource scheduling for high dynamic networks. Ubiquitous smart wireless connectivity is critical for future large-scale industrial tasks, services, assets and devices in future 6G systems. This project aims to bring AI-driven multi-agent Deep Reinforcement learning (DRL) to perform resource allocation over and beyond massive machine-type communications with new spectrum links including THz and optical wireless communications (OWC) to enhance the performance with regard to capacity, reliability and latency for future industrial networks. The research will involve both experimental and theoretical studies working and involve close collaboration with experts in Viavi Solutions(UK), ISEP(FR), Brunel University(UK), Fraunhofer IIS(DE), University of the West of Scotland(UK), Eurecom(FR), Oledcomm SAS(FR), Thales(FR), Bosch(DE), REL(IL), Altran(PT), Telekom Deutschland(DE), Eurescom(DE) attend project meeting in Europe.
The student will be based in the School of Engineering. The research involved will take a multi-disciplinary approach to tackle big and open challenges with an aim to enable AI-driven multi-agent Deep Reinforcement learning (DRL) to perform resource allocation for high dynamic and ultra dense D2D clusters. It will include
1) Design and analyse of new radio waveform for the THz and OWC, where the impact of practical impairments (such as higher phase noise , extreme propagation loss, high atmospheric absorption in certain frequencies, lower power amplifier efficiency, and strict transmitted power spectral density regulatory requirements, when compared to lower frequency bands) on the overall system performance will be considered for the waveform;
2) Design and model of the MESH network enhanced with the new spectrum links for THz and OWC transmission with underlaid high dynamic D2D clusters using Software Defined Radio platform
3) Model and analyse the grant-free based NOMA coordinated with OMA medium control scheme over the high dynamic ultra-dense D2D CF network. This task will provide the processing algorithms with NOMA in DRL agent. Hybrid power and code domain SCMA will be used in UL. And new PHY designs are developed for distributed asynchronous multi-user THz and OWC systems that can cover the different transmitter and receiver. This task also defines the interface to the multi-agent DRL scheme for the DRL model which will be used for the NOMA encoding network.
Entry requirements
The applicant should have (or expect to obtain by the start date) at least a good 2:1 (and preferably a Masters degree) in engineering or a related subject such as physics or materials, strong mathematical ability and an enquiring and rigorous approach to research together with a strong intellect and disciplined work habits. Good team-working, observational and communication skills are essential. The applicant should have strong program skills in Matlab, C/C++. The knowledge of Linux will be essential.
The applicant should have (or expect to obtain by the start date) at least a good 2:1 (and preferably a Masters degree) or overseas equivalent in engineering or a related subject such as physics or materials, strong mathematical ability and an enquiring and rigorous approach to research together with a strong intellect and disciplined work habits. Good team-working, observational and communication skills are essential. The applicant should have strong program skills in Matlab, C/C++. The knowledge of Linux will be essential.
University of Leicester English language requirements apply.
How to apply
Please refer to the application advice and application link at: https://le.ac.uk/study/research-degrees/funded-opportunities/
Key dates
Application deadline: 15th Feb 2021
Start date: May 2021
Continue with Facebook
