Astute Submarine

Astute submarine
Earlier today HMS Astute, the first boat in new submarine class for the UK Royal Navy was launched (‘New UK nuclear submarine launched’, BBC News). This is a significant event, as the Astute class submarine is an incredible technological achievement. For example: it can traverse the globe without surfacing because it can generate oxygen from sea water; and the Thales 2076 sonar (Aviation Week) has the world’s largest number of hydrophones of any sonar system in service – it is claimed that under the right conditions an Astute in the English Channel could detect the QE2 cruise ship leaving New York harbour on the other side of the Atlantic Ocean (BAE SYSTEMS press release). You can read more about these amazing facts on the UK Royal Navy website; there’s also a rare video clip of the inside of the Astute on the BBC News website, and there are some nice photos on the Gizmodo blog.

I have to admit that I am bit of a fan of submarines, probably from watching the film ‘The Hunt for Red October‘ (Internet Movie Database, Wikipedia). I remember one of the lines spoken in the film was about making submarines to appear to be "a hole in the water". Hollywood, of course, has a tendency to stretch what’s technically possible at any given point in time, but often it’s not that far from reality, or at least the near term future.

Astute is now the UK Royal Navy’s largest attack submarine at 97 metres long, so it is vital to keep it invisible from sonar detection from surface ships and other submarines. The Royal Navy website mentions that Astute implements passive sonar cancellation through the use of 39,000 acoustic tiles on the submarine’s hull to minimize detection (analogous to the radar-absorbing materials used on stealth aircraft). Unfortunately, the website doesn’t mention whether Astute employs any forms of active sonar cancellation.

One of the ways in which active measures can be taken to cancel out incoming sonar waves which are reflected off the submarine’s hull is by generating waves of identical frequency, amplitude and direction, but 180 degrees out of phase, thus canceling out the reflected wave. It is theoretically easier to achieve active sonar cancellation than active radar cancellation, because the speed of sound in water is around 1500m/s, whereas the speed of light (and radio waves) in air is 300,000,000 m/s (wikipedia). So, if  sensors were installed one metre from the submarine’s hull this would mean that there would be less than a millisecond to sample incoming sonar waves across multiple frequencies and perform the complex computations to determine the cancellation pulses which would need to be generated. (That’s a nice example of a hard real-time requirement – if the deadline is missed, there’s no second chance).

This of course would require a massive amount of computing power, and ten years ago when I was working on large multiprocessor VME DSP systems which typically provided around 720 MFLOPs per VME slot, this didn’t seem to provide sufficient processing density, especially given that the space is a lot more restricted on a submarine than on a surface ship. Ten years later, there are now available multiprocessor DSP and multicore CPU computing blades providing between 10 and 100 times the computing power, and with multi-gigabit per second interconnects, so active sonar cancellation could be  more achievable.

We’ll probably never know, because if this system exists, it will go undetected. So the next time you’re out at sea, you never know, an Astute submarine might be passing silently by in the depths beneath you….