Clearer sight for submarines

Submarines could observe the ocean floor with 10 times better resolution and detect underwater mines with improved accuracy thanks to a new form of sonar imaging technology.

That is the hope of the UK's Defence Science and Technology Laboratory (DSTL), which is supporting researchers at Heriot-Watt University in their work to improve sonar target detection in cluttered underwater environments.

The academic team will develop algorithms and 3D image processing techniques for interferometric synthetic aperture sonar (SAS). These sonar combine a number of 'acoustic pings' to form images that are said to have vastly higher resolution than conventional sonar.

'You can get pictures that are approaching optical quality by using that sort of technique,' said Keith Brown, one of the Heriot-Watt University researchers. SAS, he added, are used almost exclusively for research purposes and have yet to be deployed in military platforms.

The Heriot-Watt researchers will combine the SAS with more conventional side-scan sonar to determine if their joint deployment can improve the detection and classification of targets under water.

Side-scan sonar works by transmitting a narrow fan-shaped acoustic ping perpendicular to its direction of travel. If the sonar waves hit an object in the surrounding environment, sound energy will reflect back in the direction of the sonar.

The time it takes for the pulse to return is recorded, together with its amplitude, and is sent to a console for interpretation and display. The data from the successive pulses is then stitched together to create a continuous image of the seafloor.

According to Brown, current sonar detection technology suffers from false alarms. 'There are still problems with actually classifying something as a mine, for example, as opposed to a rock,' he said.

The Heriot-Watt team believes that such false alarms could be avoided by imaging the objects in 3D so that characteristics such as texture can be better analysed. Brown said that his team will review several approaches to gathering and reconstructing 3D images using SAS.

At the same time, he added, the team may also consider applying some of its previous research into bio-inspired signal processing. Brown said his group has demonstrated an underwater object detection techniques that mimic dolphins.

'So rather than doing the classification from the image, which is what you do with synthetic aperture sonar and 3D reconstruction, this approach would use just a simple side scan to identify objects of interest,' he explained. 'It would use a signal like a dolphin uses to get more information.'

Brown said this signal processing technology could also have applications beyond the military realm. He believes that it could be used by the oil-and-gas industry to monitor subsea installations.

'We've had some very good preliminary results in detecting the difference between plastic pipes and metal pipes, which look exactly the same with traditional sonar,' said Brown. 'The dolphin signals allow us to see them differently.'

He added that the group also hopes its research into SAS and 3D reconstruction could be used by the oil-and-gas industry or even aviation authorities investigating aircraft crashes at sea.

'This project has been funded partly by the DSTL so they're particularly interested in the detection of mines, but there are many non-military applications where people are interested in finding things on the sea floor,' said Brown.

Siobhan Wagner