How Connect is finding new ways for data to speak to society


13 Apr 2021390 Views

Prof Dan Kilper. Image: Connect

Sam Cox discovers the future of networks of all shapes and sizes being investigated at the Connect research centre in Trinity College Dublin.

The Covid-19 pandemic has brought the critical significance of internet connectivity to the fore. With millions stuck at home, communications networks have become appreciated as an essential utility. Connectivity to the internet has allowed continued communication, work and entertainment amid broad restrictions on movement.

These communication networks are the focus of the research carried out by Connect, the Science Foundation Ireland (SFI) research centre for future networks and communications. Headquartered in Trinity College Dublin and with researchers across nine other higher education institutes, Connect’s focus on reliable, sustainable and dependable networks in all of their forms has contributed to its recent success in attracting significant renewed public investment for the next six years.

‘The thing about communication networks is they evolve really quickly. We are in a very fast-paced area of technology’
– DR BRENDAN JENNINGS

Prof Dan Kilper has just taken the reins as Connect’s new director. Coming from the University of Arizona and having collaborated with Connect via SFI’s US-Ireland research partnership programme, he was already very familiar with the centre’s work.

“One of the really attractive elements of Connect’s research programme is its holistic approach to networks. The team takes an end-to-end view with expertise ranging from sensor design to wireless networks to optical communications. This ensures strong multidisciplinary approaches to the research challenges we address,” said Kilper.

“There is also a strong emphasis on testbeds, which facilitate really exciting academic and industry collaboration. Our Pervasive Nation [internet of things] testbed and our Open Ireland open networking testbed in Trinity are enabling new international partnerships to address a variety of research challenges.”

Connect also works closely with industry partners. According to Kilper, some 44 companies invested more than €14m over the past six years. “This provides a strong applied focus for the research,” he said.

“There is also a strong focus on commercialisation and five spin-out companies, which now employ 41 people with a further 14 people to be hired in 2021, have emerged over the past few years.”

Kilper also noted that Connect researchers have won more than €25m in funding from non-exchequer, non-commercial sources such as the European Union’s Horizon 2020 fund. This, said Kilper, is an indication of the centre’s “world-leading quality”.

Changing connections

Dr Brendan Jennings from Waterford Institute of Technology (WIT) is one of Connect’s principal investigators. Looking ahead to the centre’s next few years of research, he said: “The thing about communication networks is they evolve really quickly. We are in a very fast-paced area of technology. It is very hard to identify what’s going to be interesting in communications networks research in four years’ time.”

Jennings highlighted satellite networking and quantum communications as two areas that will likely become a focus in the coming years, but he feels that surprises are always around the corner. Take, for instance, the Covid-19 tracker app developed by the HSE.

“It uses Bluetooth capability on the phone. If you asked me last year if there was anything especially interesting about Bluetooth technology, I would have said no. It is really mature, it works really well. But, suddenly, last March, people started using it for a purpose other than it was designed for – to estimate the physical distance between pairs of phones. This was repurposing a technology,” said Jennings. “So, the question for us was: how well will it work?”

Some of the researchers at Trinity involved with Connect undertook measurement-based studies to gauge how accurate Ireland’s Covid-19 tracking app would be in real-world environments. And this wasn’t the only way the pandemic has impacted their research into future networks.

“If the pandemic had happened 10 years ago, the economic hit would’ve been worse because many of us would not have been able to work. We would not have had the capacity or the tools. It has shown how critical it is to have these video-conferencing tools,” said Jennings.

In terms of Connect’s upcoming six-year research cycle, Jennings (who served as interim director for the centre) identified three key pillars. First is flexibility, allowing people to use networks in different ways. Second is resilience: can the network perform to a sufficient degree when damaged, or indeed self-repair? And third is sustainability. As in, if we use millions of interconnected devices, how do we design these objects to be sustainable.

Jennings noted that the immediate future may be very different, given our current experience of how communications networks can, as a result of design choices, change how we interact with our colleagues, our workspaces and our environments. This is what Connect is doing – reimagining these networks, from the nanoscale to the whole of society.

Molecular communications

So, let’s start small – really small – with Dr Sasitharan Balasubramaniam’s research at the Walton Institute in WIT. He is reimagining the body at a molecular level through a communication networks framework.

Within the gut, there are billions of bacteria communicating with their environment and with each other. It is their presence that keeps our gut stable. They’re digesting food and they’re influencing molecules that feed back to the brain. It’s a symbiotic relationship that is necessary for life.

But what happens when there is a malfunction or change in the dynamics? Because of infections or other causes, this communication network can become destabilised. Balasubramaniam suggested this might be why certain people are allergic to particular foods.

If this communication system could be understood, Balasubramaniam believes we could engineer new populations of bacteria that could compensate for the deficiency. Cell DNA could be ‘programmed’ to change its function and influence certain, specific molecules, creating an alternative way to transmit information within the human body. Once these bacteria start communicating within the network, the instability could be corrected, and the network would be set right.

‘They’re very complex systems compared to the internet’
– DR SASITHARAN BALASUBRAMANIAM

While much of this work is theoretical, there are immediate applications. “We submitted a paper where we simulated the lung network. It’s basically just a network of pipes. It takes air in and it feeds into the lung tissue. We can simulate the quantity of virus you inhale and how far it will go down – and we applied this to Covid-19,” said Balasubramaniam.

This project investigated different rates of breathing and the condition of your throat, and how these factors interplay with viral-to-cell penetration and replication. Of particular interest in this researcher is to discover smokers’ vulnerability to Covid-19. Dr Gerald Barry from University College Dublin provided the necessary expertise in virology, feeding the necessary details into the researchers’ network frameworks. Meanwhile, the Connect team are on a learning curve taking them from cellular networks to ‘cell networks’.

“I’m learning as we’re going,” said Balasubramaniam. “They’re very complex systems compared to the internet. They are very stochastic – they die, they replicate, they can mutate their genes and change their behaviour properties. You don’t really find that in internet systems. In your mobile phone, the worst you have is a hardware system failure or you download malware that disrupts your security function. But you don’t have mobile devices that mutate in the same way. You have so many subsystem levels of cells, to organs, to systems, and there are so many levels of different properties.”

Enabling smart cities

From complex internal systems to global ones, Prof Siobhán Clarke told me about Earth Overshoot Day. “An international group assesses at what point in the year that the world has used up the resources the world can provide that year, and is borrowing from future [generations],” she explained.

“That date has been getting earlier and earlier, and in 2019 it was 29 July. Even with the lockdowns, and fewer people flying, Earth Overshoot Day happened on 22 August in 2020.”

This frightens Clarke and she emphasised the need to better use our global resources. She is head of Enable, a large research programme hosted by Connect through the SFI Spokes programme – a mechanism introduced to encourage expansion of additional academic and research partners as a centre moves through its research cycle.

Enable is focused on connecting communities with smart urban environments through the internet of things. Clarke explained that they want to enable cities to better use what they have available to them. “A smart city can gather information about what’s happening in the city in real-time, be able to analyse the information in real-time, and make good decisions in real-time, to make the city work better,” she said.

A blonde short-haired woman in a twill blazer stands smiling with her arms folded.

Prof Siobhán Clarke. Image: Connect

While this may sound like a utopia, there is a pressing need to decrease energy consumption and to make smarter decisions around managing infrastructure, according to Clarke. On a practical level, this involves embedding sensors within the environment to feed back and allow for ongoing analysis.

Take, for instance, bins. Traditional bin systems have trucks collecting waste at a set time on a set route. If the bin is too empty at this particular collection time, fuel is wasted and there is also additional traffic congestion. If the bin is too full by the time the truck arrives, there is litter and people are angry. A smart bin would continuously send information on how full it is, enabling bin collections and routes to adapt in real-time.

That is what can be done with one data point. But what if the bin also registers how often it is being opened? Now you can anticipate when the bin will need to be collected. More data points mean better prediction.

Connect’s role in this project is to feed the information back for analysis and to look at it in real time. “Getting a city to work well requires really strong decision-making in real time. The better the information, the better decisions you can make,” Clarke explained. “If, for example, you get data related to traveller demand for decision-making to change the routes of buses, you might make that change. But you also have to know how to loop back the data on whether that was a positive or negative change.”

There are ethical considerations with this that Clarke acknowledged. It is important that all socioeconomic areas are treated fairly by the algorithms. Citizen engagement is a large part of Enable’s portfolio, both in terms of data collection and finding what works for its users. There are big questions, however, relating to how our environment could and should monitor our actions. While there is an obvious need for better resource management, how much power should technology have to shape our lives?

Artists and ethics

One aspect of Connect that sets it apart is a group of artists that consider these questions. Dr Tom O’Dea and Dennis McNulty are part of Connect’s Orthogonal Methods Group and are committed to exploring some of the centre’s multidisciplinary questions. The group was founded by former Connect director and newly elected provost of Trinity College Dublin, Prof Linda Doyle.

McNulty trained as a civil engineer while O’Dea trained in mechanical engineering. O’Dea highlighted how a fundamental principle of engineering is abstracting the environment in such a way as to simplify and process it. By doing this, engineering tools and equations can be used to manipulate the world mathematically. In traditional engineering, it is easier to remove humans from the context of design.

McNulty agrees. “If you study engineering, on a day-to-day basis, you study certain equations related to a specific field but, over time, what you’re really learning is a way of thinking, a very particular way of processing the world,” he said.

“A few years out of college, I began to understand the influence my engineering education had had on my perspective. And later on, after a decade or so working as an artist, I realised I needed to reconcile the engineering and artist parts of myself.”

Managing this complexity is part of McNulty and O’Dea’s role within Connect. As part of the Designing for the Unknown research group, they met with two other artists and four engineers every fortnight for workshops and research discussions. They found their discussions regularly orbited around ethics, with one of the central questions being: who are the people affected by your research? This led to the production of an ethics toolkit for researchers which has now been made available on YouTube.

As the world settles into the new normal, Connect will be working to improve the resilience, flexibility and sustainability of networks whether the node is your neuron, your laptop or your conference call. Isolated, these data points are powerful but, once they are linked together, their properties can take on new dimensions addressing everything from social isolation to climate change. From the minute to the massive.

By Sam Cox

Sam Cox was named the science and technology winner in the 2020 National Student Media Awards (Smedias). This award category is sponsored by Science Foundation Ireland and includes a €1,000 bursary to support and encourage up-and-coming science and technology journalism.

The 2021 Smedias are now open for entries. The deadline for applications is 15 April 2021.