A Particle Filtering Approach for Enabling Distributed and Scalable Sharing of DSA Network Resources

Abstract: Handling the massive number of devices needed in numerous applications such as smart cities is a major challenge given the scarcity of spectrum resources. Dynamic spectrum access (DSA) is seen as a potential candidate to support the connectivity and spectrum access of these devices. We propose an efficient technique that relies on particle filtering to enable distributed resource allocation and sharing for large-scale dynamic spectrum access networks. More specifically, we take advantage of the high tracking capability of particle filtering to efficiently assign the available spectrum and power resources among cognitive users. Our proposed technique maximizes the per-user throughput while ensuring fairness among users, and it does so while accounting for the different users' quality of service requirements and the channel gains' variability. Through intensive simulations, we show that our proposed approach performs well by achieving high overall throughput while improving user's fairness under different objective functions. Furthermore, it achieves higher performance when compared to state-of-the-art techniques.