Abstract

In recent years, Non-Orthogonal Multiple Access (NOMA) has emerged as a promising technique for enhancing the capacity and throughput of wireless communication systems. This thesis investigates the potential of NOMA in improving the performance of multiuser multicast systems, focusing on multibeam satellite communication systems in the forward link, throughput enhancement, and power optimization. We propose a novel framework that combines a NOMA scheme with multibeam architecture and frequency reuse in multicast transmission. The proposed framework enhances system throughput by optimizing power allocation.
First, we present a comprehensive review of the principles and techniques related to NOMA and multibeam multicast systems, highlighting their unique challenges and potential benefits. Next, we introduce our proposed framework in 4-color frequency reuse satellite systems. In 4-color frequency reuse, each user receives signals from other co-channel beams. However, the level of isolation is such that the interbeam interference can be treated as background noise without significant performance degradation. This means that there is no collaboration between beams, and each beam can be isolated from the rest. Therefore, NOMA is considered in single-beam multicast satellite communication systems. The optimum power allocation to maximize the minimum fairness rate and sum-rate is derived for a given user clustering in a single beam. Moreover, an optimum user clustering is derived, which improves the system throughput.
Next, we investigate our proposed framework in full frequency reuse satellite systems under perfect channel state information at the transmitter (CSIT). The proposed framework integrates the NOMA scheme in multicast multibeam architecture. Linear precoding techniques, such as zero-forcing (ZF) and minimum mean square error (MMSE), are used to cancel interbeam interference while NOMA is applied on a beam basis. NOMA and linear precoding are adopted for the proposed framework in multicast transmission. A low-complexity user scheduling is proposed to deal with the trade-offs between optimum user scheduling for linear precoding and the NOMA scheme. Moreover, a low-complexity linear precoding in multicast transmission is proposed based on unicast linear precoding methods and a mapper which deals with the lack of spatial degrees of freedom. To improve the performance of linear precoding, we present three mappers, where the proposed singular-value-decomposition (SVD) mapper demonstrates the best performance.
To improve system throughput, power allocation should be optimized. In this thesis, we consider two objective functions: max-min fairness rate (MMF) and sum-rate. This thesis introduces a technique for addressing the non-convex MMF optimization issue in the proposed framework by employing auxiliary variables to convert it into a semi-definite programming problem, which can then be resolved using linear programming solvers. This thesis also suggests an approach to tackle the non-convex sum-rate maximization goal function in MB-MC-NOMA systems by constructing Lagrangian multipliers concerning the constraints. By employing quadratic transformations on the sum-of-ratios, the problem is restructured within an iterative sum-rate power optimization algorithm.
This thesis considers a realistic scenario with imperfect CSIT. To combat the effect of imperfect CSIT in multibeam multicast satellite communication systems, a rate-splitting approach is proposed. An averaging rate (AR) framework for MMF rate and sum-rate optimization considering ICST is proposed. To render the formulated MMF and sum-rate problems convex, we utilize the Weighted Minimum Mean Square Error (WMMSE) method. We first derive a rate-WMMSE relationship and then, using this relationship along with a low-complexity solution based on Alternating Optimization (AO), we transform the problems into equivalent convex ones.
To validate the effectiveness of our proposed frameworks, we conduct extensive simulations and comparisons with state-of-the-art schemes. The results demonstrate significant improvements in throughput and power efficiency, confirming the potential of NOMA-based multiuser multicast systems for future wireless communication networks.
Finally, we discuss potential future research directions, including the integration of the proposed frameworks in the cellular networks, calculating the transmitter and receiver complexity of the proposed techniques, considering higher layers of RS. This thesis contributes to the ongoing development of next-generation wireless communication systems, paving the way for more efficient and reliable data transmission in multiuser multicast environments.