On the connectivity of multi-band cognitive radio ad hoc networks

The connectivity of multi-band cognitive radio ad hoc networks is studied using a new natural connectivity based metric and compared to the traditional probability of finding a route metric. The analysis is developed based on a generalized cognitive radio networks model where multiple primary users with independent activity factors operating in multiple frequency bands are considered. The adjacent graph matrix elements, representing the weights associated to the network graph edges, are set proportional to the activity factors of the primary users. It is shown that the proposed connectivity metric is convex with the primary users activity factors and can efficiently capture the behaviour of the probability of finding a route metric with a significant computational complexity reduction. Finally, the proposed metric presents a solid analytical tool for planning and studying the effect of different network parameters on the connectivity of cognitive radio networks.