Open Radio Access Networks “O-RAN” SystemsThroughput improvement)

Abstract

Open Radio Access Networks (O-RAN) are expected to revolutionize
communication systems. O-RAN enables virtualized Radio Access Networks
where distributed components are attached through open interfaces and
intelligent controllers optimize the performance. A new conceptual design has been created for network configuration, deployment, and functions. This design uses interchangeable components from various vendors and can be optimized for performance through a centralized inference layer and data-driven control. However, due to the large number of users and limited capacity of the fronthaul in dense wireless systems, it becomes challenging to achieve optimal resource allocation in such massive systems. In this work, we balance the spectral efficiency and the required power, which will reduce the signalling overheads and processing in high-density radio access networks. More specifically, a linear channel estimation (based on the MMSE technique) is employed to design the Conjugate Beamforming vectors. The suggested iterative approach,
namely LSF-IV, leverages the large-scale fading detection (LSF) and the
intermediate value method (IV) to attain the balancing between users' uplink spectral performance. Concerning the algorithm validity and for a specific parameter setting scenario, the proposed approach can significantly enhance the spectral performance for the users in the worst channel conditions compared with the typical fractional PA. Adjusting a specific parameter can improve spectral performance by 45% with a 95% likelihood. This study found that using a more extended training sequence for channel estimation can result in a 27% improvement in spectral performance, based on a 95% likely percentage.