Precoding designs under limited capacity fronthaul for C-RAN

Date

2023-8

Author

Odunkıran, Ahmet Furkan

Metadata

Show full item record

This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.

Item Usage Stats

99
views

66
downloads

Cite This

Massive multiple-input multiple-output antenna structures used in 5G systems greatly improve throughput and coverage. However, they also bring challenges in areas such as power consumption, complexity, and hardware. Centralized or cloud radio access network structures centralize baseband operations in a single unit and allow for the separation of baseband processing and radio frequency processing functions for a MIMO environment and allow for the design of multiple radio units with smaller array sizes compared to the traditional radio access networks which reduce the production cost. The fronthaul connection between the baseband and radio units in a radio network has strict capacity limits with common processing distribution interfaces between these units. Using different kinds of functional split procedures that are brought by technological improvements helps to reduce the load on fronthaul. This thesis considers a two-stage precoding approach for the precoding operation in a fronthaul capacity constrained wired fronthaul C-RAN system with multiple access points that have reduced complexities compared to a single access point system. First, a maximization problem that consists of the sum spectral efficiency of the users and its constraints has been stated. Then, the constraints are relaxed and added inside the main maximization problem using mathematical relations and the main maximization problem is expressed in a differentiable form. In this thesis, an optimization method over the differentiable spectral efficiency expression in a gradient-based form is proposed to maximize the sum spectral efficiency. The performance of the proposed method is discussed under different conditions.

Subject Keywords

Precoding, Fronthaul, Gradient descent, Cloud radio access network

URI

https://hdl.handle.net/11511/105313

Collections

Graduate School of Natural and Applied Sciences, Thesis