Microwaves Research Group web banner 3a

Microwaves and Antenna Engineering Research Group

Pioneering the analysis, design and integration of high frequency electronics devices and systems

Prof. Tim O'Farrell, The University of Sheffield, UK 

Title: Energy Efficiency Evaluation Framework for Ultra-Dense Cellular Networks

Abstract: Small cell densification is an appealing technology to increase the capacity of cellular radio access networks (RANs). However, intense densification increases the network’s energy consumption, which is a key concern for 5G RAN deployment. In this talk, a joint energy and spectral efficiency evaluation framework is described based on three figures of merit: Data Volume Gain (DVG), Energy Consumption Gain (ECG), and Energy Efficiency Gain (EEG). The evaluation framework has been applied to a 5G RAN involving ultra-dense, homogeneous, small cell deployments. Our results indicate that, while small cell densification increases RAN capacity, as desired, the energy efficiency is compromised by the unbounded densification if no interference mitigation strategies are introduced. For example, compared to the baseline macro-RAN of 500 m inter-site-distance (ISD), the pico-RAN with ISD of 5 m could achieve a DVG of 117 when empty cells (i.e. cells with no users) do not transmit whereas this gain is reduced to 7 if empty cells continue to transmit signalling data. The RAN energy consumption increases monotonically with densification level unless empty cells are completely turned off, including the backhaul. Also, we observe that small cells, which typically have one user per cell, lose their multi-user diversity opportunity making round robin a sufficient scheduler policy for very small cells. The talk will explore the energy vs. spectral efficiency trade-off involved in ultra-dense small cell networks and the implications for network deployment.Abstract: Small cell densification is an appealing technology to increase the capacity of cellular radio access networks (RANs). However, intense densification increases the network’s energy consumption, which is a key concern for 5G RAN deployment. In this talk, a joint energy and spectral efficiency evaluation framework is described based on three figures of merit: Data Volume Gain (DVG), Energy Consumption Gain (ECG), and Energy Efficiency Gain (EEG). The evaluation framework has been applied to a 5G RAN involving ultra-dense, homogeneous, small cell deployments. Our results indicate that, while small cell densification increases RAN capacity, as desired, the energy efficiency is compromised by the unbounded densification if no interference mitigation strategies are introduced. For example, compared to the baseline macro-RAN of 500 m inter-site-distance (ISD), the pico-RAN with ISD of 5 m could achieve a DVG of 117 when empty cells (i.e. cells with no users) do not transmit whereas this gain is reduced to 7 if empty cells continue to transmit signalling data. The RAN energy consumption increases monotonically with densification level unless empty cells are completely turned off, including the backhaul. Also, we observe that small cells, which typically have one user per cell, lose their multi-user diversity opportunity making round robin a sufficient scheduler policy for very small cells. The talk will explore the energy vs. spectral efficiency trade-off involved in ultra-dense small cell networks and the implications for network deployment.

Prof Tim OFarrellBio: I received my B.Sc. degree in Electrical and Electronic Engineering from the University of Birmingham and my M.Sc. and Ph.D. degrees in Electrical and Electronic Engineering from the University of Manchester, respectively. Currently, I am Chair Professor of Wireless Communications at the University of Sheffield. 

My research is focused on wireless communications systems specialising in physical layer signal processing, radio resource management and wireless network planning. I have pioneered research on energy efficient mobile cellular communications, the mathematical modelling of CSMA based MAC protocols for WiFi, iterative block coding for wireless communication systems and spreading sequence design for CDMA wireless communications. 

I was an entrepreneur, being the cofounder and CTO of Supergold Communication Limited (1997-2007), a start-up that participated in the standardisation of IEEE 802.11g with the M-ary Bi-Code Keying proposal. In the framework of Mobile VCE, I was the Academic Coordinator of the Core 5 Green Radio project (2009-2012) and a leader in establishing energy efficiency as a global research field in wireless communications systems. 

I have led 21 major research projects as the principal investigator and participated in 2 as a coinvestigator with a total research spend of approximately £12M. My current EPSRC project portfolio includes the FARAD and SERAN projects as well as CommNet2. I have published 312 journal & conference papers, book chapters, patents and technical reports; graduated 23 Ph.D. and M.Phil. students; and participated in standards, consultancy and expert witness activities within the wireless sector. 

Currently, I am leading the UK Research Strategy Community Organisation in Communications, Mobile Computing and Networking within the EPSRC portfolio (www.commnet.ac.uk). I am also a Director of the mVCE (www.mobilevce.com), a Chartered Engineer and a member of the IET and IEEE.

 

Abstract: Small cell densification is an appealing technology to increase the capacity of cellular radio access networks (RANs). However, intense densification increases the network’s energy consumption, which is a key concern for 5G RAN deployment. In this talk, a joint energy and spectral efficiency evaluation framework is described based on three figures of merit: Data Volume Gain (DVG), Energy Consumption Gain (ECG), and Energy Efficiency Gain (EEG). The evaluation framework has been applied to a 5G RAN involving ultra-dense, homogeneous, small cell deployments. Our results indicate that, while small cell densification increases RAN capacity, as desired, the energy efficiency is compromised by the unbounded densification if no interference mitigation strategies are introduced. For example, compared to the baseline macro-RAN of 500 m inter-site-distance (ISD), the pico-RAN with ISD of 5 m could achieve a DVG of 117 when empty cells (i.e. cells with no users) do not transmit whereas this gain is reduced to 7 if empty cells continue to transmit signalling data. The RAN energy consumption increases monotonically with densification level unless empty cells are completely turned off, including the backhaul. Also, we observe that small cells, which typically have one user per cell, lose their multi-user diversity opportunity making round robin a sufficient scheduler policy for very small cells. The talk will explore the energy vs. spectral efficiency trade-off involved in ultra-dense small cell networks and the implications for network deployment.