An exercise environment to experiment with acoustic arrays using sonar as an example.

As the environment consists of a single Python class, no installation of sonar is required. Placing sonar.py in the working directory should suffice. A Python interpreter is, however, required. We recommend using Anaconda. The field simulations are implemented using the Python package pyfds, which can be installed using pip (comes with Anaconda) by the following command:

pip install pyfds

Here is a minimal working example (to be placed in the same directory as sonar.py):

from sonar import Shallows

if __name__ == '__main__':

    field = Shallows(scenario='2020')
    result = field.ping(
        positions=[0, 1],
        delays=[0, 0.1e-3],
        show=True
    )

The first line imports the class Shallows from the sonar module (the Python file sonar.py). The following statement (placement in if __name__ == '__main__':) is necessary for the animation of the field simulation to work. Next up, an object of the type Shallows called field is created, using the scenario for 2020. The last statement allows you to send a number of pings (acoustic impulses) at given positions and with delays. Note that all delays should be positive. The example sends pings at 0 m (center of the field) and 1 m with the first ping having no delay and the second one being delayed by 0.1 ms. The display of the field simulation can be disabled by setting show=False.

This scenario constitutes a classic echolocation example and can be used to experiment with sonar imaging algorithms. Note that not all excitation positions are valid, as the field is spatially quantized with an increment of 0.05 m. After the simulation is finished, the return value of ping contains a list of echo signals recorded at the positions the pings where send at. The transmitted signals are also present in these signals.

This scenario simulates medical ultrasound applications, e.g. lithotripsy. A small, high-density object is placed at a randomized position in the field. The field is spatially quantized with an increment of 0.5 mm, so excitation positions need to adhere to this quantization. The return value of ping is a tuple that contains the signal at the center of the high-density object and a list of echo signals recorded at the positions the pings where send at.

This scenario simulates echo ranging disturbed by a step on the floor of the simulation regime. The aim is to test different transducers apertures for robustness to the interfering signals reflected by the step. The transducer's aperture is defined indirectly by the size in two dimensions. Note that in this case the class Step has to be used. The size of the step and the distance from the acoustic axis of the transducer can be set when crating the Step object. The size of the transducer is specified with the ping method, which returns the signal transmitted and received by the transducer, e.g.:

from sonar import Step

if __name__ == '__main__':

    field = Step('2024', height=10, distance=10)
    signal = field.ping(size=0.5, show=True)