RAT2

  • Tuesday, 13 June, 16:30-18:00, Room Library Auditorium
  • Session Chair: Luciano Leonel Mendes (Inatel, Brazil)

 

 

16:30 The Relaxed Power Control Algorithm

Markus Klügel, Michael Newinger, Wolfgang Utschick and Wolfgang Kellerer (Technische Universität München, Germany)

With the progress of wireless communication technology, networks become more and more interference limited. One core aspect for guaranteeing required signal qualities in interference limited networks is power control. A simple, yet powerful power control algorithm (PCA) is Yates' iterative \textit{standard} PCA, which is well known and can be applied to various physical layer technologies. However, for some technologies, e.g., multi-antenna (MIMO) communication systems, its update rules cannot be explicitly calculated, thus rendering the power control problem complicated. For these cases we introduce \textit{relaxed} PCAs, whose update steps are within certain bounds around that of the standard PCA. We prove that relaxed PCAs and standard PCAs converge to the same fixed points. We apply the introduced algorithm to MIMO wireless networks and develop two manageable power control update rules. We verify their convergence behavior with simulations.

 

16:48 Comparison of Interference Control Methods in Large Heterogeneous Networks

Ole Grøndalen and Kashif Mahmood (Telenor, Norway); Olav Norvald Østerbø (Telenor Corporate Development, Norway)

Efficient interference control is a prerequisite for realizing the possible capacity gains in large heterogeneous networks. An efficient way to do this is to mute some Base Station (BS) transmissions in some time frames to let other BSs serve their users. Two approaches for doing this is studied; a simple one similar to the schemes used in current mobile technologies and a more complex one with higher performance gains. The complex approach jointly optimizes user attachment, scheduling and muting and requires that a large convex optimization problem is solved. By exploiting structure and sparseness properties it is shown that the computational complexity can be significantly reduced, so that the method can be feasible even in relatively large networks. The performance of the two approaches were compared in a simulation study. It was found that the simple method gave significant performance improvements compared to the case without interference control and that it offered flexibility to trade off improved edge user throughput against a lower average user throughput. The more complex approach gave better performance, even though the muting configuration candidates were selected randomly. The results indicate that significant improvements are possible by selecting the configuration candidates in a more optimal way.

 

17:06 User Satisfaction Based Resource Allocation Schemes for Multicast in D2D Networks

Jagadeesha Rb, Jang-Ping Sheu and Wing-Kai Hon (National Tsing Hua University, Taiwan)

The D2D communication has high potential to serve multiple users with high data rate within a proximity in the next generation cellular networks. In this paper, we consider a multicast scenario of D2D users where each user wishes to receive the same multicast data at varying data rates. Based on our best knowledge, finding an optimal solution to satisfy the user request in the said context requires unreasonable time. Therefore, we have proposed approximation algorithms with two objectives (i) maximize the satisfied throughput, (ii) maximize the number of satisfied users when the available resource blocks are limited. The simulation results show that the proposed algorithms offer a worst-case performance guarantee and outperform the other conventional schemes in terms of throughput, satisfied users count, and fairness.

 

17:24 System Level Analysis of Multi-Operator Small Cell Network at 10 GHz

Petri Luoto (University of Oulu, Finland); Antti Roivainen (Keysight Technologies Finland Oy, Finland); Mehdi Bennis (Centre of Wireless Communications, University of Oulu, Finland); Pekka Pirinen (University of Oulu, Finland); Sumudu Samarakoon (Centre for Wireless Communications, University of Oulu, Finland); Matti Latva-aho (UoOulu, Finland)

Due to higher cost and spectrum scarcity, it is expected that an efficient use of spectrum in fifth generation (5G) networks will rather rely on sharing than exclusive licenses, especially when higher frequency allocations are considered. In this paper, the performance of a dense indoor multi-operator small cell network at 10 GHz is analyzed. The main goal is to show the benefits obtained at higher carrier frequency due to network densification while mobile network operators are sharing the spectrum. The analysis is assessed through extensive system level simulations. The main performance metrics are user throughput and signal-to-interference-and-noise ratio. Results show that when 10 GHz carrier frequency is used it allows higher network densities while satisfying user throughput requirements. However, when network is sparse lower carrier frequency leads to better performance. When network is dense, on average 2 Mb/s better mean throughput is achieved at 10 GHz when compared to traditional cellular frequency.

 

17:42 Joint Optimization of Energy Efficiency and Spectrum Efficiency in 5G Ultra-Dense Networks

Mary Adedoyin and Olabisi Emmanuel Falowo (University of Cape Town, South Africa)

The heterogeneous deployment of ultra dense small cells such as femtocells in the coverage area of the traditional macrocells is seen as a cost-efficient solution to provide network capacity, indoor coverage and green communications towards sustainable environments in the fifth generation (5G) wireless network. However, the unplanned and the ultra-dense deployment of femtocells will lead to increase in total energy consumption, cross-tier interference (interference between macrocells and femtocells), co-tier interference (interference between neighbouring femtocells) and inadequate QoS provisioning. Therefore, there is a need to develop a radio resource allocation (RRA) algorithm that will jointly maximize the energy efficiency (EE) and the spectrum efficiency (SE) of the overall networks. Unfortunately, maximizing the EE results in low performance of the SE and vice versa. This paper investigates how to balance the trade-off that arises when maximizing both the EE and the SE simultaneously. The joint EE and SE maximization problem is formulated as a multi-objective optimization problem (MOP), which is later converted into a single-objective optimization problem (SOP) using the weighted sum method. An iterative algorithm based on the Lagrangian dual decomposition (LDD) method is proposed. Simulation results show that the proposed algorithm achieves an optimal trade-off between the EE and the SE with fast convergence.