Influence of spatial statistics of spectrum use on the performance of cognitive wireless networks
Janne Riihijärvi, Jad Nayef Nasreddine, Petri Mähönen · 2012
We study the performance characteristics of cognitive wireless networks under different dynamic spectrum access scenarios. Our focus is especially on the influence of spatial structures of the primary and secondary user networks on the achievable performance of the secondary network. We adopt techniques from spatial statistics to develop stochastic models for the structure and interaction of these networks. The chosen models are based on Gaussian random fields and Gibbs point processes, and are firmly grounded on empirical data. We then apply extensive Monte Carlo simulations to study the behavior of these models and their related performance properties. The models and the applied techniques are applicable also to a wider variety of networking problems. Our results provide first quantitative assessments on the influence of the spatial structure, and especially correlation properties, of the involved networks on the expected performance and thus on the utility of dynamic spectrum access based systems.