TEMPORAL CONNECTIVITY AS A MEASURE OF ROBUSTNESS IN NONORTHOGONAL MULTIPLE ACCESS WIRELESS NETWORKS

Abstract

Nonorthogonal multiple access (NOMA) is recognized as an important technology to meet the performance requirements of fifth generation (5G) and beyond 5G (B5G) wireless networks. Through the technique of overloading, NOMA has the potential to support higher connection densities, increased spectral efficiency, and lower latency than orthogonal multiple access. The role of NOMA in 5G/B5G wireless networks necessitates a clear understanding of how overloading variability affects network robustness. This dissertation considers the relationship between variable overloading and network robustness through the lens of temporal network theory, where robustness is measured through the evolution of temporal connectivity between network devices (ND). We develop a NOMA temporal graph model and stochastic temporal component framework to characterize time-varying network connectivity as a function of NOMA overloading. The analysis is extended to derive stochastic expressions and probability mass functions for unidirectional connectivity, bidirectional connectivity, the inter-event time between unidirectional connectivity, and the minimum time required for bidirectional connectivity between all NDs. We test the accuracy of our analytical results through numerical simulations. Our results provide an overloading-based characterization of time-varying network robustness that is generalizable to any underlying NOMA implementation.