This work considers two‐way communication between each pair of users with highly delay‐aware applications. We formulate a joint uplink and downlink resource allocation problem in a cloud radio access network. Assuming average end‐to‐end (E2E) delay of each user pair and practical limitation such as maximum transmit power, we maximize the total throughput of all pair of users in the cloud radio access network. In this setup, we consider that each user can be connected to at most one remote radio head and a limited capacity fronthaul link between each remote radio head and baseband unit. To present the resource allocation problem in a more tractable manner, we replace the E2E delay limitation with its equivalent throughput‐based formulation. Due to inherent NP‐hard and nonconvex nature of the proposed problem, we apply successive convex approximation to reach a two‐step iterative algorithm where, in each step, a specific set of optimization variable derived while other variables are fixed. The problem of each step is transformed into the standard geometric programming via the arithmetic‐geometric mean approximation. Simulation results reveal that our proposed joint uplink‐downlink resource allocation algorithm outperforms a case that uplink and downlink resources are allocated separately in terms of total throughput and outage probability of E2E delay, ie, a chance that E2E delay does not hold.
The concept of cloud radio access network (C-RAN) architecture is being proposed to fully meet the requirements of 5G mobile networks. Thanks to the centralized cloud baseband unit (BBU), C-RAN reduces the energy consumption and cost of deployment significantly. However, it suffers from stringent fronthaul capacity and latency which are substantial in delay critical applications. Splitting up the processing functionalities between the control unit (CU) and distributed units (DUs) can mitigate fronthaul load and relax their requirements with the expense of an increase in power consumption. In this paper, we investigate joint access and fronthaul resource allocation problem for delay critical applications where the objective is minimizing the sum of normalized total power and fronthaul bandwidth consumption. We consider a downlink scenario and incorporate the total end-to-end delay components. For simplicity, linear models are assumed between function splitting (FS) levels and decreased fronhaul load as well as increased processing power. Different delay requirements affect our objective function and enforce a different FS level. We establish a flexible decision about the best FS level, which minimize our objective. Simulation results demonstrate that the delay constraint has a significant impact on the required fronthaul bandwidth and power consumption, which are directly related to the cost of the network. Moreover, flexible selecting function split level can achieve up to 40% gain in reducing the total utility (i.e., the sum of normalized total power and fronthaul required bandwidth).
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.