An energy efficient fair node selection for cooperative in-band and out-of-band spectrum sensing

In a cooperative cognitive network, multiple secondary user (SU) nodes are selected for spectrum sensing to detect the presence of primary user (PU) nodes in the currently used channel as well as identify other available unused licensed channels. Periodic sensing of such in-band (IB) and out-of-band (OB) channel consumes a considerable amount of energy which is a huge liability to an energy constrained networks. Additionally, SU nodes are selected to meet the requirements such as detection and false alarm probabilities. This results the consecutive selection of same nodes, which depletes the energy of those nodes rapidly shortening the lifetime of nodes and network itself. In this paper, we propose a fair and energy efficient node selection scheme for joint IB and OB spectrum sensing. The SU nodes are grouped based on the energy stored in it referred as residual energy. A joint optimization problem is formulated to minimize energy consumption while ensuring the selected nodes belong to specific group depending on whether sensing is IB or OB. Considering the nature of formulated problem, we use particle swarm optimization to solve it. The results show that proposed scheme not only consumes relatively less energy and reduces the number of dead nodes, but also allows to maintain fair distribution residual energy within the network. Consequently, the network can prolong its spectrum sensing activities for longer duration.