# GigaMobile: Gigabit Mobile Networking using Incentivised Operator Controlled Device-to-Device Communications

Lead Research Organisation: Queen's University of Belfast
Department Name: Electronics Electrical Eng and Comp Sci

### Abstract

The increasing popularity of smartphones and other mobile devices, where the end user has access to a host of rich multimedia functionality, means that the current mobile network architecture is struggling to meet surging data demands. Smartphone ownership in the UK alone has risen from 38% of the population in early 2010 to 60% in 2013 with the ONS reporting that over 32% of adults now access the Internet using their smartphone every day. This figure is expected to grow significantly in the coming years with Cisco predicting that worldwide demand for mobile data traffic will outstrip fixed data, reaching 11.2 Exabytes per month by 2017. Although the global rollout of 4G networks is well underway, it is unlikely that that 4G alone will be able to service the growing data requirements of mobile users. Furthermore, while voice, data, and compressed streaming media are now the norm, it is future social networking applications which will undoubtedly present mobile network designers and operators with their greatest challenge.

Both Google and Samsung, through their Glass and Galaxy Gear based concepts respectively, have given us a glimpse of some of the exciting new pervasive technologies that will push the boundaries on the maximum rate at which data can be communicated over mobile networks. For example, using Google Glass, users will no longer just shop and download compressed audio and video, instead they will be immersed into a completely new augmented reality in which they can share their immediate perception and senses with friends and colleagues in the cloud. While this technology will revolutionise social networking it will add further stress to already overburdened mobile networks. To avoid future congestion caused by this huge influx of data, a major change in the way mobile networks are setup and operated is required. However, considerable academic, industrial and regulatory challenges remain and this is the focus of the proposed research programme.

To help overcome the future "communications bottleneck" in mobile systems, this project proposes a new paradigm for ultra-high capacity mobile networks by simultaneously and jointly addressing the bandwidth problem and the dynamic network management issues associated with device-to-device communications. Combining the complementary expertise of research teams in Queens University Belfast and Cardiff University, this project will focus on understanding and exploiting incentivised, multimode user equipment operating as an ultra-high capacity underlay network featuring real-time opportunistic adaptive routing all overseen by a context aware mobile network infrastructure.

The significance and potential impact of the proposed project for the UK cannot be overstated. The original concept of using cellular radio based D2D communications to supplement mobile communications has been pioneered in the UK although at the time of conception it would have been difficult to envisage the explosion in data requirements that would occur in the future, particularly from social and multimedia applications, and hence the higher capacity technologies that would be required. It is therefore critical that the UK continues to remain at the forefront of future evolutions in this area and is the first to adopt and implement new cutting-edge technologies that will help address the mobile networking aspect of "the communications bottleneck". The new high capacity D2D links proposed in this project will pave the way for a host as of yet unimagined mobile, gaming and social networking applications which will have the potential to revolutionise society in the UK. According to a recent Gartner report, worldwide revenue from app stores will increase in 2013 by 62%, bringing the total industry revenue to $25 billion dollars. Clearly, using the UK as a test bed for multi-gigabit mobile networking, the mobile apps industry in the UK will gain a significant advantage over their international competitors and be first in line to reap the economic benefits. To further enhance the economic benefits, both institutions will work closely with their respective technology transfer offices and industrial collaborators to ensure the rapid transfer of knowledge to the strategically important ICT sector. Directly relevant to this call, this project will also aim to maximise impact by addressing the following sub-challenges: Protocols for a 21st Century Internet - the work conducted here, including the channel models and network simulator will inform global standardisation bodies and alliances (such as 3GPP and WiGig) and provide them with the necessary tools and knowledge required to design, test and make recommendations for future wireless protocols and hardware to be used for ultra-high bandwidth D2D communications. Additionally crowd-sourcing of these distributed resources for communication will be addressed; Energy Efficient Communication and Data Systems - by using operator controlled D2D communications, we can off-load some of the traffic from the base station and support efficient data transportation as local data stays local, i.e., doesn't have to routed through the eNodeB. This not only saves power, but also frees up precious data bandwidth and spectrum space. By allowing the D2D communications to be controlled we will also be able to implement judicious adaptive power control mechanisms. Both of these measures will lead to lower power mobile networks, reducing the UK's overall carbon footprint and helping towards future goals in carbon emission; Building Context and Content Aware Networks - by listening to on-going transmissions during idle periods, UEs can build up a picture of surrounding networks which use similar WiGig technology such as operator controlled 'hotspots' in city centres that may provide an alternative route to the cloud. This can then be reported to the eNodeB which in turn can create a synthesis which combines the optimum configuration of all available networks (dependent on data type/content) to route data; Seamless Mode Adaptive Communications - by allowing the operator to control D2D communications we reduce the complexity of the network setup and maintenance associated with truly mobile ad hoc networks. Furthermore, because the operator controls the technology used to implement cellular communications i.e. 2.45/5 or 60 GHz D2D, UE to eNodeB or UE to local WiGig hotspot etc., the switching between networks will completely transparent to the end user. ### Organisations ### Publications 10 25 50 Bhargav N. (2015) Signal reception characteristics in the proximity of alice and bob for secure indoor peer-to-peer communications at 2.45 GHz in 2015 9th European Conference on Antennas and Propagation, EuCAP 2015 Description It is broadly recognised that the current generation of mobile networks is being put under growing strain due to a surge in the number of wireless devices, many of which are consuming ever increasing amounts of data. This has led to the evolution of the so-called fifth generation (and beyond) of wireless technologies, commonly referred to as '5G', which, among many goals, aim to provide ubiquitous, reliable, high-speed and low-latency links to network users. In this project we sought to address the requirement for ultra-high capacity communications in mobile networks by simultaneously and jointly addressing the bandwidth problem and the dynamic network management issues associated with device-to-device (D2D) communications. Combining the complementary expertise of research teams in Queen's University Belfast and Cardiff University, industrial collaborators Keima, O2/Telefónica UK and Samsung, the project focused on understanding and exploiting incentivised, multimode user equipment operating as an ultra-high capacity underlay to future mobile networks. This was achieved through real-time opportunistic adaptive routing all overseen by an operator-controlled context aware mobile network infrastructure. The first stage of the work addressed the impact that human device users can have on D2D links. We were able to demonstrate that unlike the microwave operating frequencies used by technologies such as Wi-Fi and Bluetooth, the human body and how the user equipment (UE), in this case a mobile handset, is handled can have a huge determination on whether D2D communications will be successful. For example, even over short distances of a few metres when both D2D users are facing in opposite directions, it may not be possible to achieve reliable communications. This situation is exacerbated when one or both users become mobile. Through the multi-frequency measurements conducted within the work, we have been able to devise how closely people are located within D2D networks, simply by observing the instantaneous connectivity information. Also building upon the experimental work conducted during the project, we developed a number of measurement informed stochastic geometry based mobile network models. These allowed us to fully encapsulate the physical effects of the D2D channel and at the same time investigate different strategies such as underlaid and overlaid D2D communications and how these can enhance the capacity of the future infrastructure-based mobile networks. We observed that the D2D link provides higher rates than those of the cellular link when the spectrum partition factor was appropriately chosen. However, depending on the network settings, while encouraging more UEs to use the D2D mode will increase the average rate, it may increase the potential for higher levels of interference and degraded outage probability. This work was summarised in two papers which were published in IEEE Transactions on Wireless Communications, both of which remained in the popular documents (most downloaded) list of the publisher's website for a significant number of months after first appearing. Relating to the human behavioural aspect of the work, a discovery was made of a fundamental nature. In particular, a human inspired disposition to make social comparison of reputations was found important to the problem of cooperation for scenarios involving donation of resources to third parties that might not be encountered again, such as in D2D cooperation. The research found that a simple heuristic, donating to those with at least the same reputation, was sufficient to promote both cooperation for collective and payoff for the individual. This was also found to be implicitly present in previous research. The associated paper describing the work was published in Nature Scientific Reports and received attention from 50 media outlets across the globe, including the USA, Europe and India. It is worth highlighting that cooperation of this nature also has relevance for future machine based and autonomous systems, such as robotics. Additionally, it is also relevant to the emerging field of human-machine cooperation. All of the aforementioned outcomes were combined in the final output of the project, which integrated social comparison with decision to carry (equivalently relay) information for other user's in D2D communications. Prior to the outcomes of this project, many of the D2D related contributions which appear in the literature, make the assumption that that relay nodes cooperate unequivocally. This can greatly overestimated things like capacity, especially since the natural human instinct can be to preserve resources and therefore defect rather than cooperate. Through our work, we investigated the performance gap between the ideal case of 100% cooperation and practical scenarios with a lower cooperation probability. We have demonstrated that practical scenarios achieve lower transmission capacity and higher outage probability than idealistic network views, which assume full cooperation. However, after a sufficient number of generations, the cooperation probability will follow the natural rules of evolution and the transmission performance of practical scenarios approach that of the full cooperation case. This indicates that all D2D relay nodes adapt the same dominant cooperative strategy based on social comparison, without the need for external enforcement. This work was also published in IEEE Transactions on Wireless Communications and at the time of writing this text (March 2018) has been in the most popular paper list since first appearing on the publisher's website (December 2017), achieving 763 downloads. Exploitation Route Our major industry partners (Samsung and O2) provided strong support throughout the project and were enthusiastic with the outcomes. It is in these areas where this work will most likely be exploited first, i.e. by those involved in digital/communication/information technologies such as mobile device manufacturers, network operators and standards bodies. For example, the measurements and subsequent channel models may be used immediately to inform mobile hardware manufacture. A side product of the work (and not explicitly mentioned in the key findings above) was the development of a number of new statistical distributions (i.e.$\kappa$-$\mu$/ inverse Gamma and$\eta$-$\mu\$ / inverse Gamma models). While clearly applicable in the immediate area of wireless communications, they may also find use in other areas of engineering and science where the statistical modelling of physical data is required. The results of the work, will also be beneficial for those working on the interaction of human behaviour and the cooperation of (relay) nodes in wireless networks as they now have models to characterise these effects. Similarly, the measurement informed system models which have been proposed may be used by network designers to infer the capacity and area spectral efficiency for future D2D networks.
Sectors Digital/Communication/Information Technologies (including Software),Electronics

Description Centre for Defence Enterprise
Amount £110,000 (GBP)
Funding ID CDE40578
Organisation Defence Science & Technology Laboratory (DSTL)
Department Centre for Defence Enterprise
Sector Public
Country United Kingdom
Start 01/2016
End 09/2016

Description US-Ireland R&D Partnership Programme
Amount £320,000 (GBP)
Funding ID USI080
Organisation Government of Northern Ireland
Department Department for Employment and Learning Northern Ireland (DELNI)
Sector Public
Country United Kingdom
Start 03/2016
End 09/2019

Description Distributed Analytics and Information Sciences International Technology Alliance (DAIS-ITA)
Organisation IBM
Department IBM T. J. Watson Research Center, Yorktown Heights
Country United States
Sector Private
PI Contribution Development of models capturing cooperation behaviours for socio-technical modelling.
Collaborator Contribution Collaborative links to Army Research Laboratory and DSTL. This is now supporting three PhD studentships (2017-18 start).
Impact Collaborative links to USA partners.
Start Year 2016

Description New Research Link with Prof. Allen MacKenzie at Virginia Tech, USA
Organisation Virginia Tech
Country United States
PI Contribution We have worked with Prof. Mackenzie at Virginia Tech to help them attract funding from NSF as part of a tri-party funding arrangement between the USA, Ireland and Northern Ireland under the auspices of the SF-US-Ireland R&D Partnership.
Collaborator Contribution Prof. Mackenzie at Virginia Tech worked with us to attract funding from DEL NI as part of a tri-party funding arrangement between the USA, Ireland and Northern Ireland under the auspices of the SF-US-Ireland R&D Partnership.
Impact See funding part of report.
Start Year 2015

Description New Research Link with Prof. George K. Karagiannidis at Aristotle University of Thessaloniki, Greece
Organisation Aristotle University of Thessaloniki
Department School of Electrical and Computer Engineering
Country Greece
PI Contribution The PI (Simon Cotton) and a PhD student working closely with the GigaMobile Project have worked together with Prof. George K. Karagiannidis to develop two new fading models that will be used to characterise the device-to-device channels.
Collaborator Contribution Prof. George K. Karagiannidis has provided assistance with the development of physical signal models and communications performance measures. This has resulted in two conference papers which were presented at PIMRC 2015.
Impact See publications list
Start Year 2015

Description New Research Link with Prof. Harpreet Dhillon at Virgina Tech, USA
Organisation Virginia Tech
Country United States
PI Contribution The PI (Simon Cotton), PDRA (Young Jin Chun) and a PhD student working closely with the GigaMobile Project have worked together with Prof. Harpreet Dhillon at Virginia Tech in the US. We have had two notable project related outcomes, namely two research papers which were published in IEEE Transactions on Wireless Communications (listed below). It should be noted that these two papers have remained in the Top 50 most downloaded list since first publication.
Collaborator Contribution Prof. Dhillon provided much needed guidance on the system model and helped with the mathematical derivations that were required. He also assisted with the preparation of the journal papers.
Impact [1] Y. J. Chun, S. L. Cotton, H. S. Dhillon, F. J. Lopez-Martinez, J. F. Paris and S. K. Yoo, "A Comprehensive Analysis of 5G Heterogeneous Cellular Systems Operating Over ?-µ Shadowed Fading Channels," IEEE Transactions on Wireless Communications, vol. 16, no. 11, pp. 6995-7010, Nov. 2017. [2] Y. J. Chun, S. L. Cotton, H. S. Dhillon, A. Ghrayeb, M. O. Hasna, "A stochastic geometric analysis of device-to-device communications operating over generalized fading channels," IEEE Transactions on Wireless Communications, vol. 16, no. 7, pp. 4151-4165, July 2017.
Start Year 2016

Description New Research Link with Prof. John Thompson at the University of Edinburgh
Organisation University of Edinburgh
Department Edinburgh Neuroscience
Country United Kingdom
PI Contribution We are currently working with Prof.Thompson and a PDRA at the University of Edinburgh to incorporate some of our mm-wave channel characteristics/models obtained from the Gigamobile project with findings from an EPSRC project they are leading (which was also funded under the same TI3 call).
Collaborator Contribution See above
Impact Still working on integration of the mm-wave channel models developed by QUB with the mm-wave device-to-device system models developed at the University of Edinburgh.
Start Year 2016

Description New Research Link with Prof. Mikko Valkama at Tampere University of Technology, Finland
Organisation Tampere University of Technology
Country Finland
PI Contribution The PI (Simon Cotton) and a PhD student working closely with the GigaMobile Project have worked together with Prof. Mikko Valkama to develop two new fading models that will be used to characterise the device-to-device channels.
Collaborator Contribution Prof. Mikko Valkama has provided assistance with the development of physical signal models and communications performance measures. This has resulted in two conference papers which were presented at PIMRC 2015.
Impact See publications
Start Year 2015

Description New Research Link with Prof. Robert Heath Jr. at the University of Texas at Austin
Organisation Duke University
Department Electrical and Computer Engineering
Country United States
PI Contribution The PI (Simon Cotton), PDRA (Young Jin Chun) and a PhD student working closely with the GigaMobile Project have worked together with Prof. Robert Heath Jr. at the University of Texas at Austin to characterise the small cell UE to eNB channel at 60 GHz.
Collaborator Contribution Prof. Heath provided expert input on the millimetre-wave systems aspects of the work. This has results in a publication in IEEE Wireless Communications Letters.
Impact See publications.
Start Year 2016

Description Research Link with Prof. Claude Oestges at the Université catholique de Louvain, Belgium
Organisation Catholic University of Louvain
Country Belgium
PI Contribution The PI (Simon Cotton) and a PhD student working closely with the GigaMobile Project have worked together with Prof. Claude Oestges at Université catholique de Louvain in Belgium. Through this collaboration we have been able to work on and publish the first known results relating to the impact of vehicular traffic on wearable-to-wearable (device-to-device) communications. This work has been featured in IEEE Transactions on Antennas and Propagation.
Collaborator Contribution Prof. Oestges provided significant input on experimental design and the analysis / modelling of the data. He also assisted with the preparation of subsequent the journal paper.
Impact M. G. Doone, S. L. Cotton and C. Oestges, "An Experimental Investigation into the Impact of Vehicular Traffic on Interpersonal Wearable-to-Wearable Communications Channels," IEEE Transactions on Antennas and Propagation, vol. 65, no. 10, pp. 5418-5430, 2017.
Start Year 2014

Description Research Link with Prof. Michel Yacoub at the University of Campinas, Brazil
Organisation State University of Campinas
Country Brazil
PI Contribution The PI (Simon Cotton), PDRA (Young Jin Chun) and a PhD student at QUB working closely with the GigaMobile Project have worked together with Prof. Michel Yacoub to develop a new analytical model for co-channel interference (CCI) with background noise (BN). This will new model will be used to understand the role of CCI+BN in device-to-device communications.
Collaborator Contribution Prof. Yacoub and one of his PhD students have assisted with the formulation of the new expressions - in particular their proof and simplification. This work has led to a publication in IEEE Communications Letters and to a further study which will help understand the role of relaying in device-to-device communications.
Impact See publications list.
Start Year 2016

Description 2nd NEMOs Workhop on mmWave Communications (July 31st and August 1st, 2018)
Form Of Engagement Activity Participation in an activity, workshop or similar
Part Of Official Scheme? No
Geographic Reach International
Primary Audience Other audiences
Results and Impact This was a workshop which took place on July 31st and August 1st, 2018 on mmWave Communications. It was organised by researchers from the CONNECT Centre and held at Trinity College Dublin. The event brought together key (national and international) stakeholders in the development and adoption of mmWave communications, it attracted in excess of 200 participants from Industry, Academia and the general public. The PI provided a keynote at the event.
Year(s) Of Engagement Activity 2017

Description Future Wireless Mini-Symposium (April 28th, 2016)
Form Of Engagement Activity Participation in an activity, workshop or similar
Part Of Official Scheme? No
Geographic Reach International
Results and Impact This was a mini-symposium which took place on April 28th 2016 on the theme of "High Capacity Networking Opportunities for Future Cellular Networks". It was organised by the PI (Simon Cotton) and held at the ECIT Institute of Queen's University Belfast. The symposium brought together key (national and international) stakeholders in the development and adoption of future high capacity wireless communications.

The symposium attracted in excess of 100 participants from Industry, Academia and the general public. It featured talks from leading industrial and academic speakers from around the globe. Among these, were:

- Mike Short (Vice President of O2 / Telefonica Europe and former President of the IET) Linkedin Profile: http://tinyurl.com/zl2q5sp

- Howard Benn (Head of Standards and Industrial Affairs at the Samsung Electronics Research Institute, UK) Linkedin Profile: http://tinyurl.com/j3esscl

- Prof. Luiz DaSilva (FIEEE, Trinity College Dublin, Ireland) Linkedin Profile: http://tinyurl.com/jm8lm6r

- Prof. Allen MacKenzie (Virginia Tech, USA) Linkedin Profile: http://tinyurl.com/jq26w44

The symposium has facilitated the creation of a number of new collaborative links (between audience members and speakers!), the sharing of data between academia and industry as well as knowledge exchange helping to shape future research challenges. The symposium was so successful, that it has been made an annual event with Trinity College Dublin agreeing to host the next event in July 2017.
Year(s) Of Engagement Activity 2016