Orbit determination/prediction PhD opportunity at University College 
London:

Long Term Prediction of GNSS Spacecraft Orbits for Integration with 
Chipset Location Devices

3.5 year fully funded industrial CASE PhD studentship based at University 
College London, and in collaboration with Glonav Ltd., supervised by Dr 
Marek Ziebart (UCL) and Dr Peter Fleming (Glonav)

All applications of GPS technology rely upon understanding and predicting 
how the system spacecraft move in orbit about the Earth. This in turn 
requires modelling of the forces acting on the spacecraft and modelling of 
measurements made between the satellites and tracking stations on the 
Earth’s surface. With the explosive growth of applications of space-based 
navigation technology there is great demand for individuals with PhDs in 
this area. The project (described in more detail below) would suit 
graduates with backgrounds in physics, space science, mathematics, 
electrical engineering, computer science or similar numerate disciplines. 
The funding covers all tuition fees and an annual (tax free) stipend of 
£12,500 to £13,000.

Project details:

The market growth in integration of GPS positioning capability with 
handheld devices such as mobile phones and PDAs has been staggering. There 
are an estimated 200 million GPS-enabled chipsets currently in use 
worldwide, and this market is predicted to grow to circa one billion in 
the next five years. The uptake of technology in this market sector is 
driven by innovation. Moreover, this innovation is the key differentiator 
between competitors in the market place.
The advent of high sensitivity GPS receivers has greatly expanded the 
scope of the user environment. Whilst discontinuous tracking of satellite-
to-receiver ranges in ‘difficult’ environments (such as within cities) is 
not greatly problematic this lack of continuity, and its resulting 
episodic capture of the GPS navigation message, means that alternative 
mechanisms of accessing GPS orbit data become very important. One powerful 
way of achieving this is to predict the orbits of the GPS satellites over 
long arcs - that is, over much longer arcs than the conventional broadcast 
ephemeris parameters – which are only valid for a few hours. The challenge 
is to predict the orbits over many days with an accuracy of a few metres 
(or less). These long arc orbits would then be uploaded to the device when 
it is docked or synched, or they could be delivered via wireless 
communications. This would give added benefits to improve ‘customer 
satisfaction’: positioning ‘hot start’ would be at the level of 1 second 
and the device capability would not be vulnerable to changes in the 
broadcast ephemeris parameters (these can change each hour).
In this project a method will be developed to enable accurate long term 
GPS/GNSS satellite prediction over weeks/months, using an optimal 
combination of state of the art orbit modelling techniques (developed at 
University College London) with a square root information filter to 
incorporate ranging observations and other sources of ephemeris data. This 
will result in users being able to utilise positioning devices in 
difficult environments without needing continual tracking of the GPS/GNSS 
navigation message. The key aim here is to make these long term orbit 
predictions more accurate by an order of magnitude, and to make them valid 
(within a certain specification) over longer periods of time (that is, 
over weeks and possibly months).
UCL has a global reputation in the area of satellite orbit modelling, and 
is the most active research centre in the world in developing the field of 
GPS orbit prediction. Spacecraft dynamic models developed by UCL are used 
as operational standards by NASA Jet Propulsion Laboratory and by NASA 
Goddard Space Flight Centre. The UCL group has twice been commissioned by 
the US Air Force to develop spacecraft models for GPS satellites. Applying 
this experience in the domain of long term orbit prediction is a natural 
extension of current research capability and should lead to significant 
gains in this area. 
The project industrial partner is Glonav Ltd, a recent start up company 
but built around a UK industry team with over 20 years experience in GPS 
user equipment. Glonav was formed in June 2006 from the GPS division of 
Ceva, and is a “fabless semiconductor” company concentrating on providing 
high sensitivity chips for integration into cellphones and Personal 
Navigation Devices. Glonav has 13 employees in the UK, and less than 50 
worldwide. In the 1980s, the GPS team, based in Daventry, Northants, 
(formerly Navstar), pioneered the delivery of GPS satellite data over 
wireless links, now recognised as Assisted GPS. The core of this team 
remains together today, and brings to the new markets its 10th generation 
of GPS technology development. In 2006 ~ 5 million GPS engines using the 
Glonav core GPS technology were shipped to consumers. The results of the 
CASE studentship will be channelled into the Glonav host software for 
deployment with customers worldwide.

If you are interested in the position please send a CV and covering letter 
(via e-mail or surface mail) to:

Dr Marek Ziebart
Department of Geomatic Engineering
University College London
Gower Street
London
WC1E 6BT

e-mail: [log in to unmask]

Closing date for applications: 20th August, 2007

For a full award students must have:
Relevant connection with UK (usually established by residence) 
Normally a degree at 2:1 or better (or equivalent)

For additional details on EPSRC iCASE studentships see:

http://www.epsrc.ac.uk/PostgraduateTraining/IndustrialCASE/FrequentlyAskedQ
uestions.htm