Calculating the pulsed response of linear arrays: accuracy versus computational efficiency

Abstract

The accuracy and computational efficiency of different methods for calculating the field profile from linear arrays are examined. Specifically, various approximations used for time and frequency domain solutions are studied and compared to the exact time domain impulse response method, which is taken to be the "gold standard". The accuracy is measured in terms of the error in the 6 dB spatial resolution, whereas the efficiency is represented by the normalized computation time. The results demonstrate that in the frequency domain the Fresnel approximation yields accurate results (errors <1%) for the unsteered array, but that the accuracy degrades with increasing steering angles (>10/spl deg/). It is also shown that, in general, this method is computationally more efficient as compared with the exact method since the efficiency of the latter depends highly on the field point location and array geometry.