Capacity and delay in mobile cognitive networks with random way-point mobility model

We study the throughput and delay scaling laws of the mobile cognitive network. There are nn primary users and mm secondary users in the primary network and the secondary network, respectively. They are distributed uniformly and independently on the surface of a unit torus, moving according to the random way-point mobility model (RWMM). The primary users have priority to access the spectrum, and do not alter their protocol in the presence of the secondary network. The secondary users opportunistically use the licensed spectrum when it remains unused. Under the improved 2-hop relay policy, we derive the constant throughput of Θ(1) per primary source–destination (S–D) pair when the radius of the preservation region rp=Θ(1/m). The expected delay scales as Θ(n/v(n)), where v(n)v(n) is the average speed of the primary nodes. Moreover, if the secondary network is denser than the primary network, i.e., m=nγm=nγ with γ>1γ>1, the constant throughput of Θ(1) per secondary S–D pair and the expected delay of Θ(m/v(m)) can be derived surprisingly when the radius of the avoidance region rs=Θ(1/n) and rp=Θ(1/m), where v(m)v(m) is the average speed of the secondary nodes.