Energy-Efficient Quantized Data Fusion Based on Differential Mechanism in Cognitive Radio Networks

In the field of cognitive radio (CR), cooperative spectrum sensing (CSS) utilizes the spatial diversity of each secondary user (SU) to accurately detect spectrum hole. Although the information exchange and decision fusion in CSS improves sensing accuracy, it also brings additional communication overhead and energy consumption. Since the energy of SUs is limited, it is necessary to improve the sensing performance while reducing the energy consumption, i.e., maximizing the energy efficiency (EE). In this paper, two effective bitwise differential voting rules (DVRs) are proposed to reduce the number of bits reported without compromising sensing performance. Meanwhile, this paper derives theoretical performance expressions for the proposed bitwise DVRs and proposes a differential mechanism based quantized data fusion (QDF) to achieve optimal EE. The proposed QDF optimizes the quantization interval and the global decision threshold to achieve the highest EE and regulates the number of quantization bits through the differential mechanism. Finally, the numerical simulation results confirm the effectiveness and robustness of the proposed bitwise DVRs. Moreover, the proposed energy-efficient differential mechanism based QDF shows superiority over other energy-efficient QDF techniques that only apply DVR in terms of EE.