Joint Beam-Forming, User Clustering and Power Allocation for MIMO-NOMA Systems

Area of Science:

• Wireless communication systems
• Signal processing
• Optimization theory

Background:

• Non-orthogonal multiple access (NOMA) enhances spectral efficiency in wireless networks.
• Multiple-input multiple-output (MIMO) systems offer spatial multiplexing gains.
• Integrating MIMO with NOMA (MIMO-NOMA) presents complex resource allocation challenges.

Purpose of the Study:

• To develop an efficient framework for optimal resource allocation in MIMO-NOMA systems.
• To address the interconnected problems of beam-forming, user clustering, and power allocation.
• To improve system performance, particularly spectral efficiency and data rates for edge users.

Main Methods:

• A three-step framework is proposed: beam-forming, user clustering, and power allocation.
• Beam-forming utilizes fractional transmitting power control (FTPC) and the limited-memory Broyden-Fletcher-Goldfarb-Shanno (L-BFGS) method.
• User clustering considers channel differences and correlations, generating multiple schemes.
• Power allocation is formulated as a difference of convex (DC) programming problem solved via successive convex approximation (SCA).

Main Results:

• The proposed beam-forming algorithm optimizes intra-cluster power allocation.
• User clustering schemes are evaluated based on performance metrics.
• The SCA method provides a robust solution for power allocation.
• Simulations demonstrate significant improvements in spectral efficiency and edge user data rates.

Conclusions:

• The integrated three-step framework effectively solves the multi-dimensional resource allocation problem in MIMO-NOMA systems.
• The proposed methods, including L-BFGS and SCA, achieve optimal or near-optimal solutions.
• The system demonstrates enhanced performance, particularly for users at the cell edge.