Real time-frequency active sonar processing: a massively parallel approach
A paradigm for massively parallel processing of matched filters, replica correlators, ambiguity functions and time-frequency distributions using an SIMD (Single Instruction Multiple Data) programming methodology is presented. It is shown that active sonar detection algorithms, as implemented by frequency domain processing, can be a "natural match" to an SIMD methodology. Using the Connection Machine, an 8192 node SIMD parallel processor, this methodology holds great promise in meeting the extensive computational needs of enhanced active sonar systems. The decomposition process is presented and examples, output of the computer program CMASP (Connection Machine Ambiguity Surface Processor), are given. CMASP can provide real time simultaneous multiple beam, Doppler and waveform replica correlations. Synthetic data are processed and the corresponding CMASP outputs are displayed as three-dimensional ambiguity surfaces on networked graphic workstations. Because of efficient problem decomposition, in addition to the target bearing, range and velocity information as provided by continuous ambiguity surfaces, other time-frequency processing can be exploited. Specifically, instantaneous-like time-frequency distributions can be realized (e.g. Wigner, Rihaczek distributions) in which the data set is presented and processed as time-varying spectral reresentations.
SourceIn: Low frequency active sonar (SACLANTCEN Conference Proceedings CP-42), vol. 2, 1993, pp. I/28-1 - I/28-29.
Zvara, George P.