Metaverse's stochastic optimization for enhanced computation and communication

• Main Author: Ng, Wei Chong
• Other Authors: Dusit Niyato
• Format: Thesis-Doctor of Philosophy
• Language: English
• Published: Nanyang Technological University 2024
• Subjects: Computer and Information Science; Metaverse; Resource allocation; Stochastic integer programming; Edge computing; Computation offloading; Wireless power transfer; Coded distributed computing; Semantic communication
• Online Access: https://hdl.handle.net/10356/175933
• Institution: Nanyang Technological University

The emergence of the Metaverse, a collective virtual shared space, represents a paradigm shift in how we perceive and interact with digital environments. Beyond mere entertainment, the Metaverse promises transformative applications in education, professional collaboration, healthcare, social interaction, and even in reshaping economic models through virtual commerce and real estate. This thesis offers an in-depth exploration of the technological and computational challenges and solutions pivotal to these applications. Central to this exploration is the role of Virtual Service Providers. Their challenges, especially in terms of latency reduction and efficient data communication, are dissected, revealing the intricate balance required to maintain real-time interactions in such expansive virtual spaces. The research introduces the mobile cloud-edge-end collaborative computing paradigm, a cornerstone for the Metaverse's infrastructure. This paradigm promises to revolutionize data processing and delivery, ensuring seamless experiences for users across various applications, from virtual classrooms to immersive therapeutic environments. A significant portion of the study is dedicated to understanding the Wireless Charging Hybrid Network. The nuances of its optimization, especially when considering stochastic offloading, are presented, highlighting its potential to support the vast array of devices that users might employ within the Metaverse. Furthermore, the thesis delves into advanced computational models, with a particular emphasis on BERT. The potential of such models in semantic data processing is explored, emphasizing their role in ensuring that data delivery in the Metaverse is not only efficient but also contextually relevant, enhancing user experiences across diverse applications. Through rigorous analysis, case studies, and a synthesis of current technological trends, this research paints a comprehensive picture of the Metaverse's current state, its potential applications, and its future trajectory. It underscores the importance of continuous innovation, the challenges that lie ahead, and the immense potential the Metaverse holds for reshaping digital interactions in various sectors.