Joint Optimization of Power and Fronthaul Compression for Data and Pilot Signals in Uplink C-RANs

Abstract

For efficient implementation of 5G base stations, cloud-radio access network (C-RAN), where distributed units (DUs) are connected to a central unit (CU) through a fronthaul, has been considered. To support 5G with a wide bandwidth and large number of antennas, the fronthaul capacity becomes one of the limitations. To mitigate this problem, moving more blocks from the CU to the DU has been considered. One of the split options for the uplink is to deliver data and pilot symbols separately from a DU to a CU. For more efficiency in the fronthaul, the compression of data and pilot symbols has been adopted. As an objective of optimization, both the maximization of the uplink user rate and minimization of the fronthaul rate are considered. The optimization of the transmit powers and distortion variances of data and pilot symbols is considered. The closed-form expressions of the optimal variances of the data and pilot signal distortions with fixed data and pilot power are derived. Additionally, the optimal power allocation between the data and pilot signals are obtained when the data and pilot distortion variances caused by compression are fixed. Using an alternating optimization, the convergent solutions to the joint optimization problem can be obtained.