Exploiting the Location information for Adaptive Beamforming in Transport Systems

The transport sector is a major driver of economies across the globe. The digital augmentation of the sector has helped created new businesses worldwide. Further growth of the so-called intelligent transport systems is contingent on the synergy with the wireless telecommunication networks that provide the resources for fast, reliable and robust connectivity. This synergy is key for the new  EIT Digital industrial doctorate position at Telefonica UK Limited and the University of Glasgow.

The transport sector market is estimated at £5 trillion. New businesses created via digital augmentation of the sector currently contribute approximately £275 billion to economies worldwide.

Millimetre-wave (mm-wave) communication, a wireless technology, can support ultra-fast and high data-rates that are significantly higher than the conventional microwave frequency mobile networks. However, there are several physical and engineering challenges that currently limit a widescale deployment of mm-wave networks. One of them is the moving vehicle, for which several precursory measures need to be taken to establish a reliable communication link. These pre-configuration overheads not only make the overall communication system inefficient in terms of time but energy expenditure as well.

To overcome these propagation constraints, new hardware systems and architecture frameworks such as large-scale antenna array systems must be used, which are empowered with a computationally-intensive process of adaptive beamforming. However, at mm-wave frequencies the cost of beamforming becomes prohibitively high. Therefore, a major challenge in the realization of mm-wave communications is to reduce the beam-training overhead.

In the public transport sector such as buses, trams and trains, all the vehicles operate at fixed routes and schedules, and therefore, through this prior, contextual information, the location of vehicles can be predicted with high accuracy at any given time. It has been shown that this knowledge about the transport infrastructure can be used to reduce the time and space complexity of the beam-training overhead required for establishing communication links.

Approach

The overarching aim and innovation of this industrial doctorate position are to leverage the prior, on-hand location information of vehicles, learned in part from the routes and schedules, and develop a self-organizing mobile network architecture that optimizes the radio link-budget. The inherent challenges of the mm-wave communication network will be addressed through a resilient beamforming strategy that can be reconfigured based on the geospatial information from the vehicles which is partly inferred from the transport infrastructure.

The industrial doctorate candidate will develop novel, machine learning-based, real-time direction estimation algorithms through which the antenna beam patterns can be efficiently coordinated and aligned in the required directions to provide maximal signal strengths at the receiver end of the communication system. The PhD will also use technologies such as machine learning to develop predictive models from the contextual information of the location.

The novelty of the PhD lies in exploiting the location information of moving vehicles to reconfigure the mobile network as needed. The proposed project completely aligns with the research directions of the industrial partner, Telefónica UK Limited, which is actively working with public transport infrastructure entities in the UK such as TfL and Network Rail to develop 5G-enabled ITS.

Expected outcome

The industrial doctorate candidate will publish two journal articles, two conference papers and a PhD Thesis. On the business side,  the outcomes of the project will primarily consist of a self-organising network (SON) scheme for an intelligent transport system (ITS) that leverages the location information of vehicles in configuring the network parameters, which is realised through adaptive beamforming of the radio links. The new technologies developed in the project will result in patent(s) applied through the support of the business and innovation office at the University of Glasgow, and the industrial partner, Telefónica UK Limited.

To further test the concept in a physical environment, we will seek the transfer of knowledge to create a spin-off for which we will seek support from the Telefónica’s open innovation hub, Wayra UK in the Intelligent Mobility Accelerator program, that only provides Telefónica’s network resources but business development support as well.

Impact

Data-rich, mobile user analytics solutions have already proven to be a huge success in gauging customer behaviour and marketing trends. However, such solutions often involve huge risks as decisions are primarily taken from the data collected in the past. The current global pandemic event has disrupted every aspect of our lives today. There is now an ever-greater need to develop analytics-based solutions through which decisions can be made based on real-time events. The work done in this project directly falls under the business and research directions of the industrial partner, Telefónica UK Limited, especially, the O2 motion lab which works in exploiting the mobile user location data to identify customer behaviour and generate marketing insights. This phD project will enable real-time decisions for network reconfiguration,  and expand Telefónica’s portfolio in the ITS.

Location

The doctoral student involved in this programme will share its time between the University of Glasgow, the nearby Co-Location Centre of the EIT Digital Node, the local government body Transport for London, and Network Rail.  In addition to the industrial doctorate research, the PhD student will also be following leadership seminars.

Facts

Apply

You can read the full description here. Those interested in applying should complete the EIT Digital application form. You are asked for your motivation, CV, and documents showing their academic track records.

© 2010-2021 EIT Digital IVZW. All rights reserved. Legal notice