Node Selection Scheme for Cooperative Spectrum Sensing in Cognitive Radio

Abstract

Spectrum sensing is an important part of cognitive radio where the goal is to find out whether primary user is present or absent. This can be done in numerous ways such as by cooperative spectrum sensing, non-cooperative spectrum sensing and interference based spectrum sensing. We however use the cooperative spectrum sensing where the data sensed by a secondary user is shared among all the nodes. Since sharing of data with every node increases the data overhead largely so instead we use a base station called the fusion center to which all the data is sent. Since all nodes take part in the cooperation of data, the overall energy consumed is high. A single node has to sense the spectrum, make a local decision and then transmit it further. All these steps may consume different energy for differently located nodes. Hence some nodes may not have sufficient energy to complete these steps. Due to the lack of residual energy, these nodes may provide wrong information to the fusion center and reduce the reliability of the system. To tackle the problems such as high energy consumption, data overhead, and effect of malicious nodes we use a node selection scheme. In this scheme our motive is to filter out ineffective nodes and cooperate only between the genuine nodes.