认知无线电中的频谱感知与干扰检测技术研究

发布时间：2018-06-23 06:03

本文选题：认知无线电 + 频谱感知　；参考：《华中科技大学》2014年博士论文

【摘要】：随着无线通信技术的飞速发展,相关的设备和应用也与日俱增,同时,有限的无线频谱资源也变得日趋紧张,越来越难以满足日益增长的用户需求。另一方面,在现有的机制下,许多无线频谱资源被授权给了特定的用户却并没有得到很好的利用,大量的授权频谱由于授权用户的不存在或是不活动被闲置从而造成了频谱资源的浪费。为了解决这一矛盾,Mitola博士于1999年提出了认知无线电的概念,通过动态频谱接入使得感知用户可以使用闲置的授权频段的频谱资源,从而提升频谱的利用率,缓解频谱资源紧张问题。频谱感知在认知无线电系统中是一种基础且重要的功能,认知无线电系统的正常工作依赖于可靠的频谱感知结果,因此如何提升频谱感知的可靠性一直以来都是研究领域的重点。干扰是无线通信领域中的一个常见共性问题,在认知无线电系统中若存在无线干扰则授权用户和感知用户的正常通信均会受到影响。为保证系统能够应对复杂的无线通信环境,一个完整的认知无线电系统需要具备检测当前是否存在干扰的能力。围绕上述内容,本论文从以下三个方面开展相关的研究工作： 1.在认知无线电系统进行联合频谱感知时,如果忽视当前的网络参数而刻板地使用固定的数据融合规则常常造成感知可靠性不足或是系统开销过大。针对这一问题,本文提出了一种联合频谱感知数据融合规则选择方案。该方案从常用的数据融合规则中选出典型方案组成规则库,首先仿真分析在不同网络条件下规则库中各数据融合规则的性能,再根据上述结果设计一种能够根据当前的网络环境选择感知可靠性最高(若多个规则性能相近则选择响应更快的规则)的数据融合规则的算法。其中,上述算法中的待定参数通过遗传算法得出。仿真结果表明,相比较于使用单一的数据融合规则,利用此数据融合规则选择方案可以提供更高的频谱感知可靠性。此外,只需辅以低复杂度的信任值计算方法,本方案可以以更短的系统响应时间输出与其他高复杂度的方案相似的频谱感知正确率。 2.基于信任机制的联合频谱感知中,当一个感知用户汇报错误的本地感知结果时,数据融合中心通常是不分具体原因进行统一处理去降低一个节点的可信度,这种可信度计算方法并不能真实地反映一个节点的恶意程度。针对这一问题,本文提出了一种新的基于信任机制的联合频谱感知方案。该方案首先依据各节点的网络环境估算出相应的感知能力值,在此基础上结合感知结果历史数据使用最大似然估计的方法估算出用户的可信度,在数据融合中心进行数据融合时综合考虑用户的能力值和可信度输出最终的联合感知结果。根据仿真结果,使用此方案所计算出的用户可信度能够很好地捕捉感知用户的主观恶意程度,并且可以有效地提升存在篡改感知数据攻击情况下的频谱感知正确率。相较于传统的基于信任机制的联合频谱感知方案,本方案能提供更高的感知正确率,在最优情况下感知正确率最大可以提升约38%。 3.在所有的干扰类型中,学界普遍认为最难以检测的是被动型干扰。若存在被动型干扰对认知无线电系统进行干扰将对其正常运行带来极大的危害。针对这一问题,本文提出了一种针对被动型干扰源的联合检测与定位方案。该方案利用多个授权用户之间的协作,根据干扰对邻近授权小区中授权用户的影响建立常态档案,然后根据当前测量值与常态档案的相似度进行被动干扰源的检测,并且在检测出干扰源存在的情况下可以基于同样的输入进行对干扰源的实时定位。通过仿真验证,该方案检测速度较快,并且在最佳状况下可以实现被动干扰源检测的虚警概率和漏警概率同时低于10%,在干扰源定位方面,该方案最差情况下也有70%以上的概率保证干扰源定位误差在一跳之内。综上所述,本文以提升认知无线电中联合频谱感知的可靠性以及对干扰源的检测和定位能力为目标,从三个方面对相关领域的研究进行了补充,其中前两项研究针对认知无线电中的频谱感知,后一项研究针对认知无线电中的干扰检测。
[Abstract]:With the rapid development of wireless communication technology, the related equipment and applications are increasing. At the same time, the limited wireless spectrum resources are becoming increasingly tense, and more and more difficult to meet the growing user needs. On the other hand, under the existing mechanism, many wireless spectrum sources have been authorized to specific users but not very good. In order to solve this contradiction, Dr. Mitola proposed the concept of cognitive radio in 1999. Through dynamic spectrum access, the user can use the spectrum resources of the idle authorized band, from the dynamic spectrum access. And improve the utilization of spectrum, alleviate the problem of spectrum resource shortage.
Spectrum sensing is a fundamental and important function in cognitive radio systems. The normal work of cognitive radio systems depends on reliable spectrum sensing results. Therefore, how to improve the reliability of spectrum sensing has been the focus of the research field. Interference is a common common problem in the field of wireless communication. If there is wireless interference in a wire system, the normal communication between the authorized user and the perceived user will be affected. In order to ensure that the system can cope with the complex wireless communication environment, a complete cognitive radio system needs to have the ability to detect whether there is interference at present. The following three aspects are carried out in this paper. Related research work:
1. when the cognitive radio system is aware of the joint spectrum sensing, if the current network parameters are ignored, the fixed data fusion rules are often used to cause insufficient perceived reliability or excessive system overhead. In this paper, a combination rule selection scheme is proposed in this paper. In the data fusion rules, a typical scheme is selected to form a rule base. First, the performance of the data fusion rules in the rule base under different network conditions is simulated and analyzed. Then a kind of rules which can select the highest perceived reliability according to the current network environment is designed according to the current network environment (if the performance of multiple rules is similar, the rule of faster response is chosen). An algorithm for data fusion rules. Among them, the undetermined parameters in the above algorithm are obtained by genetic algorithm. The simulation results show that compared with the use of a single data fusion rule, this data fusion rule selection scheme can provide higher spectrum sensing reliability. In addition, the method of trust value calculation with low complexity is only necessary. The case can output a shorter system response time and similar spectrum sensing accuracy with other high complexity schemes.
2. in the joint spectrum sensing based on trust mechanism, when a perceived user reports the local perceived result of the error, the data fusion center is usually handled without a specific reason to reduce the credibility of a node. This credibility calculation method does not truly reflect the malicious degree of a node. In this paper, a new spectrum sensing scheme based on trust mechanism is proposed. This scheme first estimates the corresponding perception ability according to the network environment of each node. On this basis, the user's reliability is estimated by using the maximum likelihood estimation method based on the historical data of the perception results, and the data fusion center is used in the data fusion center. According to the simulation results, the user credibility can capture the subjective malicious degree of the perceived user well, and can effectively improve the spectrum perception accuracy in the case of tamper sensing data attack. In the traditional spectrum sensing scheme based on trust mechanism, this scheme can provide higher perception accuracy. In the optimal case, the maximum perception accuracy can be improved by about 38%.
3. among all the types of interference, the most difficult to detect is the passive interference. If there is a passive interference jamming to the cognitive radio system, it will bring great harm to its normal operation. In this paper, a joint detection and location scheme for passive interference source is proposed. Cooperation between multiple authorized users, based on the impact of interference on the influence of authorized users in the neighbouring authorized community, a normal file is set up, then the passive interference source is detected based on the similarity between the current measured value and the normal file, and the real-time determination of the interference source can be carried out based on the same input in the presence of the presence of the interference source. By simulation, it is proved that the detection speed is faster, and the false alarm probability and the leakage probability of the passive interference source detection can be lower than 10% at the best condition. In the worst case of the interference source, the scheme also has more than 70% probability to guarantee the interference source localization error in one jump.
To sum up, in order to improve the reliability of spectrum sensing in cognitive radio and the ability to detect and locate interference sources, the research in the related fields is supplemented from three aspects. The first two studies are aimed at spectrum sensing in cognitive radio, and the latter research is aimed at interference detection in cognitive radio.
【学位授予单位】：华中科技大学
【学位级别】：博士
【学位授予年份】：2014
【分类号】：TN925

【参考文献】