UHF频段频谱资源和认知TD-LTE系统接入关键技术研究
发布时间：2019-02-11 07:56
【摘要】：新一代宽带无线移动通信网络的顺利发展,需要足够宽的频谱来支持其通信的要求。作为我国拥有自主知识产权4G技术,TD-LTE目前获得的频率资源不但量少,并且都在2GHz以上。在WRC07大会上,ITU通过决议将由广播电视系统进行数字化改造而释放的数字红利频段(700MHz频段和800MHz频段)用于IMT。然而,我国的模拟电视数字化进程进展缓慢,这就导致数字红利频段难以分配给TD-LTE使用。相对于欧美已经完成了数字红利释放的国家,我国解决TD-LTE频率短缺、尤其是低频频率缺乏的需求更为急切。 近年来研究表明,认知无线电技术通过认知和重新配置获得最好的可用频谱,对已分配的频谱资源再次利用,是缓解频谱供需矛盾、提高频谱利用率的有效技术手段之一,为解决该问题提供了另一种途径。将认知无线电技术和TD-LTE系统结合起来,同时研究利用UHF频段的空闲频率,不但有助于解决我国的频谱短缺及管理问题,同时也有助于在目前积极发展的商用TD-LTE基础上研究认知无线电技术,对新一代宽带无线移动通信网开拓新的频谱资源有着重大意义。 本文主要进行了两方面的研究,一方面是数字红利频段的频谱资源使用情况,另一方面是认知TD-LTE系统接入的关键技术之一——频谱感知算法。首先,数字红利频段频谱资源使用情况研究主要针对694-806MHz频段,选取北京五个典型地点,根据测试地点、测量时间、测量设备和参数设置确定测试方案,进行了为其一周的频谱监测。最后利用Matlab进行数据处理,生成了一周频谱占用度光谱图和一周频段占用度条形图,数据处理时采用动态门限法来区分噪声和信号。数据处理结果表明,五个地点中频谱占用度最高的密集城区也只有25.40%的信道占用,说明TV频段比较适合作为认知无线电的候选频段接入。 其次,本文结合广播数字电视信号的特征,以经典频谱感知算法为基础,提出了优化的基于OFDM信号的多循环频率协作频谱感知算法。该算法的主要思想是,利用循环特征检测,生成单循环频率的判决结果,然后结合Chair-Varshney最优融合理论,将单点的循环频率判决结果进行融合,得到最后的判决结果。该算法既提高了频谱感知的精确度,又降低了传统协作频谱感知的复杂度。仿真时采用蒙特卡洛模型进行仿真。仿真结果证明,该算法的检测性能明显优于单循环频率检测,并且在低信噪比的情况下依然能实现较准确的频谱检测。
[Abstract]:The smooth development of the new generation broadband wireless mobile communication network requires a wide spectrum to support its communication requirements. As a 4G technology with independent intellectual property in our country, TD-LTE has not only a small amount of frequency resources, but also more than 2GHz. At the WRC07 conference, ITU passed a resolution to use the digital dividend bands (700MHz and 800MHz) released by the digital transformation of the radio and television system for IMT. However, the progress of analog TV digitization in China is slow, which makes it difficult to distribute the digital dividend band to TD-LTE. Compared with the countries in Europe and America where the digital dividend has been released, it is more urgent for our country to solve the shortage of TD-LTE frequency, especially the lack of low-frequency. Recent studies have shown that cognitive radio technology can obtain the best available spectrum through cognition and reconfiguration, and reuse the allocated spectrum resources, which is one of the effective technical means to alleviate the contradiction between supply and demand of spectrum and to improve spectrum efficiency. It provides another way to solve the problem. Combining cognitive radio technology with TD-LTE system, and studying the idle frequency of UHF band, it is not only helpful to solve the problem of spectrum shortage and management in our country. At the same time, it is helpful to study the cognitive radio technology on the basis of commercial TD-LTE, which is of great significance for the new generation of broadband wireless mobile communication networks to develop new spectrum resources. This paper mainly focuses on two aspects: on the one hand, the use of spectrum resources in the digital dividend band; on the other hand, the spectrum sensing algorithm, one of the key technologies of cognitive TD-LTE system access. First of all, the study on the use of spectrum resources in the digital dividend band is mainly aimed at the 694-806MHz frequency band. Five typical locations in Beijing are selected to determine the test plan according to the test site, measuring time, measuring equipment and parameter setting. The frequency spectrum is monitored for one week. Finally, Matlab is used to process the data, and a one-week spectrum chart and a one-week frequency band occupancy bar chart are generated. The dynamic threshold method is used to distinguish the noise from the signal in the data processing. The data processing results show that only 25.40% of the channel occupancy is found in the dense urban area with the highest spectrum occupancy among the five locations, which indicates that the TV band is more suitable to be used as a candidate frequency band for cognitive radio access. Secondly, based on the classical spectrum sensing algorithm, an optimized multi-cycle frequency cooperative spectrum sensing algorithm based on OFDM signal is proposed. The main idea of this algorithm is to generate the decision result of single cycle frequency by using cyclic feature detection, and then combine the optimal fusion theory of Chair-Varshney to fuse the decision result of single point cyclic frequency and get the final decision result. This algorithm not only improves the accuracy of spectrum sensing, but also reduces the complexity of traditional cooperative spectrum sensing. Monte Carlo model is used for simulation. Simulation results show that the detection performance of this algorithm is obviously better than that of single-cycle frequency detection, and it can still achieve more accurate spectrum detection under the condition of low signal-to-noise ratio (SNR).
【学位授予单位】：北京邮电大学
【学位级别】：硕士
【学位授予年份】：2014
【分类号】：TN929.5
本文编号：2419542
[Abstract]:The smooth development of the new generation broadband wireless mobile communication network requires a wide spectrum to support its communication requirements. As a 4G technology with independent intellectual property in our country, TD-LTE has not only a small amount of frequency resources, but also more than 2GHz. At the WRC07 conference, ITU passed a resolution to use the digital dividend bands (700MHz and 800MHz) released by the digital transformation of the radio and television system for IMT. However, the progress of analog TV digitization in China is slow, which makes it difficult to distribute the digital dividend band to TD-LTE. Compared with the countries in Europe and America where the digital dividend has been released, it is more urgent for our country to solve the shortage of TD-LTE frequency, especially the lack of low-frequency. Recent studies have shown that cognitive radio technology can obtain the best available spectrum through cognition and reconfiguration, and reuse the allocated spectrum resources, which is one of the effective technical means to alleviate the contradiction between supply and demand of spectrum and to improve spectrum efficiency. It provides another way to solve the problem. Combining cognitive radio technology with TD-LTE system, and studying the idle frequency of UHF band, it is not only helpful to solve the problem of spectrum shortage and management in our country. At the same time, it is helpful to study the cognitive radio technology on the basis of commercial TD-LTE, which is of great significance for the new generation of broadband wireless mobile communication networks to develop new spectrum resources. This paper mainly focuses on two aspects: on the one hand, the use of spectrum resources in the digital dividend band; on the other hand, the spectrum sensing algorithm, one of the key technologies of cognitive TD-LTE system access. First of all, the study on the use of spectrum resources in the digital dividend band is mainly aimed at the 694-806MHz frequency band. Five typical locations in Beijing are selected to determine the test plan according to the test site, measuring time, measuring equipment and parameter setting. The frequency spectrum is monitored for one week. Finally, Matlab is used to process the data, and a one-week spectrum chart and a one-week frequency band occupancy bar chart are generated. The dynamic threshold method is used to distinguish the noise from the signal in the data processing. The data processing results show that only 25.40% of the channel occupancy is found in the dense urban area with the highest spectrum occupancy among the five locations, which indicates that the TV band is more suitable to be used as a candidate frequency band for cognitive radio access. Secondly, based on the classical spectrum sensing algorithm, an optimized multi-cycle frequency cooperative spectrum sensing algorithm based on OFDM signal is proposed. The main idea of this algorithm is to generate the decision result of single cycle frequency by using cyclic feature detection, and then combine the optimal fusion theory of Chair-Varshney to fuse the decision result of single point cyclic frequency and get the final decision result. This algorithm not only improves the accuracy of spectrum sensing, but also reduces the complexity of traditional cooperative spectrum sensing. Monte Carlo model is used for simulation. Simulation results show that the detection performance of this algorithm is obviously better than that of single-cycle frequency detection, and it can still achieve more accurate spectrum detection under the condition of low signal-to-noise ratio (SNR).
【学位授予单位】：北京邮电大学
【学位级别】：硕士
【学位授予年份】：2014
【分类号】：TN929.5
【参考文献】
相关期刊论文 前2条
1 黄奇珊;彭启琮;路友荣;韩猛;;OFDM信号循环谱结构分析[J];电子与信息学报;2008年01期
2 杨小牛;;从软件无线电到认知无线电,走向终极无线电——无线通信发展展望[J];中国电子科学研究院学报;2008年01期
，本文编号：2419542
本文链接：https://www.wllwen.com/falvlunwen/zhishichanquanfa/2419542.html
