Multi-objective Optimization of Resource Allocation for Uplink Transmission in Two-Tier Heterogeneous Cellular Networks

With the rapid development of Internet of Things, spectrum efficiency (SE) and energy efficiency (EE) become two key indicators for the future green cellular networks, but it is hard to balance the tradeoff when maximizing them simultaneously. This paper consider the uplink resource allocation probl...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:2019 IEEE International Conference on Smart Internet of Things (SmartIoT) S. 275 - 282
Hauptverfasser: Wang, Jianhui, Liu, Haolin, Cao, Xianxian, Deng, Qingyong, Pei, Tingrui
Format: Tagungsbericht
Sprache:Englisch
Veröffentlicht: IEEE 01.08.2019
Schlagworte:
Energy efficiency
Internet of Things
Manganese
Multi-objective optimization
Optimization
Pareto-optimal
Spectrum efficiency
Online-Zugang:Volltext
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Beschreibung
Zusammenfassung:With the rapid development of Internet of Things, spectrum efficiency (SE) and energy efficiency (EE) become two key indicators for the future green cellular networks, but it is hard to balance the tradeoff when maximizing them simultaneously. This paper consider the uplink resource allocation problem of a two-tier heterogeneous cellular network whose macrocell base staion (MBS) and femtocell access points (FAPs) are operating under a shared spectrum scenario. Unlike traditional researches that use the Lagrangian dual method to convert a multi-objective optimization problem into a single-objective optimization problem by a weighted parameter, we propose a multi-objective memetic algorithm(MOMA) termed as MOMA_JUACAPA to achieve good balance in the EE-SE tradeoff subject to the quality-of-service (QoS) constraints for user equipments(UEs), and then the user association, the spectrum allocation and the power allocation are jointly optimized. What's more, a novel hybrid coding and two local search strategies are designed in the MOMA_JUACAPA to accelerate the speed of convergence and improve the distributivity meanwhile. The experimental results show that the performance of the MOMA_JUACAPA algorithm is better than the traditional NSGA-II in the EE and the SE, and the longest intercept method is adopted to get a knee point from the Pareto Front, which is considered as the best equilibrium solution among the whole Pareto-optimal set.
DOI:10.1109/SmartIoT.2019.00049