基于多维空间的射频频谱检测技术研究

发布时间：2018-01-04 19:21

本文关键词：基于多维空间的射频频谱检测技术研究　出处：《东南大学》2016年博士论文　论文类型：学位论文

【摘要】：随着无线通信技术的应用和发展,频谱资源相对于日益增加的通信需求变得十分有限,频谱检测是发现空闲频谱和提高频谱资源利用率的关键技术。对应于信号传输的多维性,空闲频谱也具有多个维度。除了频率作为固有维度外,另外两个最基本的维度就是时间和空间。本学位论文深入研究认知无线电(CR, Cognitive Radio)网络环境在时间维度、空间维度以及联合时间与空间维度上的多维空间频谱检测。首先在时间维度上研究了CR网络频谱感知中能量检测的最优门限、CR网络中基于Dempster-Shafer(D-S)证据理论的合作频谱感知方法和CR网络噪声不确定环境下频谱感知中的双门限检测方法；然后在空间维度和时间维度上研究了适用于CR网络典型场景的空闲频谱检测方法和CR网络中采用平均检测概率约束的空时二维感知方法。论文的主要研究成果如下：1.针对频谱感知中的能量检测,得到了一个本地检测的最优门限和采用大数融合规则进行多用户合作频谱检测的次优门限。分析与仿真结果表明,单用户检测时,本地最优门限使得虚警概率和漏检概率相加后的系统错误概率最小；在较低信噪比(SNR, Signal Noise Ratio)且采用大数融合规则进行多用户合作检测时,次优门限对应的系统错误概率接近最小值。2.针对合作频谱感知,提出了一种基于D-S证据理论的频谱检测方法,该方法对冲突证据源和融合规则进行了修正。分析与仿真结果表明,同“与”、“或”融合及其他D-S证据理论融合方法相比,所提方法进一步提高了CR网络检测性能,并可作为加性高斯白噪声(AWGN, Additive White Gauss Noise)信道和Rayleigh信道等网络环境的通用频谱感知方法。3.针对噪声不确定环境下频谱感知中的能量检测,提出了一种双门限的检测方法。该方法根据噪声不确定范围设置高、低两个检测门限,只对介于两检测门限之外的检测结果进行判决。分析与仿真结果表明,该方法克服了噪声不确定的影响,并且噪声不确定环境下得到的检测性能要优于噪声确定环境下采用精确噪声功率得到的检测性能。在AWGN信道及Rayleigh信道下应用该方法,检测性能均有显著改进。4.针对CR网络中若干典型场景,提出了一种空闲频谱检测方法,通过排除空间虚警(SFA, Space False Alarm)影响,得到单个认知用户(从用户)在路径损耗和Rayleigh衰落影响下的空闲频谱检测概率,同时得到路径损耗影响下多用户合作检测及单用户在宽带环境下的空闲频谱检测概率。分析与仿真结果表明,所提方法的空闲频谱检测概率与采样次数有关；在该方法下采用多用户合作检测时的空闲频谱检测概率未必高于单用户检测；Rayleigh衰落会使空闲频谱检测概率进一步减小,而增加带宽则会使空闲频谱检测概率大大提高。5.针对CR网络的空闲频谱检测,提出一种采用平均检测概率约束的空时二维感知方法。以检测概率在整个检测区域内的平均值,作为从用户接入授权用户(主用户)频谱的约束,并由此设置检测门限,在理论上得到时域、空域及空时二维空闲频谱接入概率。分析和仿真结果表明,该方法所得到的时域虚警概率和空域虚警概率均降低,空时二维空闲频谱的接入概率显著提高。
[Abstract]:With the development and application of wireless communication technology, spectrum resources relative to the increasing demand for communication is very limited, the spectrum detection is to discover the idle spectrum and improve the spectrum utilization rate. The key technology of multi-dimensional corresponding to the signal transmission, the idle spectrum also has multiple dimensions. In addition to the inherent frequency as the dimension, the other two most the basic dimension is time and space. In this thesis, the research of cognitive radio (CR Cognitive Radio) network environment in the time dimension, space dimension and multidimensional space spectrum detection combined with time and space dimensions. First, in the time dimension of the optimal threshold energy spectrum sensing detection in CR network, Dempster-Shafer network based on CR in (D-S) evidence theory of cooperative spectrum sensing method and CR network environment uncertainty noise detection method in spectrum sensing threshold; Then in the space and time dimensions of the average detection probability constraint space for CR network scenario idle spectrum detection method and CR network in two dimensional perception method. The main research results of this paper are as follows: 1. for energy detection spectrum sensing, get the suboptimal threshold of an optimal local threshold detection and the large number of fusion rules for multi-user cooperative spectrum detection. Analysis and simulation results show that the single user detection, local optimal threshold makes the false alarm probability and the failure probability obtained by the system error probability is small; in the low signal-to-noise ratio (SNR Signal, Noise Ratio) and the use of large numbers of fusion rules multi user cooperative detection, the minimum value of.2. for cooperative spectrum sensing suboptimal threshold corresponding to system error probability, proposes a detection method based on spectrum D-S evidence theory, The method to correct the conflict of evidence sources and fusion rules. The analysis and simulation results show that with the "and" and "or" fusion and other D-S evidence theory fusion methods, the proposed method improves the detection performance of CR network, and can be used as additive white Gauss noise (AWGN, Additive White Gauss Noise) the channel and Rayleigh channel network universal spectrum sensing method for.3. uncertain environment noise energy detection spectrum sensing, a new method is proposed to detect double threshold. According to the noise uncertainty range is set high, two low detection threshold, only to detect between two detection threshold judgment results analysis and simulation results show that the method can overcome the influence of noise and uncertainty, and the uncertainty noise detection performance is superior to the environment under noise environment using precise determination of noise power The detecting performance is improved. The application of this method in AWGN channel and Rayleigh channel, detection performance significantly improved.4. for some typical scenes in CR network, proposed an idle spectrum detection method, by eliminating false space (SFA, Space False Alarm), a single cognitive user (from the user) in path loss and Rayleigh fading idle spectrum detection probability under the influence, at the same time be in the broadband environment idle spectrum detection probability of multi user detection and cooperative path loss under the influence of a single user. Analysis and simulation results show that the idle spectrum detection probability of the proposed method and sampling times; in the method using the idle spectrum detection probability when the user cooperative detection may not be higher than the single user detection; Rayleigh fading makes the detection probability of idle spectrum is further reduced and increased bandwidth will make the idle spectrum detection probability The rate of.5. greatly improve the CR network idle spectrum detection, proposes a two-dimensional sensing method using average detection probability constraints. The space-time average detection probability in the detection area, as an authorized user from user access (primary user) spectrum constraint, and then set up a threshold, obtained in the time domain theory airspace, and space-time spectrum access probability. Time domain analysis and simulation results show that the false alarm probability and false alarm probability space by the method are reduced, the probability of access idle spectrum significantly improved two-dimensional space-time.

【学位授予单位】：东南大学
【学位级别】：博士
【学位授予年份】：2016
【分类号】：TN92;TM935.21

【相似文献】