中文摘要：

频谱感知是认知无线电网络中关键的一个环节,为了保证主用户不受干扰,要求感知算法必须具有较高的检测效率和检测精度。该文主要研究MIMO场景下的频谱感知问题,利用非圆信号的特性,提出一种基于局部最大功效不变检验（LMPIT）的频谱感知方法。根据渐近分布理论,推导了所述方法的理论检测门限。最后,采用蒙特卡洛仿真方法,分别分析了不同信道环境下该方法的检测性能,并与相关的感知算法进行对比。结果表明：在相同的环境下,文中提出的方法相比其他方法检测性能更高,且所需的采样点数更小,能够实现快速且精确的检测。

英文摘要：

Spectrum sensing is a key technology in the cognitive radio network, in order to protect the primary user, the sensing algorithms must have a high detection efficiency and detection accuracy. This paper mainly focuses on the spectrum sensing in MIMO environment. Considering that the non-circular signal is usually used in the communication system, a novel spectrum sensing method is proposed for non-circular signals based on the Locally Most Powerful Invariant Test（LMPIT）. The theoretical threshold is derived according to the asymptotic distribution theorem. Finally, the detection performance comparisons with other methods in various channels are simulated respectively. The results show that the proposed method outperforms other algorithms and only need small sample numbers, thus having higher sensing accuracy and efficiency.