基于随机线性网络编码的频谱感知算法研究

发布时间：2018-08-11 18:55

【摘要】：随着移动互联网与智能移动终端的高速发展,频谱供需矛盾日渐突出,成为了制约无线通信技术发展的新瓶颈。然而,根据FCC报告指出,当前的平均频谱利用率仅为15%～85%,频谱资源浪费严重。认知无线电技术通过频谱感知利用频谱空洞进行通信,提高频谱利用率,成为了解决频谱资源短缺的关键技术之一。频谱感知是认知无线电技术实现的前提,因此,如何实现频谱状态的快速感知已成为认知无线电研究的重点问题。本文首先以减小检测时延、提高系统吞吐率为目标,在认知无线电中引入随机线性网络编码,对基于随机线性网络编码的快速频谱检测算法进行了研究。其次,为了减小检测时延的影响、降低虚警概率,本文利用隐马尔可夫模型对主用户信道进行建模,深入研究了基于随机线性网络编码的频谱预测算法。本文的主要工作和创新点总结如下：1、针对传统认知无线电网络中频谱状态转换频繁和频谱检测时延过长的问题,提出基于随机线性网络编码的累积和能量检测算法(RLNC-CUSUM)。该算法在主用户信道中引入随机线性网络编码,利用随机线性网络编码对频谱状态的整形作用,使频谱状态转换稀疏、频谱结构更规律化,进而减小频谱检测时延、提高系统吞吐率。针对传统的累积和能量检测算法(CUSUM)抗衰落性能差的问题,本文对五种衰落信道进行建模分析,并比较RLNC-CUSUM算法在不同衰落信道下的检测性能,进而验证该算法良好的抗衰落能力。实验结果表明,在一定的虚警概率下,该算法能够有效减小检测时延,提高吞吐率及抗衰落能力,能够更好地适应复杂的衰落信道环境。2、基于在认知无线电网络中引入随机线性网络编码带来的频谱预测性和隐马尔可夫模型对频谱的预测作用,提出基于随机线性网络编码的退避频谱预测算法(Back-off-SP)。该算法利用隐马尔可夫模型对主用户信道进行建模分析,引入退避预测方法,并对传统的基于隐马尔可夫模型的频谱预测算法进行改进,以提高频谱感知性能。通过实验仿真验证,该算法能够有效提高频谱预测概率,降低虚警概率,进而降低次用户对主用户的干扰,提高次用户吞吐率。
[Abstract]:With the rapid development of mobile Internet and intelligent mobile terminal, the contradiction of spectrum supply and demand is becoming more and more prominent, which has become a new bottleneck restricting the development of wireless communication technology. However, according to the FCC report, the current average spectrum efficiency is only 15% 85%, and the waste of spectrum resources is serious. Cognitive radio technology is one of the key technologies to solve the shortage of spectrum resources. Spectrum sensing is the premise of cognitive radio technology. Therefore, how to realize the fast sensing of spectrum state has become an important issue in cognitive radio research. In order to reduce detection delay and improve system throughput, this paper introduces stochastic linear network coding into cognitive radio, and studies a fast spectrum detection algorithm based on stochastic linear network coding. Secondly, in order to reduce the influence of detection delay and the probability of false alarm, this paper uses hidden Markov model to model the primary user channel, and deeply studies the spectrum prediction algorithm based on stochastic linear network coding. The main work and innovations of this paper are summarized as follows: 1. Aiming at the problems of frequent spectrum state transition and long spectrum detection delay in traditional cognitive radio networks, a cumulative and energy detection algorithm based on stochastic linear network coding (RLNC-CUSUM) is proposed. In this algorithm, random linear network coding is introduced into primary user channel, and the shaping effect of random linear network coding on spectrum state is used to make spectrum state conversion sparse, spectrum structure more regular, and spectrum detection delay reduced. Improve system throughput. Aiming at the problem of poor anti-fading performance of traditional cumulative and energy detection algorithm (CUSUM), five fading channels are modeled and analyzed in this paper, and the detection performance of RLNC-CUSUM algorithm in different fading channels is compared to verify the good anti-fading performance of the algorithm. The experimental results show that the proposed algorithm can effectively reduce the detection delay and improve the throughput and anti-fading ability under a certain false alarm probability. It can better adapt to the complex fading channel environment. It is based on the prediction of spectrum caused by the introduction of stochastic linear network coding in cognitive radio networks and the effect of hidden Markov model on spectrum prediction. A Backoff Spectrum Prediction algorithm (Back-off-SP) based on stochastic linear network coding is proposed. The algorithm uses hidden Markov model to model and analyze the primary user channel, introduces Backoff prediction method, and improves the traditional spectrum prediction algorithm based on hidden Markov model to improve spectrum sensing performance. The experimental results show that the proposed algorithm can effectively improve the spectrum prediction probability, reduce the false alarm probability, and then reduce the secondary user's interference to the primary user and improve the secondary user's throughput.
【学位授予单位】：广西大学
【学位级别】：硕士
【学位授予年份】：2016
【分类号】：TN925

【相似文献】