Cheaper, simpler THz imaging the promise of new technique

The technique could also have implications for the design of new, high resolution radar and sonar systems. In a digital camera, the lens focuses the incoming light so that light reflected by a small patch of the visual scene strikes a correspondingly small patch of the sensor array. In lower frequency imaging systems, however, an incoming wave strikes all of the sensors in the array. The system determines the origin and intensity of the wave by comparing its phase when it arrives at each of the sensors. As long as the distance between sensors is no more than half the wavelength of the incoming wave, that calculation is fairly straightforward. But if the sensors are spaced farther than half a wavelength apart, the inversion yields more than one possible solution. Those solutions are spaced at regular angles around the sensor array, a phenomenon known as spatial aliasing. In most applications of lower-frequency imaging, however, any given circumference around the detector is usually sparsely populated – which is the phenomenon the new system exploits. "Think about a range around you, around 5ft," said engineering professor Gregory Wornell. "There's actually not that much at five feet around you. Or at 10 feet. Different parts of the scene are occupied at those different ranges, but at any given range, it's pretty sparse. Roughly speaking, the theory goes like this: If, say, 10% of the scene at a given range is occupied with objects, then you need only 10% of the full array to still be able to achieve full resolution." The trick is to determine which 10 percent of the array to keep. Wornell says. "Keeping every tenth sensor won't work: it's the regularity of the distances between sensors that leads to aliasing. Arbitrarily varying the distances between sensors would solve that problem, but it would also make inverting the sensors' measurements prohibitively complicated." As such, Wornell and his team developed a detector along which the sensors are distributed in pairs. The regular spacing between pairs of sensors ensures that the scene reconstruction can be calculated efficiently, but the distance from each sensor to the next remains irregular. The researchers also developed an algorithm that determines the optimal pattern for the sensors' distribution. In essence, the algorithm maximises the number of different distances between arbitrary pairs of sensors. The researchers have performed experiments at radar frequencies using a one-dimensional array of sensors deployed in a car park, which verified the predictions of the theory. With a two dimensional array, the researchers say savings of 100x and 10x can be obtained in each dimension.