In this paper, a blind algorithm for detecting active users in a DS-CDMA system is presented. This probabilistic algorithm relies on the theory of hidden Markov models (HMM) and is completely blind in the sense that no knowledge of the signature sequences, channel state information or training sequences is required for any user. Additionally, observation through an array of sensors is also considered. Performance is verified via computer simulations, showing the near-far resistance of the analyzed procedure.