I. Introduction
Future satellite communication systems are expected to provide high speed multimedia and broadband services. As a matter of fact, the satellite industry is targeting at not only areas without backbone connectivity (e.g., maritime, aeronautic, and other extreme remote sites), but also high dense populated scenarios with an existing communication infrastructure to decongest the terrestrial wireless network [1]. However, under exclusive regulation, currently deployed spectrum can hardly meet the ever increasing spectrum demand of resource-consuming multimedia applications in satellite communications [2], [3]. To accommodate more wireless services within the limited spectrum, cognitive radio has emerged as a promising technology to alleviate the scarcity of spectrum for satellite communications by enabling spectrum sharing between satellite and terrestrial networks, referred as cognitive satellite terrestrial networks [4], [5]. Various spectrum sharing approaches are suggested for cognitive satellite terrestrial networks, e.g., underlay, overlay, and interweave [6]. The underlay mode is especially attractive due to its high spectral efficiency, which allows cognitive systems to reutilize the spectrum where incumbent signals are present without obstructing the normal operation of the primary licensed systems.