Understanding the Global Navigation Satellite System

| November 22, 2017

Have you ever wondered why your GPS receiver acts up when driving around the city? It may have decided that you have instantaneously teleported from one street to another or that you’re driving through the hotel lobby instead of passing by. The answer is multipath.

GPS satellites broadcast a type of signal which enables a receiver to measure the distance between itself and the satellite. With enough of these measurements from different GPS satellites, the receiver can then calculate it’s position. Multipath is a phenomenon that occurs when the signal from the GPS satellite bounces off an object, like a building in the city, before arriving at the receiver along with the signal that has propagated directly to the receiver. Because the signal that has reflected off the building takes longer to arrive at the receiver, the measured distance between the GPS satellite and receiver now contains errors resulting in the calculated position being incorrect.

For years engineers have investigated ways of eliminating multipath and the errors it introduces. Over these years we have also learnt that multipath isn’t always a bad thing. If we design a receiver properly, we can actually use the reflected signal to infer properties about what it reflected off.

If you look at how sunlight reflects off the surface of the ocean you will notice how smooth waters result in a mirror like reflection while rough waters will cause a larger patch of the ocean to glisten. A similar process occurs to the GPS signals reflected off the ocean surface and we can measure how smooth or rough the water is from the reflected signal. Using models we can also estimate the speed of the wind at the surface of the ocean. It is also possible to estimate tidal and wave heights of the ocean along with the moisture level of soil, depth of snow, age of ice and the thickness of vegetation using the reflected signal. All of this from a phenomena that is considered by most to be a nuisance!

The Global Positioning System (GPS) is the U.S satellite navigation system and is the most widely known system. Global Navigation Satellite System (GNSS) is an umbrella term encompassing all global systems including GPS along with the Russian GLONASS, European Galileo and Chinese Beidou systems. Using reflected GNSS signals for remote sensing is known as GNSS Reflectometry (GNSS-R) and is a type of bi-static radar because the transmitter (the GNSS satellite) and receiver are not at the same location.

A major advantage of this configuration is that when we build a GNSS-R receiver we do not need to include a transmitter and carry that along with us as most traditional remote sensing systems do. This drastically reduces the mass, volume, power and overall cost when compared to traditional systems remote sensing systems.

There is a GNSS hotspot in Australia’s corner of the globe which enjoys the highest number of visible GNSS and augmentation system satellites. Accompanying this is the highest number of GNSS reflections and observations of the Earth made possible. Combined with the low cost of GNSS-R receivers compared to traditional remote sensing systems, we must consider the use of GNSS-R for remote sensing in Australia as part of developing our space capabilities. To date, we’ve depended on foreignly owned satellites for Earth observation data and have either paid for it or relied on the goodwill of some institutions who make it publicly available.

Ben Southwell

Benjamin J. Southwell received his Bachelor of Robotic and Mechatronic Engineering from the University of Western Sydney (Honours Class I, University Medal) in 2012. He is currently a PhD. candidate at the Australian Centre for Space Engineering Research (ACSER), The University of New South Wales, AUSTRALIA. His current research interests include remote sensing with reflected Global Navigation Satellite System (GNSS) signals and other signals of opportunity using airborne and spaceborne receivers. He is also a member of ACSER’s CubeSat team and has been responsible for both development and operations of UNSW-EC0 which is currently flying in low Earth orbit as part of the QB50 mission.


  1. Alan Douglas

    November 28, 2017 at 2:34 pm

    I found this article fascinating and would like to know more. Could you please explain why, if we are in a hot spot, do we also need to get into space ourselves? Surely, if these satellites are all transmitting we should be able to use the data by comparing the signals from different devices? Or have I misunderstood what you wrote?

    • editor


      November 29, 2017 at 8:45 am

      From the author:
      In general, we need to get into space ourselves so that we own the infrastructure and are not reliant on others to either provide us with access to the data or control of their satellites through goodwill or contractual obligations. We need to develop satellites that are designed specifically to serve Australia’s needs. Additionally, the space industry is worth hundreds of billions of dollars and is growing, why would any country sit this one out?

      The acronyms can be a little confusing. GNSS satellites are navigation satellites, other types of satellites include those used for communications and remote sensing. GNSS-R is the use of the reflected GNSS signal for remote sensing. Australia sits in a GNSS hot spot which makes it an ideal place to utilise GNSS-R technology for remote sensing.

      I am not suggesting we launch our own navigation system (however the argument for a space based augmentation system (SBAS) for Australia is one to be had another day).

      I am suggesting Australia consider using GNSS-R technology on a satellite(s) designed for remote sensing applications in Australia. GNSS-R technology relies on the existing GNSS satellites to be the transmitter part of the sensing system greatly reducing the cost and complexity. We do not own any space based remote sensing satellites and the low cost of GNSS-R receivers compared to traditional remote sensing technology makes it a very attractive option for our first remote sensing satellite.

      I hope this answers your question.

  2. Alan Douglas

    November 29, 2017 at 8:58 am

    Yes. Thank you. You seem to have an unusually broad knowledge; I appreciate it.