Abstract: A novel channel reservation statistical prediction mechanism based on statistical analysis is proposed to solve the problem of performance deterioration of the random competition reservation mechanism when the load becomes heavy, which decreases the channel utilization of satellite network. The predication interval which the availability of slots for reservation (SR) in the next frame possibly belongs to, is calculated by uniformly minimum variance unbiased estimation and the Faulkenberry theory based on multiavailability division of SR. The precise predication of prospective availability of SR is obtained according to the transition probability of the Markov model and frequency estimation of SR availability appearance times in history information. User terminals adjust their transmitting probability of reservation requests according to the availability predication of SR in the next frame. The results of theoretical analysis and simulation both indicate that the performance of applicability and availability of the novel mechanism is good, while the predication accuracy has been improved obviously. The collision probability of reservation requests has been decreased and the channel utilization has been enhanced.