RAS1: Radio Access and Softwarisation

Tuesday, 16 June 2020, 12:15-14:30 CEST, Recommended re-viewing, https://www.youtube.com/playlist?list=PLjQu6nB1DfNBTsJLKGzYUHOHKxIpfVbDl

Tuesday, 16 June 2020, 12:15-17:00 CEST, Non-Live interaction (Chat),  link sent only to Registered people


Resource Allocation for a Reliable D2D Enabled Cellular Network in Factories of the Future

Idayat Sanusi (University of Greenwich, United Kingdom (Great Britain)); Karim M. Nasr (University of Greenwich & University of Surrey, United Kingdom (Great Britain)); Klaus Moessner (University of Surrey, United Kingdom (Great Britain))
We investigate and compare two radio resource management (RRM) approaches for a reliable Device to Device (D2D) enabled cellular network in an industrial environment. The D2D links have target Quality of Service (QoS) requirements in terms of minimum reliability and data rate while cellular users have minimum throughput requirements. The aim is to maximise the overall system throughput while ensuring the minimum (QoS) requirements are met. Interference is managed by jointly considering admission and power control based on interdistances between devices. The deferred acceptance (DA) game theoretic algorithm is used to perform matching and is compared to a centralised optimisation approach. Numerical simulations are used to verify the performances of the two algorithms.


QoS-Aware Spectrum Sharing for D2D Communication in Cellular Networks

Abubaker Matovu Waswa (Technische Universität Ilmenau, Germany); Stephen S. Mwanje (Nokia Bell Labs, Germany); Jens Mueckenheim (University of Applied Sciences Merseburg, Germany); Andreas Mitschele-Thiel (Ilmenau University of Technology, Germany)
The continued proliferation of new mobile applications, e.g. autonomous driving, smart cities/homes and tactile internet has increased the number and density of mobile devices requiring wireless connectivity. This creates demand for efficient utilization of the available frequency spectrum apportioned to cellular networks. Device-to-device communication (D2D) as an underlaying transmission within cellular networks presents improved spectral efficiency through spectrum sharing with appropriate management of the resultant interferences. This paper proposes two resource allocation schemes that facilitate the assignment of suitable D2D-pair subsets to share the different cellular users’ spectrum in a given cell while concurrently meeting all users’ quality of service (QoS) requirements.


5G-RANGE: A Transceiver for Remote Areas Based on Software-Defined Radio

Wheberth Damascena Dias, Alexandre Ferreira, Roberto Kagami, Juliano Silveira Ferreira, Daniely Gomes Silva and Luciano Leonel Mendes (Inatel, Brazil)
The 5G-RANGE project has as the main goal to develop a cost-effective solution to enable the deployment of Internet connectivity on long-distance and low populated areas. This work presents a 5G-RANGE PHY implementation based on SDR. This approach aims to reduce the network deployment cost by using affordable off-the-shelf computer hardware, reducing the barrier for small operators to deploy such networks and promote competition. Another benefit of the SDR is its ability to upgrade to future releases, and reduced development cycle time, compared with hardware development.
This paper also describes some of the high-performance tools and techniques used in the development and the system performance achieved with the proposed approach.


On the Impact of Normalized Interference Threshold for Topological Interference Management

Hassan Kallam (Université de Lyon & INRIA, INSA Lyon, CITI-INRIA, France); Leonardo S. Cardoso (Université de Lyon & INRIA, INSA-Lyon, CITI-INRIA, France); Jean-Marie Gorce (INSA-Lyon & CITI, Inria, France)
This paper presents a new formulation to build an interference topology for the multi-user unicast Topological Interference Management (TIM) based wireless network problem. Based on our interference topology formulation, we are able to evaluate the achievable rate’s theoretical limit, in the asymptotic signal to noise ratio (SNR) regime, for the underlying wireless network and not just for its topological interference representation. This new formulation allows us to cope with the finite SNR regime and not just with the asymptotic SNR regime with the Degrees of Freedom (DoF) analysis. A new SNR independent interference threshold parameter is proposed and we evaluate the achievable symmetric rates of the wireless network in both the finite SNR regime and the asymptotic SNR regime. Finally, we present outer bound solutions on the new normalized interference threshold parameter for interference topologies with half-DoF-feasibility, considering both an orthogonal resource allocation and Interference Alignment (IA). These bounds specify if a given half-DoF-feasible interference topology can be, in terms of the achievable rate, the best topology or not. Using this result, we limit the search space in the normalized interference threshold parameter range, to find half-DoF-feasible interference topologies having the possibility to be the best topologies in terms of the achievable rate.


Unsupervised Learning for Detection of Mobility Related Anomalies in Commercial LTE Networks

Jessica Moysen (Elisa Oyj, Finland & Fundaciò i2CAT, Spain); Furqan Ahmed (Elisa Corporation, Finland); Mario Garcia-Lozano (Universitat Politècnica de Catalunya, Spain); Jarno Niemelä (Elisa Corporation, Finland)
We propose an unsupervised learning based anomaly detection framework for identifying cells experiencing performance degradation due to mobility problems, in LTE networks. Handover failure rate is used as a performance metric, whereas the mobility problems considered include too early and too-late handovers. In order to facilitate unsupervised learning, the framework leverages existing datasets in commercial LTE networks (e.g. performance management counters, configuration
management data, geographical locations, and inventory etc). To this end, the first step is data pre-processing, followed by feature extraction based on principal component analysis and clustering. For implementation, we use real data from
an operational commercial LTE network. Results show that clustering is highly effective in understanding and identifying mobility related anomalous behaviour, and provides actionable insights for automation and self-optimization, paving the way for efficient mobility robustness optimization, which is an important self-optimization use-case for contemporary 4G/5G networks.


Outage Prediction for URLLC in Rayleigh Fading

Andreas Traßl (Technische Universität Dresden & Centre for Tactile Internet with Human-in-the-Loop, Germany); Lucas Scheuvens (TU Dresden, Germany); Tom Hößler (TU Dresden & Barkhausen Institut, Germany); Eva Schmitt (TU Dresden, Germany); Norman Franchi and Gerhard P. Fettweis (Technische Universität Dresden, Germany)
A major challenge for the realization of URLLC systems is the fast fading of wireless channels and the therewith associated outages. To overcome the fading, diversity concepts, as well as transmissions with low modulation order and low code rate are widely considered. However, these methods have an undeniably high resource consumption. In order to reduce it, fading prediction methods can be utilized as part of the scheduling strategy.We propose a novel outage prediction approach based on a Wiener filter that significantly helps to increase URLLC system monitoring and radio resource scheduling. In this paper we focus on Rayleigh fading channels and classical Doppler spectra. The proposed outage predictor is found to work on a wide range of sampling times and also with only a few observed fading samples available. During operation, future outage probabilities can be calculated beforehand and reliability guarantees can be made. Utilizing the proposed outage prediction method increases the achievable reliability of a link by several orders of magnitude compared to best effort transmissions on a single link. Links that are operable are well identified when little movement is present and prediction of several milliseconds is sufficient.