Minimum stable convergence criteria for stochastic diffusion search

An analysis of Stochastic Di®usion Search, a novel and e±cient opti-misation and search algorithm, is presented, resulting in a derivation of the minimum acceptable match resulting in a stable convergence within a noisy search space. The applicability of SDS can therefore be assessed for a given problem. Introduction: Stochastic Di®usion Search (SDS), ¯rst described in [1], is an e±cient probabilitistic multi-agent global search and optimisation technique that has been applied to diverse problems such as site selection for wireless net-works [2], mobile robot self-localisation [3], object recognition [4] and text search [5]. Additionally, a hybrid SDS and n-tuple RAM [6] technique has been use