提升小波与能量检测联合优化下的频谱感知方法北大核心CSCD

传统的频谱感知方法易受噪声波动的干扰,而且在低信噪比的无线通信条件下检测精度较差。通过结合提升小波去噪与动态门限能量检测算法,能有效提高传统频谱感知方法的抗噪声性能和检测精度。首先对含噪信号进行奇偶抽样,分解信号,去除噪声部分,再重构为去噪新信号,然后通过能量检测方法来统计信号的能量积累,设置动态门限,最后以动态门限判断用户信号是否存在。提升小波去噪能够有效地去除采样信号中的噪声,减少噪声对能量检测法检测精度的影响,动态门限能根据噪声波动进行调整来适应复杂的噪声环境。仿真结果表明,提升小波去噪结合动态门限能量检测算法相比于传统的频谱感知要有更优的检测精度。此方法不但提高了其对不确定噪声的抵抗性,使之能适应复杂的通信环境,而且提高了频谱感知过程的可靠性。     

Adaptive Spectrum Sensing Algorithm in Cognitive Ultra-wideband Systems

Energy detection is a simple spectrum sensing technique that compares the energy in the received signal with a threshold to determine whether a primary user signal is present or not. Setting the threshold is very important to the performance of the spectrum sensing. This paper proposes an adaptive spectrum sensing algorithm where an optimal decision threshold of energy detection is derived based on minimizing the weighted sum of probabilities of detection and false alarm. Since the optimal decision threshold is dependent on the noise power and signal power, a simple, practical frequency domain approach is devised to estimate both. The algorithm can be used for the detection of various kinds of signals without any prior knowledge of the signal, channel or noise power, and is able to adapt to noise fluctuation. Simulations for detecting narrow-band and wideband signals (phase shift keying signal, frequency shift keying signal, orthogonal frequency division multiplexing signal) and ultra-wideband (UWB) signals (direct sequence spread spectrum signals) in an IEEE 802.15.3a UWB band are presented. The results show that the proposed algorithm has excellent robustness to noise uncertainty and outperforms the existing spectrum sensing algorithms in the literature.     

基于主用户信号谱特征的频谱感知方法

频谱感知是认知无线电的关键技术之一。首先讨论了一种基于信号频谱特征的频谱感知方法,提出了一种判决统计量,克服了噪声功率不确定性的影响。给出了算法流程,并采用ATSC（ad-vanced television systems committee）信号作为主用户信号对算法性能进行了仿真分析。结果表明：选择较多点快速Fourier变换（FFT,fast fourier transform）估计得到的理想主用户信号功率谱作为模板所得的检测概率较高;FFT点数相同情况下,增加谱模板估计时的周期图平均次数对检测性能没有多大改进。另外,不同观测时间下的仿真结果还表明增加接收信号的观测时间有利于改进算法检测性能。     

Degree of polarization detection: a dual‐polarized antenna based spectrum sensing algorithm for cognitive radio

A new spectrum sensing algorithm‐degree of polarization (DoP) sensing algorithm is proposed in this paper. By exploiting a pair of dual‐polarized antenna at the receiver, DoP of the received vector signal is estimated and utilized to detect the presence of primary users based on the polarization characteristics of electromagnetic waves. The dual‐polarized narrowband and broadband systems are both considered for DoP detection. In theoretical analysis, we derive the probability of detection, the probability of false and detection threshold of the proposed algorithm. It is shown that our algorithm overcomes the noise uncertainty problem. Considering the polarization‐sensitive channel impairments, the impact of polarization mode dispersion on DoP detector is discussed. This method can be utilized for various signal detection applications without requiring the knowledge of signal, transmission channel, and noise power. In simulations based on wireless microphone signals, by applying polarization information signal carries, DoP achieves a better detection performance than arithmetic‐to‐geometric mean detector, the maximum‐to‐minimum ratio eigenvalue detector, and energy detector with noise uncertainty. The simulations based on digital video broadcasting‐terrestrial signals are also presented, which may show the detection performance of the proposed method may be affected by polarization mode dispersion. Copyright © 2013 John Wiley & Sons, Ltd.     

基于图像K-means聚类分析的频谱感知算法

近年来,基于能量检测的协作频谱感知算法被广泛应用于频谱感知领域。由于该方法在计算能量检测的判决门限受噪声影响较大以及受限于认知用户的数量等问题,导致其检测性能受到影响。为了解决这一问题,本文提出一种基于图像K-means聚类分析的频谱感知算法。这种方法利用主用户信号存在与否的两种认知信号状态映射成图像,经过调整图像强度和高斯滤波预处理之后利用提取图像像素分布直方图的方法提取出特征向量,然后利用改进的K均值聚类算法对这些特征向量进行训练得到分类模型。最后利用训练好的分类模型对未知信号进行检测,从而实现频谱感知。仿真结果表明,本文所提出的频谱感知算法,在检测性能上优于传统能量检测以及协作频谱感知算法,尤其在低虚警概率、低信噪比的环境下效果更加突出。      

一种基于特征值和级联聚类的协作频谱感知方法

为了提高低信噪比(Signal-to-Noise Ratio, SNR)下频谱感知的性能，使用模糊C均值(Fuzzy C-means, FCM)和高斯混合模型(Gaussian Mixture Model, GMM),提出了一种基于特征值和级联聚类的协作频谱感知方法。从接收信号的协方差矩阵中提取特征值构造特征向量，通过在三维空间中执行聚类得到信道是否可用的分类模型，此过程无需获得主用户(Primary User, PU)信号以及噪声功率的先验信息，避免了复杂的门限计算。FCM聚类用于优化GMM聚类的初始参数，有效解决了在低SNR下GMM容易陷入局部最小值的问题。仿真结果表明，该方法降低了GMM的收敛时间并提高了模型分类的准确性，与其他主流方法相比能够有效提升频谱感知的性能。     

基于分形理论的CUWB频谱感知方法

针对超宽带频带内授权信号类型确定的特点,为了弥补盒维数检测不能够有效识别信号类型的缺点,提出通过进一步提取信息维数特征对信号调制样式进行识别,该方法融合了盒维数检测和信息维数检测的优点进行合作判决。仿真表明基于信号分形理论的频谱感知能够取得较好的检测效果,且该方法运算复杂度低,对噪声不敏感,能够有效区别噪声与授权信号,抵御模拟授权用户攻击,检测效果优于循环谱检测和能量检测。     

基于循环前缀频域自相关的OFDM信号频谱感知

针对无线通信频谱资源有限并且利用率非常低的问题,研究了认知无线电系统中基于信号典型特征的频谱感知策略,并进行动态频谱检测.提出了一种基于循环前缀频域自相关的频谱感知算法,利用正交频分复用(Orthogonal Frequency Division Multiplexing,OFDM)信号的循环前缀具有循环平稳特性,在信...     

基于分形盒维数的双门限合作频谱感知方法

在信号分形盒维数特征的基础上,提出利用噪声与授权信号分形盒维数的差异对授权用户是否存在进行检测.为了使合作感知性能趋于更优,多用户采用双门限策略进行分步合作.该方法运算复杂度低,对噪声不敏感.仿真结果表明,双门限合作相比单门限分形盒维数检测和能量检测,系统检测率更高,所需频谱感知时长较短,同时减轻了融合控制中心以及传输...     

Smart antenna‐assisted spectrum sensing for robust detection in cognitive radio networks

In this paper, we consider the problem of multiband spectrum sensing by employing smart antenna arrays at the cognitive receiver. Although energy detection is widely used for spectrum sensing in cognitive radio networks because of its simplicity and accuracy, it is severely deteriorated by the noise uncertainty. This paper introduces robust spectrum sensing techniques to circumvent this difficulty, which operate simultaneously over the total frequency channels rather than a single channel each time. To enhance the detection performance, the proposed schemes jointly utilize the information of eigenvalues and eigenvectors, signal and noise subspace components in conjunction with the likelihood functions and Gerschgorin radii. Neither subjective decision threshold setting nor the estimation of noise power is required in our schemes, making them robust to noise uncertainty. Simulations are presented to validate the performance of the proposed schemes, and the results show that our schemes can outperform other existing spectrum sensing methods. Copyright © 2014 John Wiley & Sons, Ltd.     

联合循环平稳特征PCA与RVM的频谱感知

针对无线信道环境中低信噪比情况下主用户信号检测率较低的问题,提出了一种基于循环平稳特征主成分分析(PCA)与相关向量机(RVM)的认知网络频谱感知算法。该算法结合了主成分分析算法与相关向量机分类方法,应用于解决认知网络频谱感知问题。首先对信号循环平稳特征参数进行特征提取,通过主成分分析进行降维提取信号主成分,生成训练样本和待测样本,并完成对相关向量机的训练,再采用训练完成的相关向量机算法分别对有无主用户情况下的信号进行分类检测,最后获得主用户信号存在性的感知判断。仿真实验表明,与人工神经网络、支持向量机和最大最小特征值算法相比较,所提算法在低信噪比情况下具有较高的分类检测性能,检测率最大可提高61.6%,有效地实现了对主用户信号的感知。     

基于稀疏重构的色噪声背景下未知线谱信号估计

针对色噪声背景下的未知线谱信号估计问题,该文提出一种基于分子频带处理的稀疏重构类线谱估计方法。首先,利用多速率余弦调制滤波器组对观测信号进行子带分解,得到功率谱相对平坦的子带信号。之后,在每个子带信号上,利用基于迭代最小化的稀疏学习方法进行线谱估计,并将各子带上的线谱估计结果进行频域综合滤波以及门限判决等处理。最终得到色噪声背景下的线谱估计结果。理论推导及仿真实验表明所提方法在色噪声背景下具有较好的线谱估计性能。其能够有效地去除色噪声背景,同时保留稀疏重构类线谱估计方法所具有的高频率分辨力等优点。     

Non-parametric Blind Spectrum Sensing Based on Censored Observations for Cognitive radio

In cognitive radio, spectrum sensing is a challenging task. In this paper, a spectrum sensing method based on censored observations is proposed. We call it as Censored Anderson Darling (CAD) sensing. We present the performance of the CAD sensing method with receiver operating characteristics (ROC) in fading channels using simulations. It is observed that the proposed method outperforms the conventional energy detection (ED) at lower signal to noise ratio. It also provides better detection performance compared to Ordered Statistics (OS) based sensing method. We also use the CAD sensing method assuming noise uncertainty. It means noise variance at secondary user(SU) is unknown. We call it as Blind CAD sensing (B-CAD). We also show that the B-CAD outperforms the energy detection.     

认知系统中多天线协作频谱感知方案

针对认知无线电的核心问题——频谱感知,采用性能好的协作频谱感知,这里研究了认知无线电系统中一种多天线协作频谱感知方案,此方案中的噪声信号和主用户的信号均认为是独立复高斯随机信号。同时,次用户将检测到的信号通过波束成形后传向融合中心,而优化函数为发射功率受限的条件下,最大化全局的检测概率。理论推导和方针结果表明,所提出的方案有效地提高了检查概率,充分发挥了空间分集和多用户分集的优势,普遍提高了系统的感知概率。     

采用离散小波变换的认知无线电频谱能量检测

在认知无线电系统中,频谱检测是搜索空闲信道,避免对授权用户产生有害干扰的关键环节。本文提出了一种离散小波变换与时域能量检测相结合的频谱检测方法,对SU共享的宽带信道中的窄带PU信号进行预检测。首先对接收信号进行离散小波变换,获得能够反映信道频谱变化的细节小波系数,然后以该系数作为统计量,对其进行时域能量统计计算。该方法计算量较小,容易实现,可进行多分辨率分析,能够提高检测的灵敏度;不需要被检测信号的先验知识,适用于检测各种未知信号。仿真实验对无线麦克信号和地面无线数字电视信号进行了检测,验证了该方法的正确性。      

基于信号能量分布拟合优度的长短时记忆网络频谱感知算法研究

传统频谱感知算法性能在低信噪比下不够理想,在高信噪比下较好,算法性能随信噪比降低逐渐变差。本文提出了基于信号能量分布拟合优度的长短时记忆网络频谱感知算法,利用授权用户信号存在时的接收信号为基础,计算接收信号的能量分布,并将通过拟合优度算法得到的距离值作为特征构造特征向量,然后将特征向量输入长短时记忆网络训练得到模型,最后将测试数据输入训练模型进行预测,从而实现频谱感知。仿真结果表明,本文提出的新算法在信噪比为-13 dB,采样点数为28时,检测概率达到96.21%,明显优于传统能量检测算法和传统拟合优度算法。      

噪声功率不确定下协同频谱感知最优融合算法

研究了噪声功率波动性对单点和协同频谱感知性能地影响。分析了在此环境下,单点基站的检测概率与采样点个数的关系,并得出当信噪比不小于噪声功率波动系数的两倍时,基站仍可正常检测。在协同网络中,给出了此环境下数据融合方法的最优权重设定。理论分析与实验表明,相比于传统方法,文中所提最优数据融合算法在各种信噪比环境下性能均较优,尤其在各基站信噪比差异较大时,性能更为显著。     

基于随机共振和信息几何的协作频谱感知方法

为提高频谱感知系统在低信噪比环境下对微弱信号的感知性能,提出了一种基于随机共振技术和信息几何理论的频谱感知方法。首先通过随机共振技术增强输入信号的能量,以提高感知信号的信噪比。然后,基于信息几何理论将信号矩阵的协方差矩阵对应成流形上的点,并计算流形上样本点之间的散度距离作为感知信号的特征数据。最后,采用BP神经网络对信号特征数据进行分类,有效避免了决策阈值的计算,快速实现了频谱决策。仿真实验证明,所提方法在低信噪比条件下具有更好的感知性能,有效提高了复杂环境下的频谱检测概率。     

认知无线电中一种改进的AVC频谱感知算法

AVC算法是一种适用于拉普拉斯噪声环境的常用频谱感知算法,但该算法并未充分平滑拉普拉斯噪声中的“尖峰”,导致算法的检测性能不佳.针对此,提出一种改进的AVC频谱感知算法,其原理是对接收信号绝对值做开根号处理,并累加处理结果,作为检验统计量,进而判决是否存在主用户,实现频谱感知.此外,利用中心极限定理推导了所提算法检验统计量在主用户不存在时的概率密度曲线,从而给出理论判决门限.仿真表明,所提算法的检测性能分别优于AVC感知算法和拉普拉斯噪声下的能量检测算法大约1 dB和4 dB.     

Reducing the effects of the noise uncertainty in energy detectors for cognitive radio networks

Because of its ease of implementation and minimum requirements about the primary signals' information, energy detection is broadly considered for signal detection in spectrum sensing algorithms. However, the noise uncertainty phenomenon, caused by the random variations in the noise power, degrades the performance of an energy detector, particularly when the signal‐to‐noise ratio (SNR) is low. In this work, we propose to reduce the negative effects of the noise uncertainty in the performance of an energy detector by dynamically adapting its detection threshold to the noise conditions experienced at each sensing epoch. The noise power is estimated from the received signal samples using an algorithm based on a high‐pass filters bank and median filtering. With our proposal, it is possible to maintain a constant and low false alarm rate in the presence of noise uncertainty, without increasing the probability of misdetection, even in the low SNR regime, and without increasing the number of samples considered for spectrum sensing. Copyright © 2014 John Wiley & Sons, Ltd.