Super-powered GPS?

GPS is useful, but it has a number of problems, and doesn't work well inside buildings. Could a Cambridge startup have the answer?
18 November 2016

Interview with 

Ramsey Faragher, University of Cambridge


Systems like GPS bring enormous advantages, but they also have problems. ForGPS a start, they usually don't work indoors, and in cities like London dominated by high rise buildings the signal can be patchy at best, and usually right before a critical navigational decision is needed! We've probably all been there! The problem stems, in part, from the way that GPS devices handle the signals fed into them. But that might be about to change thanks to a new Cambridge-based start up who have invented technology that - they claim - can make GPS devices 1000 times more sensitive than they are at the moment. Ramsey Faragher spoke to Chris Smith about his solution...

Ramsey - GPS satellites are 20,000 kilometres away orbiting the Earth. They transmit a very weak signal, it's only 50 watts, like a light bulb but it travels all that distance. And when it reaches the surface of the Earth, it's lost so much energy that we only pick up signals that are a billionth of a billionth of a watt, so it's incredibly weak.

The sensitivity of the receivers we have today is such that they can cope with a small amount of material in the way, like the leaves of trees or maybe a very thin wooden wall. But the challenges we have in cities is that the signal can get stopped when it hits a building. You're on the other side of that building, but that signal from the satellite might reflect off a large piece of metal, or another building, or a puddle on the ground and scatter back towards you from a different direction. And that delays the timing signal that's being carried by that satellite and current receivers don't do anything overly clever to unpick and distinguish between these and filter out the reflections.

Chris - So instead of being able to see a series of independent signals coming in, all time stamped, and you're measuring the time differences, if one of them has bounced off, like a bullet ricocheting around a room, a whole bunch of buildings before it gets to the device, it's going to feed the receiving device incorrect time signals so it's going to think you're not where you are. And so, as a result it knows it's got something wrong and it just says I can't detect a valid signal?

Ramsey - Yes, exactly. And critically, the maths that's going on inside the GPS receivers always assume that you have a line of sight signal. The receivers try to do some things to tidy up that mushed, delayed timing measurement, but they still put it through the same line of maths as if it was a line of sight signal, even if it's not, and this is where we get really big errors in cities.

Chris - And also why, I mean we're inside here in a building and I just turned by phone on to see if I could localise myself and it says it can't find a GPS signal. that would be why presumably?

Ramsey - Yes. So we're only actually a few metres from a window, but the other side of that window is an urban canyon, we're in the middle of Glasgow city centre. And that receiver is not sensitive enough to pick them up through the walls and the ceiling of this building so, no GPS for you.

Chris - It's a major headache, isn't it? Because if you're a person trying to find their way around a town using your phone, you often get lost when actually you're trying to find where you're going, emergency vehicles might be held up trying to find their way to a location. So what can we do about it?

Ramsey - Well, firstly, much more seriously and concerning, if you make an emergency call on a mobile device and it tries to turn on the GPS chip in order to provide the emergency services with your location, then you're at the mercy of the environment you're in. So if you are outside in the open, then a GPS fix will occur and the emergency services can see where you are.

Chris - To what sort of accuracy?

Ramsey - Well, if you're outside in the open - ten metres. But people who are inside with no GPS fix, they don't have that luxury. And we know, at the moment, hundreds of thousands of people die worldwide every year on emergency calls speaking to the 999 operator, and there's no way of knowing where they are.

Chris - Yeah. So you've made the problem very clear to me, but have you got a solution?

Ramsey - Well, one way to try and solve this problem is to dramatically improve the importance of the GPS receivers and, at Focal Point Positioning, we've developed a new way of processing the GPS data so that its 1,000 times more sensitive than existing receiver designs. And we clean up the multipath interference, that is the reflected signal versus the line of signals.

So we get a boost in sensitivity that's beyond other people's way of detecting signals inside buildings and we clean up those signals so that the accuracy is high inside the building as well as high outside the building.

Chris - I know you don't want to blow your own patent so I'll ask you to word the answer cautiously. But are you doing it by being able to work out what's a real line of sight signal coming in from the satellite versus what must have bounced around like a bullet ricocheting off the nearby buildings and because you can tell them apart you can clean up the signal - is that how you do it?

Ramsey - There's a number of clever things we've done. We've started again, basically with a blank piece of paper and said here's where the electromagnetic radiation is hitting the antenna and lets work from here. And we're going to ask the question every stage along the way: is this the state of the art or can we do better? There's a number of stages that we've improved but, critically, what we've done is just change the software that runs on the chip.

Chris - So a person who uses this, ultimately when it becomes available, they would need to install this into their device, into the chip in their phone, for example, to make it work?

Ramsey - They would not. We are aiming to licence this technology to chips at manufacturers so that they change the software on their chips. And we're talking to people in both the smartphone industry and the automotive sector. Our technology dramatically improves GPS positioning around buildings, in cities as well, which is very important. Not just for people who want their Uber driver to turn up at the right place, but for the coming autonomous vehicles that will be safely driving us home while we sleep in the back.


Add a comment