无线通信系统中的频谱资源占用模型研究

发布时间：2019-05-10 03:38

【摘要】：现阶段无线通信系统对频谱资源的使用已经远远超出了人们的预期,日渐紧缺的频谱资源使得继续拓展频谱潜力变得极为困难。如何能够合理的对频谱资源展开使用成为了全世界关注的焦点。在本文中,笔者从无线频谱资源的占用出发,对频谱占用模型进行了全面细致的分析。对实际频谱的分析,让我们能够以空口测量和频谱感知为基础,捕获各类无线通信系统中授权用户的业务参数,这种研究在很大程度上起到了对现有频谱占用模型的梳理作用。现实中的频谱会呈现出不同的频谱使用状态,连续的时间则意味着连续的频谱状态转移。对无线通信系统而言,我们假设其各状态的驻留时间分布接近指数分布,从而引入了以马尔可夫链为基础的隐马尔可夫模型来建模频谱的使用状态。对于目标无线通信系统,本文首先针对典型的蜂窝移动通信系统GSM,进行了连续四天的功率谱密度空口测量。长时间的测量给了我们研究GSM下行频段信道状态和业务特性的机会。通过对测量功率序列使用基于隐马尔可夫模型的高斯观测BW算法,我们能够成功获取GSM的信道状态参数,在此基础上,提取出连续时间内GSM系统授权用户的业务特性参数。接下来,本文针对另一典型无线通信系统无线局域网进行了细致的分析。本文选择从802.11b的物理层结构开始研究,逐步分析了其数据包构成、DBPSK通信系统的性能,以及数据包业务对频谱资源的占用情况。在这些理论基础上,笔者用802.11b的实际空口测量数据验证了其无线数据包与协议的一致性。通过对数据包进行逐一解调,得到了由测量数据统计出的数据包业务参数。同时,根据业务源泊松发包的事实,我们证实了频谱状态的驻留时间是近似呈指数分布的,因此,利用隐马尔可夫模型对频谱状态的估计,我们得到了与真实发包速率相符的业务参数。本文验证了隐马尔可夫模型与无线频谱占用模型的高度吻合性。一方面,我们利用隐马尔可夫模型估计GSM下行链路的信道状态参数集,进而提取授权用户的业务特性参数集;另一方面,针对802.11b协议下的无线通信,我们先将相应频段中的数据包进行解调与功率调整,再由统计理论与隐马尔可夫模型参数估计得到业务源的业务特性参数。
[Abstract]:At present, the use of spectrum resources in wireless communication systems has far exceeded people's expectations, and the increasingly scarce spectrum resources make it very difficult to continue to expand the spectrum potential. How to use spectrum resources reasonably has become the focus of attention all over the world. In this paper, the author makes a comprehensive and detailed analysis of the spectrum occupation model from the occupation of wireless spectrum resources. The analysis of the actual spectrum enables us to capture the service parameters of authorized users in all kinds of wireless communication systems on the basis of empty port measurement and spectrum sensing. To a great extent, this research plays a role in combing the existing spectrum occupation model. In reality, the spectrum will show different spectrum usage states, and the continuous time means continuous spectrum state transition. For wireless communication systems, we assume that the resident time distribution of each state is close to the exponential distribution, so we introduce a hidden Markov model based on Markov chain to model the usage state of spectrum. For the target wireless communication system, the power spectral density empty port measurement for four consecutive days is carried out for the typical cellular mobile communication system GSM,. The long time measurement gives us the opportunity to study the channel state and traffic characteristics of GSM downlink band. By using Gao Si observation BW algorithm based on hidden Markov model for the measured power sequence, we can successfully obtain the channel state parameters of GSM. On this basis, the service characteristic parameters of authorized users of GSM system in continuous time are extracted. Next, this paper makes a detailed analysis of another typical wireless communication system wireless local area network (WLAN). In this paper, the physical layer structure of 802.11b is studied, and the packet composition, the performance of DBPSK communication system and the occupation of spectrum resources by packet service are analyzed step by step. On the basis of these theories, the consistency between the wireless packet and the protocol is verified by the actual empty port measurement data of 802.11b. Through the demodulation of the data packets one by one, the packet service parameters counted by the measured data are obtained. At the same time, according to the fact that the service source Poisson sends the packet, we prove that the residence time of the spectrum state is approximately exponential distribution. Therefore, the hidden Markov model is used to estimate the spectrum state. We get the business parameters that are consistent with the real rate of delivery. In this paper, the high coincidence between hidden Markov model and wireless spectrum occupation model is verified. On the one hand, we use hidden Markov model to estimate the channel state parameter set of GSM downlink, and then extract the traffic characteristic parameter set of authorized users. On the other hand, for wireless communication under 802.11b protocol, we first Demodulate and adjust the data packets in the corresponding frequency band, and then obtain the service characteristic parameters of the traffic source from the statistical theory and hidden Markov model parameters estimation.
【学位授予单位】：电子科技大学
【学位级别】：硕士
【学位授予年份】：2014
【分类号】：TN925.93

【相似文献】