A comparison of computing architectures for ultrasonic signal processing

The design of advanced sonar modules implies the use of more and more complex processing algorithms, in order to obtain as much information as possible from the environment. The use of simultaneous emission and reception techniques in certain transducers, that form an ultrasonic sensorial module, allows to increase the benefits and the performance of this type of systems. These developments make possible to obtain more information for every ultrasonic emission, diminishing the scanning time and improving the interpretation and usefulness of the obtained results. These techniques are usually based on the encoding of the ultrasonic emission, so the received signals can be correlated in order to search For possible echoes. As a consequence, the computational complexity necessary to carry out the implementation of the new algorithms increases considerably. The analysis of possible computing platforms for the real-time processing of the new algorithms becomes a task of great importance. In this work, some computing architectures are analysed and compared, in order to determine which scheme is most suitable for the ultrasonic signal processing.