When a New York Times journalist asked Albert Einstein to explain radio communication, he reportedly replied: ‘You see, wire telegraph is a kind of a very, very long cat. You pull his tail in New York and his head is meowing in Los Angeles. Do you understand this? And radio operates exactly the same way: you send signals here, they receive them there. The only difference is that there is no cat.’
That was in the 1920s. Even Einstein may have struggled to imagine that less than a century later the ‘magic’ of radio communication would have developed so fantastically that the world would be on the cusp of being powered by the Internet of things (IoT).
Professor Costas Constantinou is an internationally leading expert on communications engineering, specifically radiowave (RW) propagation – the behaviour of radio waves as they travel, or are propagated, from one point to another or to permeate the world around us – whose groundbreaking work will help to make the IoT a reality, most likely within the next few years. He recently presented his Inaugural Lecture, ‘The Magic of Radio’, and opened it with this lesser-known Einstein quote.
‘What I did in my lecture was to weave into a story the history of radio communication, which if it isn’t magic, I don’t know what is,’ observes Costas, Professor of Communication Electrodynamics in the School of Engineering.
Costas’s address was part of a series of Inaugural Lectures run by the College of Engineering and Physical Sciences to showcase its leading scientists who are pushing the boundaries in their disciplines. These events, which are open to the public and free of charge, mark an academic’s promotion to Professor, but also provide a rare opportunity to hear first-hand about their research.
Before detailing his own work, Costas spoke about radio wave communication’s early pioneers, from 19th century Scottish mathematical physicist James Maxwell – ‘almost all our radio services, from television to wifi, are still based on Maxwell’s theoretical analysis’ – to the likes of Braun and Marconi, whose development of wireless telegraphy won them the Nobel Prize in Physics in 1909.
It was another ‘great’, German theoretical physicist Arnold Sommerfeld, who ignited Costas’s interest in RW engineering. ‘Sommerfeld was a genius, and my first contact with the science of RWs was when I read his book. It was he who solved the problem of trapping RWs along the earth’s surface. What really impressed me as an undergraduate was that his work demonstrated that it wasn’t enough to have a theory – you needed insights to figure out how to make it work.’
And that is what Costas, who has more than 150 research papers to his name, has gone on to do. In a career that already spans 30 years, he has made significant contributions in a broad range of topics that include electromagnetic scattering and diffraction, electromagnetic measurement, RW propagation modelling, mobile radio and future communications networks architectures. He has also contributed to two patent applications on highly resilient routing protocols in computer networks and a patent application on electrically small antennas.
One of Costas’s early successes, in the ‘90s, was the creation of a RW propagation prediction model, which to this day is the engine inside the Ministry of Defence’s battlefield radio communication planning tool.
‘This was one of the very first three-dimensional RW prediction models capable of being used in both flat terrain as well as mountainous terrain – waves propagating over hills, inside valleys and so on,’ he explains.
Costas has also carried out a systematic study of what gives rise to all the radio echoes around buildings and whether it is possible to predict accurately the physics behind the real-world echoes. This involved tackling the problem of how RWs bend around real building corners.
That work led to the forensic analysis of radio imaging around Bristol city centre and the Birmingham campus, with help from Oxford and Bristol university researchers.
‘The best way to give a flavour of what we did is to describe it as taking fuzzy video pictures of RWs in flight as they bounced towards the receiver and carefully locating precisely the object which caused each bounce, and whether we had the theoretical tools to predict the signal strength of each echo.’
Seven years into the study, Costas and his collaborators succeeded in generating the knowledge necessary for radio planning tools to make accurate predictions, feeding information to BT that made its way into the 3G mobile radio standardisation process.
‘Along the way we found that the building architecture and materials, trees, lamp posts and even metallic gutter downpipes all matter, and that we have the theoretical tools to incorporate most, but unfortunately not all, of the important sources of radio echoes into propagation prediction models.’
The problem is that the environment in which we deploy radio systems is often too complex to model. ‘Our use of statistics is a pragmatic approach that allows us to deal with our ignorance of the environment and builds a safety margin in the signal strength calculations. The fact that radio systems work at all with the levels of uncertainty that exist in their planning is nothing short of a miracle – magic, in fact!’
Costas’s work now is geared towards providing richer-than-ever-before information to the designers of radio networks and future devices.
‘This is increasingly going hand-in-hand with information on how to modify the environment, to make buildings friendlier electromagnetically. It will in future involve educating architects on wave propagation, in order to contain and minimise interference. The added challenge is that by migrating to higher frequencies, the wavelength shrinks and we have to describe the world at very high resolution.’
Above all, says Costas, the biggest challenge comes from the sheer scale of what is expected of radio communication. A recent report by Intel concluded that by 2020 there will be 212 billion connected devices in the world driving cars, monitoring the environment and health, controlling food production, medicines, and generally improving people’s lives. Nearly three-quarters of these will be connected wirelessly.
‘So to understand and successfully engineer the ground, it’s essential to overcome its problematic behaviour,’ says Ian, whose Inaugural Lecture was entitled ‘Too loose or not too loose: making problematic ground safe’. ‘To do this safely, effectively and economically requires an understanding of how the ground was formed, how this impacts behaviour and affects any interventions designed to overcome such behaviour.’
While acknowledging that technology has its limitations, Costas sees the need to introduce more automation and intelligence.
‘Ultimately, by making the environment more intelligent, we can pollute less: we can stop using energy when we don’t need it; we can make things safer – from cars to houses. By populating the world with all these sensors and making them secure, we can go to the next level of mankind’s manipulation of its environment – to mankind’s benefit but without leaving behind a footprint of devastation. What I do is a tiny thing in this bigger picture, but lots of tiny things are needed.’