In high-speed wireless communication systems, direct-conversion transceivers (DCTRs) are in demand due to low implementation cost of analog front ends. However, the DCTRs are more susceptible to radio frequency impairments like in-phase and quadrature (I/Q) imbalance. This has been shown to significantly limit the spectrum sensing (SS) capability of a secondary user to find spectrum holes. Another source of performance limitation in multiantenna cognitive radio (CR) systems is the uncalibrated receivers across the antennas. Unlike previous works, which considered these two impairments separately, we model the received signal as a single-input multiple-output CR system in the presence of transmit (TX) and receive (RX) I/Q imbalances and when the receiver is uncalibrated across the antennas. In such a scenario, we model the SS problem as a composite binary hypothesis testing problem, apply the likelihood ratio test principle, and develop a new robust SS approach. The performance of the proposed SS algorithm is evaluated using extensive Monte Carlo simulations to illustrate the superiority of the proposed detector in comparison with the existing SS algorithms in the literature. The obtained results indicate that the proposed SS algorithm has constant false alarm rate property with respect to any nonuniform noise variance uncertainties as well as RX I/Q imbalance parameters, thereby being a robust and blind detector.

中文翻译：

---

联合TX / RX I / Q不平衡和未经校准的接收器下的认知无线电频谱感知

在高速无线通信系统中，由于模拟前端的实现成本较低，因此需要直接转换收发器（DCTR）。但是，DCTR更容易受到射频损伤的影响，例如同相和正交（I / Q）不平衡。事实证明，这严重限制了次要用户查找频谱孔的频谱感测（SS）能力。多天线认知无线电（CR）系统中性能限制的另一个来源是天线上未经校准的接收机。与先前的工作（分别考虑这两个减损）不同，我们在存在发射（TX）和接收（RX）I / Q不平衡以及接收器未经校准的情况下，将接收信号建模为单输入多输出CR系统天线。在这种情况下 我们将SS问题建模为复合二元假设检验问题，应用似然比检验原理，并开发出新的鲁棒SS方法。使用广泛的蒙特卡洛模拟评估了所提出的SS算法的性能，以说明所提出的检测器与文献中现有的SS算法相比的优越性。获得的结果表明，所提出的SS算法对于任何非均匀噪声方差不确定性以及RX I / Q失衡参数具有恒定的虚警率特性，从而是一种鲁棒且盲目的检测器。使用广泛的蒙特卡洛模拟评估了所提出的SS算法的性能，以说明所提出的检测器与文献中现有的SS算法相比的优越性。获得的结果表明，所提出的SS算法对于任何非均匀噪声方差不确定性以及RX I / Q失衡参数具有恒定的误报率属性，从而是一种鲁棒且盲目的检测器。使用广泛的蒙特卡洛模拟评估了所提出的SS算法的性能，以说明所提出的检测器与文献中现有的SS算法相比的优越性。获得的结果表明，所提出的SS算法对于任何非均匀噪声方差不确定性以及RX I / Q失衡参数具有恒定的虚警率特性，从而是一种鲁棒且盲目的检测器。