Low‐complexity networking based on joint energy efficiency in ultradense mmWave backhaul networks

AbstractAbstractAs operation cost and energy consumption problems deteriorate with network densification toward 5G, traditional radio resource management and networking strategies must be upgraded. It is observed that when the access network is dealing with a large amount of user traffic, the backhaul (BH) network turns out to be the new bottleneck hindering the cellular system from further increasing its capacity and efficiency. Although wired transmission technologies, such as passive optic fiber networks, are generally available between base stations, the resulting capital expenditure and operational expenditure are prohibitive in an ultradense deployment scenario. To redesign the BH network in 5G, multihop millimeter‐wave transmission is considered as a cost‐efficient and well‐performing candidate. In this paper, we propose a low‐complexity method based on millimeter‐wave transmission to minimize the energy consumption from both the access network and the BH network through joint admission‐power‐traffic control. Concretely, a self‐organizing heuristics is created through which the access and backhaul joint energy efficiency of both the base station and the whole system is maximized under the assumption of predefined user rate requirements. Simulation results show that our proposed method achieves significant reduction (∼95%) in algorithm execution time compared with the branch‐and‐bound search at the expense of little extra energy cost (∼7%). In addition, our proposed method satisfies the quality‐of‐service requirements of user equipment in the system with a lower transmission delay in terms of hops than that of the branch‐and‐bound search. A baseline algorithm called MR‐MPF (maximum reference signal receiver power admission control with minimum power flooding) is also compared to demonstrate our algorithm's superiority in energy saving. View Figure Our method is a low‐complexity heuristics comparing with the branch‐and‐bound (BnB) search with very limited extra energy cost (∼7%). Our method yields a lower transmission delay in terms of hops than the exhaustive search method, eg, BnB. Our method transforms the original MILP energy consumption minimization problem into a convex access and backhaul joint energy efficiency maximization problem, which is a novel design for the joint optimization of both the access and backhaul networks.