RAS2 – Radio access performance optimization

Tuesday, 19 June 2018, 16:30-18:00, E3 hall
Session chair: Ingrid Moerman (Ghent University – imec, Belgium)


16:30 – Knapsack Optimisation Versus Cell Range Expansion for Mobility Load Balancing in Dense Small Cells

Karim M. Nasr (University of Greenwich & University of Surrey, United Kingdom (Great Britain)); Klaus Moessner (University of Surrey, United Kingdom (Great Britain))
We present a new approach for mobility load balancing (MLB) and user association in dense small cell scenarios. This Self Organizing Network (SON) approach relies on Knapsack Optimisation (KO) to evenly distribute users across a cluster of small cells subject to constraints. It is shown that the new technique referred to as (MLB-KO) achieves substantial improvements (better than four times reduction) in blocking ratios compared to the case when no MLB strategy is deployed. Comparisons with other approaches relying on Cell Range Expansion (CRE) and Almost Blank Subframes (ABS) are also presented highlighting the effectiveness of the new approach as a centralised self optimisation technique for future dense small cell network deployments.


16:52 – Load Balancing and Control Using Particle Swarm Optimisation in 5G Heterogeneous Networks

Tareq M. Shami (University of York); David Grace and Alister G. Burr (University of York, United Kingdom (Great Britain))
Most of users in heterogeneous networks (HetNets) associate to macro base stations (BSs) due to their high transmission power while small cell BSs are underutilized. Current research has addressed this load balancing problem based on biased user association where all small cell BSs that belong to the same tier, e.g femto or pico BSs are assigned a static biasing value in order to increase their coverage area. This work utilises particle swarm optimisation to assign each small cell BS a certain biasing value with the objective of maximising the achievable throughput and controlling the load per-BS. Simulation results show that the proposed PSO approach achieves better performance in terms of the achievable throughput. In addition, the proposed approach is able to provide fairness among users by controlling the load per-BS.



17:15 – Compute Resource Disaggregation: An Enabler for Efficient 5G RAN Softwarisation

Nikolaos Gkatzios (National and Kapodistrian University of Athens, Greece); Markos Anastasopoulos, Anna Tzanakaki and Dimitra Simeonidou (University of Bristol, United Kingdom (Great Britain))
This paper proposes the concept of compute resource disaggregation in centralized softwarised Radio Access Networks. This approach allows individual allocation of processing functions, to different servers depending on the nature and volume of their processing requirements. The benefits of the proposed approach are evaluated through experiments analyzing the BBU processing requirements of LTE using an open source suite for wireless systems and a purposely developed Multistage Linear Programming modeling framework. Our results show significant benefits of the proposed approach compared to the traditional solution in terms of energy efficiency.



17:37 – Model Predictive Network Control and Throughput Sub-Optimality of MaxWeight

Richard Schoeffauer (FU-Berlin, Heisenberg CIT Group); Gerhard Wunder (Freie Universität Berlin & Heisenberg Communications and Information Theory Group, Germany)
We present a novel control policy, called Model Predictive Network Control (MPNC) to control wireless communication networks (on packet level), based on paradigms of Model Predictive Control (MPC). In contrast to common myopic policies, who use one step ahead prediction, MPNC predicts the future behavior of the system for an extended horizon, thus facilitating performance gains. We define an advanced system model in which we use a Markov chain in combination with a Bernoulli trial to model the stochastic components of the network. Furthermore we introduce the algorithm and present two detailed simulation examples, which show general improved performance and a gain in stability region compared to the standard policy.