Abstract
Wireless-connected Virtual Reality (VR) provides an immersive experience for VR users from anywhere at anytime. However, providing wireless VR users with seamless connectivity and real-time VR video with high quality are challenging due to its requirements in high Quality of Experience (QoE), low VR interaction latency under latency threshold, and privacy. To address these issues, in this thesis, we mainly focus on optimizing the uplink, rendering, downlink, and privacy of wireless VR in 5G and terahertz (THz) networks.If the viewpoint of the VR user can be predicted, the corresponding VR video frames can be rendered and delivered in advance, which can reduce the VR interaction latency.
Thus, in the first chapter, we use offline and online learning algorithms to predict the viewpoint of the VR user using a real VR dataset. For the offline learning algorithm, the trained learning model is directly used to predict the viewpoint of VR users in continuous time slots. While for the online learning algorithm, based on the VR user’s actual viewpoint delivered through uplink transmission, we compare it with the predicted viewpoint and update the parameters and input viewpoints of the online learning algorithm to further improve the prediction accuracy. To guarantee the reliability of the uplink transmission, we integrate the Proactive retransmission scheme of 5G into our proposed online learning algorithm. Simulation results show that our proposed online learning algorithm with the proactive retransmission scheme for wireless VR networks only exhibits about 5% prediction error.
Rendering real-time VR videos with high quality demands a computing unit with high processing ability. Thus, in the second chapter, we propose a mobile edge computing (MEC)-enabled wireless VR network in 5G, where the field of view (FoV) of each VR user can be real-time predicted using Recurrent Neural Network (RNN), and the rendering of VR content is moved from the VR device to the MEC server with rendering model migration capability. Taking into account the geographical and FoV request correlation, we propose centralized and distributed decoupled Deep Reinforcement Learning (DRL) strategies to maximize the long-term QoE of VR users under the VR interaction latency constraint. Simulation results show that our proposed MEC rendering schemes and DRL algorithms substantially improve the long-term QoE of VR users and reduce the VR interaction latency compared to rendering at VR devices.
In an indoor scenario, a high data rate over short transmission distances may be achieved via the abundant bandwidth in the THz band. However, THz waves experience severe signal attenuation, which may be compensated by the reconfigurable intelligent surface (RIS) technology with programmable reflecting elements. Meanwhile, the low VR interaction latency can be achieved with the MEC network architecture due to its high computation capabilities. Motivated by these considerations, in the third chapter, we propose an MEC-enabled and RIS-assisted THz VR network in an indoor scenario, by taking into account the uplink viewpoint prediction and position transmission, the MEC rendering, and the downlink transmission. We propose two methods, which are referred to as centralized online gated recurrent unit (GRU) and distributed federated averaging (FedAvg), to predict the viewpoints of the VR users. In the uplink, an algorithm that integrates online long-short term memory (LSTM) and convolutional neural networks (CNN) is deployed to predict the locations and the line-of-sight (LoS) or non-line-of-sight (NLoS) statuses of VR users over time. In the downlink, we develop a constrained DRL (C-DRL) algorithm to select the optimal phase shifts of the RIS under latency constraints. Simulation results show that our proposed learning architecture achieves near-optimal QoE as that of the genie-aided benchmark algorithm, and about two times improvement in QoE compared to the random phase shift selection scheme.
| Date of Award | 1 Jul 2022 |
|---|---|
| Original language | English |
| Awarding Institution |
|
| Supervisor | Yansha Deng (Supervisor) |