Cognitive Radio technique has been recognized as one of the most promising solutions for the increasingly growing problem of spectrum scarcity in wireless networks, specially with the emerging of the Internet of Things. In cognitive radio networks, secondary users are allowed to intelligently access licensed bands of primary users, thus enhancing the spectrum utilization. In this context, for investigating the advantages of cognitive radio, Machine Learning techniques have been widely applied to predict primary users arrivals. However, the available simulators are usually complex and highly time consuming. Therefore, in this work, we propose a simple and intuitive primary user arrivals data generator, MARIO, that can produce random arrival data for multiple channels by employing Poisson process. This generator is validated by using the generated data to predict new sequences according to a Hidden Markov Model. Our results show that the data generator can be used to simulate various traffic patterns over different channels.