The Non-Coherent Ultra-Dense C-RAN Is Capable of Outperforming Its Coherent Counterpart at a Limited Fronthaul Capacity

Abstract

The weighted sum rate maximization problem of ultra-dense cloud radio access networks (C-RANs) is considered, where realistic fronthaul capacity constraints are incorporated. To reduce the training overhead, pilot reuse is adopted and the transmit-beamforming used is designed to be robust to the channel estimation errors. In contrast to the conventional C-RAN where the remote radio heads (RRHs) coherently transmit their data symbols to the user, we consider their non-coherent transmission, where no strict phase-synchronization is required. By exploiting the classic successive interference cancellation (SIC) technique, we first derive the closed-form expressions of the individual data rates from each serving RRH to the user and the overall data rate for each user that is not related to their decoding order. Then, we adopt the reweighted l1norm technique to approximate the l0-norm in the fronthaul capacity constraints as the weighted power constraints. A lowcomplexity algorithm based on a novel sequential convex approximation (SCA) algorithm is developed to solve the resultant optimization problem with convergence guarantee. A beneficial initialization method is proposed to find the initial points of the SCA algorithm. Our simulation results show that in the high fronthaul capacity regime, the coherent transmission is superior to the non-coherent one in terms of its weighted sum rate. However, significant performance gains can be achieved by the non-coherent transmission over the non-coherent one in the low fronthaul capacity regime, which is the case in ultra-dense C-RANs, where mmWave fronthaul links with stringent capacity requirements are employed.Index Terms-Ultra-dense networks (UDN), C-RAN, fronthaul capacity, pilot reuse, robust design.C. Pan, M. Elkashlan, and A. Nallanathan are with the Queen Mary

Intelligent Reflecting Surface Aided MIMO Broadcasting for Simultaneous Wireless Information and Power Transfer

“…However, due to the non-convexity of the EH constraint, Problem (15) is still non-convex. To resolve this issue, we observe that it can be viewed as a difference of convex (d.c.) program, which can be efficiently solved by the successive convex approximation (SCA) method [40]. In particular, we can approximate it by its first-order Taylor expansion.…”

Distributed Noncoherent Transmit Beamforming for Dense Small Cell Networks

“…We note that {γi}i∈N are achieved without phase synchronization between BSs. We also remark that {γi}i∈N are the aggregated instantaneous SINR, i.e., the total information received at Ui is log(1 + γi) [8] (the derivation of γi can also be found in [11]). We aim at designing beamforming vectors {v ik } i,k so that the WSR is maximized under the constraints of transmit power budget at the BSs.…”

“…We adopt the IA framework to develop efficient solution to (3), which was inspired by our earlier work in [19]. To do so, we first reveal the hidden convexity in (3) by transforming the problem into the following equivalent form 2 We note that the decoding order has no impact on γ i [11].…”

“…However, the requirement of strict network synchronization accuracy makes it difficult and high-cost to practically deploy. Therefore, the noncoherent JT has been received growing attention since it requires less strict network synchronization accuracy compared to coherent counterpart [8][9][10][11]. 1 In particular, beamforming designs for minimizing the power consumption subject to users' minimum data rate were investigated in [8,10].…”

“…The problem of load balancing was investigated in [9]. Very recently, beamforming design for the problem of weighted sum rate (WSR) under the limited fronthaul capacity for cloud radio access network was considered in [11]. Commonly, the prior works mentioned above mainly focus on centralized approaches, and the distributed algorithms are not available, while distributed architecture is more preferred in dense small cell networks [3].…”