Joint sensing and power allocation for hybrid spectrum sharing in fading channels

In a sensing-based hybrid spectrum sharing paradigm, cognitive radio first performs spectrum sensing to identify primary users´ states (idle/busy) and then adapts its transmit power according to sensing outcomes and channel conditions. To investigate the capacity of such systems in fading channels, existing works modelled fading channel in transmission phase while additive white Gaussian noise channel in spectrum sensing; however, sensing channels also exhibit fading characteristics in practice. Therefore a more realistic system model with channel fading in both sensing and transmission is considered in this study. Under the new system model, spectrum sensing and power allocation are coupled in the ergodic capacity and an equivalent decoupling processing is proposed via mathematical manipulations. Further, joint sensing and power allocation over Rayleigh fading is studied under average interference and transmit power constraints. The optimal and suboptimal schemes are obtained by alternating optimisation and Lagrangian dual method. Finally, system performance is evaluated via extensive numerical simulations.