FrA8 – Radio Resource Management

Friday, 21 June 2019, 8:30-10:30, Room 8


Session Chair: Jordi Pérez-Romero (Universitat Politècnica de Catalunya (UPC), Spain)


Elastic Slice-Aware Radio Resource Management with AI-Traffic Prediction

Sina Khatibi (NOMOR Research GmbH, Germany); Alba Jano (Nomor Research GmbH, Germany)
Network virtualisation and network slicing are the two essential innovations in the next generation of mobile networks also known as the 5G network. Based on these innovations, multiple network slices with different requirements and objectives can share the same physical infrastructure. The techniques to efficiently allocate the available radio resources to different slices based on their requirements and their priority, also known as inter-slice radio resource management, has been the subject of many studies. The formerly proposed algorithms either assume the slices request maximum contracted data rates or they react passively as the demands arrive. This paper proposed to use Artificial Intelligence (AI) approaches to learn the pattern of the traffic demand of each network slices and predict the demands in the next decision interval. Based on the prediction of the slices’ demands, a novel model for elastic inter-slice radio resource management has been proposed to increase the multiplexing gain while not compromising the quality of offered connectivity services to the slices. The proposed model has been evaluated using a practical scenario. The numeric results show that while the performance of the model under full demand is similar to former models, its elastic resource management enables more efficient resource allocation when the traffic demands vary during the time.


C-RAN Employing xRAN Functional Split: Complexity Analysis for 5G NR Remote Radio Unit

Jay Kant Chaudhary and Atul Kumar (Technische Universität Dresden, Germany); Jens Bartelt (Airrays, Germany); Gerhard P. Fettweis (Technische Universität Dresden, Germany)
Fronthaul (FH) bandwidth in C-RAN can be significantly reduced with an appropriate functional split by offloading more signal processing functionalities to the remote radio unit (RRU). However, this not only reduces the acclaimed centralization benefits but also increases the complexity of the RRU. Considering the practical aspects such as size, weight, power consumption, it is often desirable to make RRU as simple, yet efficient, as possible. In this paper, we compute the computational requirement of the RRU with 5G New Radio (NR) considering functional split 7.2 as recently standardized by the xRAN Forum. C-RAN with mix numerology allows to support a wide range of scenario and use-case specific requirements. In addition, we compare suitability in terms of efficiency and flexibility of the RRU being implemented using either FPGA or GPP considering their computational requirement. Based on the complexity analysis, we calculate the required number of the FPGA or GPP to handle the complexity of the RRU. We show that FPGA is more feasible option compared to x86 in terms of form factor and power consumption particularly for rooftop mounted RRU.


Edge Sectors Detection in Mobile Communications Networks

Omar Kaddoura (Ericsson & University of Malaga, Spain); Inmaculada Serrano and Juan Sánchez-Sánchez (Ericsson, Spain); Raquel Barco (University of Malaga, Spain)
This paper proposes a method to detect edge sectors in mobile communication networks. This detection has applications such as identifying results of geolocation techniques presenting low accuracy or reducing false positives in troubleshooting algorithms. On the way to achieve the desired goal, the method also finds the nodes which are part of convex and concave regions in the contour of the network. The proposed method consists of three phases: independent areas identification, contour nodes detection and edge sectors detection. All these phases can be parameterized according to the pursued objective. Experiments have been carried out to prove the goodness of the proposed method to detect sectors in which location techniques like Received Signal Strength used to estimate the location of mobile users present high error figures. Results obtained prove the validity of the proposed method.


5G Component Carrier Management Evaluation by Means of System Level Simulations

Ioannis-Prodromos Belikaidis, Andreas Georgakopoulos and Evangelos Kosmatos (WINGS ICT Solutions, Greece); Isabel de la Bandera-Cascales (University of Málaga, Spain); David Palacios (Tupl Inc., Spain); Raquel Barco (University of Malaga, Spain); Panagiotis Demestichas (University of Piraeus, Greece)
This work presents essential aspects of 5G system level simulator for enabling advanced component validations and optimizations. System level simulations in the 5G era, consider demanding use cases with high load and very limited latency in order to cover services such as enhanced mobile broadband (eMBB), massive machine-type communications (mMTC) and ultra-reliable low-latency communications (URLLC). As such, appropriate configuration, environment and network models need to be defined in order to proceed to performance evaluation. The system-level simulation platform is a discrete event simulation environment for the simulation of heterogeneous networks which is extended with new features to support the new functionalities of 5G. To show the potential of this simulation tool, in this work, a framework for multi-connectivity management has been integrated and assessed in a load-imbalanced scenario. Simulation results show how a proper assignment of component carriers (CCs) in this situation allows increasing the users’ throughput by up to a 60% when compared to a simple received power scheme for link management.


Multi-Numerology Based Resource Allocation for Reducing Average Scheduling Latencies for 5G NR Wireless Networks

Tanmoy Bag and Sharva Garg (Ilmenau University of Technology, Germany); Zubair Shaik (Technische Universität Ilmenau, Germany); Andreas Mitschele-Thiel (Ilmenau University of Technology, Germany)
The introduction of 5G New Radio (NR) promises flexibility and adaptability to cater a wide variety of services. One of the key ingredients for achieving such requirements is the employment of multiple numerologies. This versatility brings along with it some complexity as well in terms of resource allocation using multiplexed non-orthogonal numerologies. Another key enabler is the concept of Shortened TTI/ Mini-slots for addressing low latency applications. In this paper, we propose a dynamic resource allocation scheme to reduce the average scheduling latencies in a network using multiple numerologies in conjunction with mini-slots. A system level simulation has been conducted to compare different nominated combinations of numerologies (with mini-slots) and service requests and it has been observed that with a minor compromise of minimum achievable scheduling latency for high priority requests, a significant reduction of average scheduling latency is attained considering all types of diverse service requests with different priorities.


Profit-Based Radio Access Network Slicing for Multi-tenant 5G Networks

Jordi Pérez-Romero (Universitat Politècnica de Catalunya (UPC), Spain); Oriol Sallent, Ramon Ferrús and Ramon Agustí (Universitat Politècnica de Catalunya, Spain)
Network slicing is a key capability of 5G networks that facilitates the provision of multi-tenancy by allocating different slices to the tenants that share a certain infrastructure according to specific Service Level Agreements (SLAs). In this context, this paper focuses on radio admission control as the function that controls the amount of radio resources assigned to the different tenants in a 5G radio access network. Specifically, a novel approach is proposed that includes profit-related metrics in the decision-making process, accounting for the additional extra incomes that can be obtained by sporadically granting additional capacity beyond the SLA level and for the penalties incurred due to potential SLA breaches. The proposed approach is evaluated by means of simulations to assess its benefits in terms of achieved profit and throughput.