Scenario-Specific Beamforming Design and Traffic-Aware Resource Allocation for 5G-NR Systems

In the realm of fifth-generation new radio (5G-NR) communication technology, how to use limited resources and low-complexity algorithms to provide high transmission rates for massive multiple-input multiple-output (mMIMO) systems with differentiated service demands remains a challenge. This paper explores a novel scenario-specific beamforming design tailored to user distributions and propagation environments. The proposed method dynamically adjusts beam orientations and widths, enabling precise coverage of users’ active regions. To mitigate inter-beam interference in spatial multiplexing, we derive a strict orthogonality condition and propose a corresponding beam grouping strategy to enforce it. Furthermore, to address the nonconvex optimization problem of time-frequency-space resource allocation (RA), we build upon the idea of the greedy algorithm to balance transmission rate and computational complexity. Building on this foundation, we propose a traffic-aware RA strategy and introduce a novel beam selection criterion by incorporating both channel characteristics and user-specific traffic demands. These improvements jointly enhance the system’s utilization of time-frequency resources and spatial reuse capability. Simulation results validate the proposed algorithms, demonstrating notable improvements in computational efficiency, system adaptability, spectral efficiency, and average achievable rate.