It won't. That's just roughly equivalent to a towed array with a longer set of hydrophones than normal. Bearing accuracy might go up if they are integrated correctly, but there will still be two solutions for the same time-differences.

As for offset, we are talking maybe 10 lousy meters (width of sub), and unlike the fixed bow arrays, the relative positions of the two are not precisely defined (they float around in the water).

I found this interesting little tidbit on a site about the Navy's ARCI sonar improvement program:


"Specific software improvements included passive ranging, spatial vernier processing, full spectrum processing, dual towed array concurrent processing, low frequency active interference rejection, passive broadband, passive narrowband and passive detection and tracking processing, track management, on-board training, and port/starboard ambiguity resolution."

http://www.defenseindustrydaily.com/...ated/index.php

Yeah, with ARCI it'snow possible to run TB-16/23 and TB-29 thinline processing concurrently. It may or may not be used for resolving ambiguity, though I will say that the old-fashioned method has always worked just fine.

One gripe of mine in the game has always been the TA screens on subs. Our towed array broadband looks almost exactly what DW put in place for the FFG. This dual-sided nonsense is annoying, and clutters the screen in high contact density scenarios.

Ah ha! So subs are toting dual towed arrays, or at least planning too. :hmm: Thanks Bill. :smug:

.

Yeah those ships are pretty wide too, about 30meters in beam
http://www.fas.org/irp/program/collect/t19_bow.jpg
http://www.fas.org/irp/program/collect/surtass.htm
However, seeing what Bill just found seems like the projects has already been brought to subs, or at least soon will be. Intuitively, the wider the seperation, the better the resolution of the system, but the theory seems to still hold. Just cructhing some rough estimates...

...The speed of sound waves in water is roughly 1500m/s
With a distance of 10meters being the two arrays, the time delay between the two arrays intercepting the same signal should be on the order of a few hundredth to a few thousandths of a second. So if the signal processing is able to determine a 1/100th sec to 1/1000th sec time lag, in theory it shoud be able to delineate between the true and false contact vector with consistency, (perhaps not in turns as well). As far as range estimates, no idea here, as you said, range estimates are probably a bit trickier, and my recollection of advanced geometry and signal processing is too elementary/old.

So............ down with bearing ambiguity!!! :D :yep: :up:

Someone also pointed out that it's not clear whether this will be used for bearing ambiguity resolution or not. It might be, however, that improved processing might make phase differences that were previously not measurable measurable. It's hard to say, though.


It's called a hyperbolic fix. With two arrays you can't nail it down to a single point, but you can narrow it down to a range of possible points, from which you might be able to use other things to pick one.

Running with two towed arrays out would provide you with a potential problem: towed arrays getting tangled in each other. That is not simulated in DW.

Further, you'd need to have the two TAs exactly as far out, and I think preferrably the same depth. (Oops, there went the ability to choose what sort of TA you need, fast or sensitive...)

What Kazushima (:hmm: gotta look that up, I suspect it will translate) said (non-fixed offsets) doesn't apply all that much for bearing resolution, until they swirl too close, into each other, or even switch sides, but when they do... If you stream them far, so as to let them get deep, and far from ownship noise, etc, etc, the swirling around will be greater.

Potentially even switching sides; That's a type of bearing resolution I would not want.


The only way I could see this get done is to give the TAs a small "tail" with two hydrophones and a "ballast" to make sure it doesn't roll, or four hydrophones and a direction sensor, and place them on the end of the TA. The idea is that the exact separation length shouldn't matter, as long as you can determine which side the sound came from. Essentially, this is to transform the TA from a one-dimensional to a two-dimensional sensor.

This just increases the need for substantial separation between the two arrays.


Why do you say that?