Estimation based cyclostationary detection for energy harvesting cooperative cognitive radio network

The two prime dynamics that are steering the future wireless communications are energy efficiency and efficient spectral usage. Energy harvesting (EH) when integrated with cognitive radio (CR) technology guarantees to address the challenges of limited availability of energy and spectrum. In this paper, we propose an estimation based cyclostationary spectrum sensing technique in an energy harvesting cooperative cognitive radio network (EH-CRN). The analytical model for the assessment of the performance is established under various fusion rules with varying numbers of nodes in a centralized cooperative CRN. The performance is investigated with respect to the harvested energy and overall throughput of the network. We model the analytical framework for the detection performance, energy harvesting, and throughput in an EH-CRN scenario. The implication of the various network parameters viz., detection frames, number of CR users operating in cooperation, collision probability, prediction error upon the throughput of the network is studied. A comprehensive comparative analysis is made between the performance of an estimation based cyclostationary detection (ECSD) and estimated noise power based energy detection strategy (ENP ED). The results indicate that the proposed estimation based cyclostationary approach provides a better performance regardless of the low SNR and is immune to noise uncertainty as compared to other techniques.