Passive sensors localize ships without geoacoustic information

Underwater localization is important for naval applications and marine mammal monitoring and can be separated into two main categories. Active sonar works much like echolocation: it sends out a ping and waits for a reflection. Meanwhile, passive sonar uses an acoustic array to listen and requires an accurate model for the sound propagation beneath the surface of the water.

Passive techniques can take advantage of noise that is already present in the underwater soundscape without knowing where it came from. While that poses one advantage, these techniques often depend on geoacoustic models of the solid material beneath the ocean, and characterizing that material is expensive and technically challenging.

To avoid using geoacoustic models, Akins and Kuperman developed Modal-MUSIC, an array processing method to localize ships that exploits the structure of ship noise. The technique requires observations at various depths in a water column and information about the index of refraction in that regime.

“It uses this knowledge to explore the space of possible excited modes in the water columns,” said author Hunter Akins. “The technique is new because it combines the physics specific to underwater acoustics with a well-known high-resolution method to extend the capabilities of an array.”

Other passive techniques require fully-spanning water column arrays, but Modal-MUSIC works with a partially spanning array.

“The study shows that, without knowing where a ship is, the information required to localize it is contained in the structure of the noise it radiates,” said Akins.

The team plans to extend the method to estimate properties of the sea floor.

Source: “Modal-MUSIC: A passive mode estimation algorithm for partially spanning arrays,” by Hunter Akins and W. A. Kuperman, JASA Express Letters (2022). The article can be accessed at https://doi.org/10.1121/10.0012041 .