A novel approach to the general problem of signal parameter estimation is described. Though the technique (ESPRIT) is discussed in the context of direction-of-arrival estimation, it can be applied to a wide variety of problems including spectral estimation. ESPRIT exploits an underlying rotational invariance among signal subspaces induced by an array of sensors with a translational invariance structure (e.g., pairwise matched and co-directional antenna element doublets). The new approach has several significant advantages over earlier techniques such as MUSIC including improved performance, reduced computational load, freedom from array characterization/calibration, and reduced sensitivity to array perturbations. Results of computer simulations carried out to evaluate the new algorithm are presented.