Good looker

Radar technology is constantly evolving, but systems still struggle to identify exactly what a target is, spot small targets or see slow-moving objects against a shifting background.

Thales UK and University College London are collaborating in a four-year research project that aims to better understand the key issues facing radar systems. This will be used to develop new processing algorithms and inform future hardware designs.

Barry Trimmer, research technology and engineering director of Thales UK's aerospace business, said the programme aims to improve the ability of a radar-like sensor to provide vital information to military users. Potential applications include deployment on tactical UAVs operating over land and littoral maritime environments or ground-based and airborne battlefield radars.

'You could summarise this as "what is it?" and "where is it?"' he said. 'Typically radars have been quite good at where is it and not terribly good at what is it.'

Prof Chris Baker, head of radar system research at UCL, whose chair in intelligent radar systems is co-sponsored by Thales and the Royal Academy of Engineering, will lead the research.

Baker said the project would be concerned with high-resolution imaging and detection of moving targets from the ground from a single radar system. 'More specifically, it will examine novel techniques able to use single waveforms which, when combined with adaptive signal processing, can achieve both tasks simultaneously,' he said.

One method to be explored is synthetic aperture radar (SAR). 'This is a method of producing something that is more like a photograph,' said Trimmer. 'Radars use exactly the same spectrum as light, but its longer wavelength means it is inherently more blurred. SAR uses a virtual one-kilometre radar antenna, synthesised from views you take at different times as you fly across it. To get better resolution, you take a lot of looks at an object from many different directions and add them all up.'

Another technique, Ground Moving Target Indication (GMTI) helps radars detect an object against a background better than has previously been possible. 'Most of GMTI is about seeing something that's moving against a still background, which by itself is something that's been around for a while,' said Trimmer. 'The development is being able to see slower and slower moving things against a background which itself moves a bit.'

The difficulty is due to a relatively poor understanding of electro-magnetic back scattering in complex scenarios, said Baker. 'All natural scenarios tend to be complex, such as a background of trees, hedges, or buildings. We would like to achieve better performance than is possible by a human. Then the all-weather, 24-hour operation that radar is capable of will become a huge advantage.'

Sea also makes detection difficult. 'Slow-moving targets on the surface are difficult to detect when they move at a similar velocity to the apparent velocity of the sea surface and have a similar-sized echo strength,' said Baker.

Littoral environments have extra problems. 'If you're close to the coast, you've got the issues of local current, buoys and coastline features. You've got to discount all that and see the thing you want to see in the middle of all this,' said Trimmer.

The project will use Thales' database of pre-recorded real-world measurements to enable comparison with initial computer simulations, help refine models and better plan future experiments.

Trimmer cautions it can be difficult to divorce processing from hardware. 'If detection has to happen in real time, you're at the border of hardware and software,' he said. 'We develop a lot of stuff ourselves which sits at the boundary between the analogue world and the digital world. We can't afford the time to wait for a general-purpose processor to do its work.'

According to Trimmer, the difference between the military and the civilian market for radar is becoming more blurred. 'We find the same techniques increasingly used in things like drug interdiction.'

Baker said military radar has tended to lead the way by developing higher-resolution imaging systems for more detail and providing greater sensitivity in the detection of moving targets. These capabilities then tend to appear at a later date in civil systems, which are typically space-based, to provide a global monitoring capability.

'Remote sensing of the earth is becoming more important in providing quantified measurements for climate and earth resources studies,' he said. 'Applications include monitoring of ocean currents, sea height, polar ice-cap melting, rain forest destruction oil spillages, terrain and relief mapping, subsidence detection and traffic monitoring.'

He added that the ability to reliably extract greater levels of detailed information on an automatic basis is absolutely key to getting the most out of these systems, be they military or civil.