In traditional call admission control (CAC) schemes, mobile users are always the passive roles during the admission procedures and the base station determines whether to admit or reject the call requests without the involvement of mobile users. In this paper, we propose a novel stochastic CAC framework that allows the mobile user to be an active entity during the CAC process. The objective of each mobile user is to maximize its utility function based on its own decision (i.e., to join the queue or not to join the queue). The optimal stochastic decision is in the sense that the probability for a mobile user to join the queue will maximize the expectation of the utility function. In other words, in the long run, the mobile user will bene t from the optimal joining probability p∗ . We further show that the optimal join probability for a mobile user could be quite different depending on the number of mobile users waiting in the queue. Finally, we illustrate the structure of the optima...