Inefficient utilization of the authorized spectrum emerges cognitive radio (CR) as a hopeful technology for both present and future telecommunications. It is owing to the potency to leverage the obtainable bandwidth of other wireless communication networks and thereby increase its occupancy. The key feature for the cognitive radio system for distinguishing the blank spectrum is spectrum sensing. This paper is intended to establish a hybrid sensing model for spectrum detection in CR to enhance the sensing efficiency of traditional techniques of spectrum sensing, which consists of two parallel paths of hybrid detectors. The first path is formed from two sequential detector stages; in the first phase, energy detector is used to recognize the PU signal existence where the signal has not been identified. Maximum–Minimum Eigenvalue (MME) is used as a second stage to detect the PU signal presence. The second path consists of two parallel stage detectors employing separate ED and MME to detect the PU signal individually, the two results are gathered to make a decision, and then the final detection decision is determined based on the two paths’ detection combined results. The proposed hybrid sensing approach adopted for enhancing the sensing performance is validated with conventional methods. Simulation results show that the proposed approach outperforms various traditional and hybrid approaches in terms of maximizing the detection probability on the specified limitations on the false alarm probability, as it can increase the detection probability to 94% instead of 79% for the parallel detector at SNR = − 10 dB and P_fa = 0.1.

对授权频谱的低效利用使认知无线电（CR）成为当前和未来电信的一种有希望的技术。由于有能力利用其他无线通信网络的可获得带宽，从而增加其占用率。认知无线电系统区分空白频谱的关键特征是频谱感测。本文旨在建立一种用于CR中频谱检测的混合感测模型，以提高传统频谱感测技术的感测效率，该技术由混合检测器的两条平行路径组成。第一路径由两个顺序的检测器级组成；在第一阶段，能量检测器用于识别尚未识别信号的PU信号的存在。最大-最小特征值（MME）被用作检测PU信号存在的第二阶段。第二条路径由两个并行级检测器组成，分别使用独立的ED和MME分别检测PU信号，收集两个结果以做出决策，然后根据两条路径的检测合并结果确定最终的检测决策。为提高感测性能而采用的提议的混合感测方法已通过常规方法进行了验证。仿真结果表明，该方法在指定的虚警概率限制条件下最大程度地提高检测概率方面优于传统方法和混合方法，因为它可以将并行检测器在SNR时的检测概率提高到94％，而不是79％。 = − 10 dB和P _fa  = 0.1。