Comparative study of centralized and decentralized localization methods

Decentralized detection (DD) is an interesting alternative to classical centralized detection (CD) when communicational/computational constraints play an important role in the design of the decision system. The authors investigate the feasibility of decentralized techniques for realistic situations, and their relative performance with CD when the Neyman-Pearson criterion is chosen, where relatively good performance of DD is shown. Of special interest seems to be the inherent robustness of DD architectures.<>